Joe's 1000x600x200 steel and aluminium router started

[devmonkey]

This will be the 4th machine I've built but the first for 10 years and the first I have posted a thread on. After various house moves, workshop rebuilds, etc, I am actually without a functioning router for the first time in ages which is extremely frustrating!

This machine will be reasonably compact, work volume of around 1000x600x200mm. It is very similar to the first router build by Boyan, in that it has a self supporting bed frame with raised X rail supports. I need to process aluminium, carbon, modelling board and maybe hardwood.

Frame is 80x80x3 welded steel box.

Gantry is 160x80 extrusion (Item series 8 heavy or clone thereof).

Plates will be 16mm cast milled tooling plate which I will have to produce on my manual mill.

Ballscrews from Fred, 1610 on X+Y, 1605 on Z, with BK/BF12 AC support blocks.

Hiwin 20 on X and 15 on Y+Z. X rails will be epoxy levelled with West 105/209.

I'm going to use steppers (in the past I used DMMTech servos) but seemingly 3N steppers with digital drives are quite suitable for this small machine.

I have done the basic dimensioning so I can get started on the frame. The gantry plates and Z axis will be adjusted as required but the sketches below should give you (and me) and idea of where I am heading.


26759
26760
26761


Frame box section cut to a fairly good tolerance on the trusty/rusty Rage2.

26762

And frame welded up, not quite finished yet, needs some end plates and reinforcements to catch the screw supports and motor mounts. I used an inverter stick welder, quite a revelation, amazing capability in something the size of a lunch box. Was quite careful to avoid warping, not by limiting heat but rather by carefully sequencing the welds to counter each other. I can't measure any warpage so it was successful.


26763

I would would welcome any input on the basic design, but more urgently I'd like some help on the following:
1. Where is the best place to buy cast milled plate from? The guy I last bought from locally has gone to the wall.
2. What is the minimum constant temperature you would risk an epoxy pour with 105/209?
3. Has anyone bought extrusion from Alprofil? http://www.alprofil.com/aluminium_profile_frames.htm

Cheers, Joe

[AndyUK]

Regards item 1:

http://www.mycncuk.com/threads/13149-Aluminum-plate-for-gantry-side

3: No but KJN were great.

[devmonkey]

Thanks Andy, will call KJN tomorrow, they don't list the price for 160x80 on their site.

Last time I used a 6"x2"x0.25" aluminium box section, it worked ok but wasn't particularly flat or straight. I'm hoping I can use the slotted profile gantry without having to shim the rail. Milling the gantry is a pain as my mill only has 700mm of travel so would need to move an re-reference and would loose accuracy.

[routercnc]

Looking good, classic design. Ballnut brackets off gantry plates may need a bit of support?

I’ve always ordered from aluminium warehouse as you can select the exact size you need. Other members have mentioned cheaper sources elsewhere (Smiths?).

I tried to keep 21deg C for the epoxy pour but from my brief experience with it I wouldn’t be concerned going down to say 18. I guess the heat might be required more at the beginning to keep the viscosity lower for the self levelling as it should generate its own heat as it cures. Others may have more to add on that.

[devmonkey]

I've only used Aluminium warehouse for tubes in the past, couldn't complain about them, product came well packaged, but I've read on this forum their cast plate is sometimes delivered damaged? I guess the plate you received was ok?

I've reconfigured the gantry end plates and extended the gantry. I wanted to keep the nut holder fixing on the outside so that I can fit simple angle over the screw to keep debris off it. With this configuration I would recess the X bearing plate and the gantry extrusion into the gantry end plate a couple of mm to keep the gantry vertical to the X rails.It also provides enough space outside the X rail to bolt up through the X bearing plate into the extrusion.

The (red) ball nut holder assembly will be adjustable up and down and if I make the gantry a few mm wider it will allow me to shim the ball screw supports off the steel frame to get the screws properly aligned, making the absolute length of the gantry beam non-critical.On my last machine I didn't have enough adjustment to align the X ball screws properly and it was a bit of headache to sort.

26768

[devmonkey]

Made some more progress, capped box sections ends:
2679626797

I'm now waiting on the ballscrews to get here (next week) so that I can weld patches for attaching the support blocks and steppers.

I've chosen EM806 drivers. Yesterday I was about to order from AliExpress, thought I'd check Zapp first and Gary has a Black Friday deal on, managed to score 4xEM806 for £340 inc vat and delivery and rather unbelievably they turned up this morning 17hrs after ordering! What an amazing service and at the same or better price than China after duties without the wait.

Steppers are ordered from OMC-stepperonline coming from UK warehouse, should be here in the next few days.

Given the drivers arrived and I already had an enclosure to hand I've started to build up the electronics, PSU parts are from Rapid:
26798

I also ordered the tooling plate from Smiths, rather good price, 16x3000x355mm 'offcut' for £274. Loads more than I need for this machine but it is useful stuff to have around. It should arrive tomorrow so I can get to milling the plates.

I'm now trying to decide what to do about the epoxy pour given it really needs a nice constant 20 degrees. The frame is in my un-insulated garage, do I build some sort of insulated box around it and bung a heater in there or do I bring it into the house. Temperature wise the house makes sense, however the frame is very heavy and it is more likely to get disturbed indoors by the kids/dog.

[AndyUK]

Rapid progress! No messing around here! You're making me look like a snail though ;)

For the epoxy I'm in the same situation re uninsulated garage. I waited until summer looking for a week of predicted lows over 10 degrees. I remember something about an 8 degree minimum... When I tried the first time I heated the garage to around 28 - but the reaction is exothermic and excess heat sets it off faster which is very counter productive. The second time I didn't bother and it was much better but probably for other reasons.

[devmonkey]

I've actually burnt myself with epoxy before using the normal hardener not the super slow 209. I was building a fibreglass mold and mixed up about half a pint of epoxy, spent about 5 mins stirring it in a container that was far too tall and narrow and it exothermed, spat on my hand which burnt the skin and I threw it outside, the container completely melted and the epoxy was smoking. Only happened once and I've done this many times, I always mix it in a large shallow container now (like a takeaway rice container), never happened since. Trick is to maximise the surface area when mixing.

Anyway I now have a plan:

I picked up a 40w tubular heater today from Toolstation for £14. It has a thermostat. I will get some sheets of polystyrene floor insulation and build an oven around the machine on the garage floor, then pour epoxy and 'slow cook' at 20 degrees for a week. Wont be until next week as I'm still waiting on the ballscrews before I can finish the frame.

One other thing, I'm wasn't planning to put any levelling feet on this machine, it will be mostly sat on a table/bench so I'll just put some self adhesive rubber pads on the bottom. What do you think? For the epoxy pour I will level it with shims off the concrete floor.

[Kitwn]

All this talk of how to raise the temperature reminds of when I was a Pom! My wife and I are starting some epoxy resin art projects soon and with summer starting in a few days are worrying about how to keep the temperature DOWN to 20C. Overnight lows are currently about 22C just before sunrise and then things warm up rapidly. Highs in the shed/studio go to near 50C at times so the plan is to keep the resin in a wine fridge overnight and start work about 6am, possibly with the air-con already running.

I do plan to build a new epoxy-levelled frame for my router at some time, but that will have to wait until the depths of winter around July next year when daily highs in the shed can be kept reliably below 30C.

[devmonkey]

Looking for some validation of my estop circuit as I didn't need to build one on my last machine, it was all integrated into the servo BOB.

My proposed circuit is a normal start-stop relay circuit with the addition of the EM806 fault signals, is this how everyone else does it?

26811

E-stop is NC
Reset is NO momentary
FLT 1-4 are the open collector FLT outputs from the 4xEM806.
K1 is a DPST relay, first pole used for latching, second pole to switch contactor.
K2 is a SPDT relay using the NC contact

Functions as follows:
With no driver fault signals K2 is off and conducting.
E-stop is conducting and in series with K2.
K1 is off.
Reset is pressed momentarily, current flows through K1 coil, latching K1 on.

Now if any driver fault signal switches on, K2 swiches on breaking the path through K1 coil, K1 switches off. If fault clears K1 remains off until reset is pressed.
Similarly if E-stop is pressed this also breaks the path through K1 coil and the same happens.

Any additional NO sources of fault can be paralleled with the FLT outputs. Any other NC sources of fault can be added in series with E-stop + K2.

Thoughts?

[devmonkey]

I've just checked the EM806 datasheet and the fault outputs can be set to normally low impedence, i.e. normally closed. This makes things much simpler as they can just be placed in series with e-stop.

26812

[devmonkey]

My plate has finally arrived. The first piece had a bend in it:
26839

I think it must have been hit by a forklift in the warehouse. Anyway called Smiths and they exchanged it the next day, the new piece is perfect.
26840


I've roughed out some bits for the Z axis on the table saw.
26841

I got a deal on this plate because it was originally an offcut piece, however I have way too much material for this machine. I certainly have enough plate to build the gantry entirely from plate and not bother with the extrusion, has anyone done this or know of any build threads (I have looked)?

From plate I'm guessing I would form a channel 160x80 with the back open, question is would this be sufficiently twist proof?

[Voicecoil]

From plate I'm guessing I would form a channel 160x80 with the back open, question is would this be sufficiently twist proof?

Obviously it depends on how thick your plate is! BUT channels aren't generally that strong torsion (twisting) wise, a box is much stiffer weight for weight.

[devmonkey]

You are right!

I just used the demo version of shape designer, this is 16mm plate.

Torsion constants, for reference the 160x80 ITEM heavy profile 390cm^4

160x80x16mm channel is 39cm^4
160x80x16mm box is 1400cm^4

[Voicecoil]

Something else to think about..... 6" x 3" x 3/8" box section (available off the shelf): Ix = 352, Iy = 1152, torsion constant (J?) = 823 - and about 1/2 the price of Item 160x80, though you obviously have to drill some holes :05.18-flustered: A lot of the slotted structural profiles don't seem to be that amazing torsion-wise, I guess it's because a lot of the metal is a fair way inside the extrusion, whereas in a box section it's all at the outside, where it does most good.

[JAZZCNC]

Something else to think about..... 6" x 3" x 3/8" box section (available off the shelf): Ix = 352, Iy = 1152, torsion constant (J?) = 823 - and about 1/2 the price of Item 160x80, though you obviously have to drill some holes :05.18-flustered: A lot of the slotted structural profiles don't seem to be that amazing torsion-wise, I guess it's because a lot of the metal is a fair way inside the extrusion, whereas in a box section it's all at the outside, where it does most good.

Cheaper yes but it's not better unless it's prepped correctly and even then it's much more resonant. The profile is generally much flatter than box steel which isn't flat in any direction and it's less resonant, it's a lot easier to work with and does the job fine.

Torsion isn't an issue until gantry width gets wide and even then if sized correctly profile works fine for routers.

[devmonkey]

So I just did some maths, if 've got all my units correct, a 100N force on the cutter exerted at 300mm below the centroid of the gantry cross section and half way along the 1000mm Y axis would cause a deflection at the tool of approximately:

160x80x16mm channel is 39cm^4 - deflection 0.22mm
160x80 ITEM heavy profile 390cm^4 - deflection 0.022mm
160x80x16mm box is 1400cm^4 - deflection 0.006mm

I guess that 100N is unrealistic and would break the cutter first.

I'm going to have to drill all the holes anyway, is the bog standard 6" x 3" x 3/8" aluminium extruded box section flat and twist free with parallel sides?

I was looking at the Item profile extrusion since I thought it would give me parallel flat sides without having to process it. My mill table is only 700mm long, not quite enough to surface the entire gantry without re-locating the work which is difficult to do accurately in plane.

I'm also sure I've seem some pictures on here of peoples gantries constructed from plate?

[Voicecoil]

Cheaper yes but it's not better unless it's prepped correctly and even then it's much more resonant. The profile is generally much flatter than box steel which isn't flat in any direction and it's less resonant, it's a lot easier to work with and does the job fine.

Torsion isn't an issue until gantry width gets wide and even then if sized correctly profile works fine for routers.

I was meaning box section aluminium, not steel, I'm aware of the problems with that. The stuff I got was pretty straight, the only error from square was a very slight (0.3mm) but even dip along the middle of the long sides, obviously the die designer didn't get his shrinkage allowance quite right :concern:

[Doddy]

<deleted... I need to read to the end of thread, not end of page>

[devmonkey]

I was meaning box section aluminium, not steel, I'm aware of the problems with that. The stuff I got was pretty straight, the only error from square was a very slight (0.3mm) but even dip along the middle of the long sides, obviously the die designer didn't get his shrinkage allowance quite right :concern:

That is interesting, if the box is square to its corners then it is much simpler to skim it flat on the mill.I would need to buy a length in order to check it in the mill with a DTI, Aluminium warehouse if out of stock.

I'm still interested in looking into constructing the gantry from all this spare plate I have.

[Voicecoil]

I got mine from Simmal (you might need to email/ring them as they don't keep everything ont' website), I suspect other places like DoRé Metals & Smiths might have it too.

[JAZZCNC]

I'm also sure I've seem some pictures on here of peoples gantries constructed from plate?

Yes probably seen gantry I've shown built from plate and few others used it or adapted to there machines. It makes for a very stiff gantry,

However, it needs a lot more work compared to HD profile. I've used all methods of constructing gantry's and HD profile is by far simpler with more than enough strength for any Router cutting materials up to Aluminium.

Aluminium Rect section doesn't compare to HD profile in terms of resonance and vibration dampening and this is should be more of a concern than twisting deflection you seem to be chasing if you want a good Router.
There is no cutter you will use that will withstand forces enough to twist a gantry made form any of these materials before it snaps. However, Resonance will affect any cutter and the smaller the cutter the worse it gets. My advice is to Stop chasing twisting etc and look more to lowering resonance if you want a good Router.

[Voicecoil]

Aluminium Rect section doesn't compare to HD profile in terms of resonance and vibration dampening and this is should be more of a concern than twisting deflection you seem to be chasing if you want a good Router.


That's an interesting point and worthy of investigation. I've saved the impulse response files from a length of the 6 x 3 x 3/8 box, does anyone have a short-ish (~600mm) length of 160 x 80 profile they can lend me to do a comparative measurement?

[devmonkey]

Yes probably seen gantry I've shown built from plate and few others used it or adapted to there machines. It makes for a very stiff gantry,

However, it needs a lot more work compared to HD profile. I've used all methods of constructing gantry's and HD profile is by far simpler with more than enough strength for any Router cutting materials up to Aluminium.


Hi Jazz,

Thanks for the advice. Do you have any pictures or wise words on how a 'stiff' gantry was made with plate? From the calculations I did a simple channel from 16mm plate is not sufficiently stiff in torsion at 1/10th that of 160x80 profile.

I'm trying to reuse a set of HGR15 rails left over from another project on the Y. These were originally purchased to replace rails on a much larger machine that had rusted badly, since then that machine has been written off.

15mm rails are too narrow to go over an 8mm slot, this means they will need to be drilled and tapped into the centre flat section between two slots or another plate milled to bridge the slots and then mounted to that. If the latter then it isn't much more work to mill up a front plate as well.

If I already had 20mm rails for Y then the profile would be a no brainer as the rails could be bolted straight into the slot.

[devmonkey]

Right I've made a decision!

I called around a load of my usual aluminium guys, Smiths, Aalco, Blackburns, plus a couple of local metal shops. No one has any 6x3x3/8 box on the shelf (even Aalco who have a warehouse near me about 10 acres in size) and although they could get it I would need to buy a full length. To be fair I didn't try Simmal.

I've built two previous machines using 6x2x1/4 box and they worked fine albeit I had to shim the rails, only very slightly as I didn't have a mill back then and I had to bolt blocks into the ends to bolt into (not required on the 3/8 box). Anyway I'm just going to go with the original plan from the first post in this thread, will order some 160x80 Item heavy clone from KJN tomorrow.

I will first try to just mount rails directly to the central flat on the profile, if this turns out too wobbly I will plate over the slots and fix rails to plate.
26845

[devmonkey]

I've started milling parts for the Z.

Front plate and rail spacers finished, all fits well. These large plates take an age to mill out on my manual mill.
26856


At least drilling the holes accurately is simple with the DRO.
26865

Also took a delivery of bolts for the machine:
26855

[devmonkey]

Finished the Z back plate so have now assembled the Z. Everything fits perfectly. All 4 long edges of the large front and back plates were milled in a single fixture so these will all be reference edges for aligning the rails and also tramming the Z on the gantry.

26870

Pockets for the ballscrew mounts, these weren't perfect were supposed to be 1mm deep, infact 1.28mm so will need a thin shim between front plate and ball nut holder. TBH ballnut holder is not a precision item, rather it is a cast lump, so probably would have needed shimming anyway.


26871

26872 26873 26874 26875

I'm now waiting on the extrusion to arrive for the gantry as I need this to measure off for the Y bearing plates, gantry end plates, etc. Hopefully should be here early next week.

[devmonkey]

How do I make the images in these posts larger rather than just thumbnails?

[Kitwn]

Joe,
That's looking very impressive.
Clicking on the thumbnails takes you to the larger versions which, on my laptop at least, are sized to fit the screen. Clicking again will enlarge them to what looks to me to be 'actual pixels' size.

Kit

[devmonkey]

Joe,
That's looking very impressive.
Clicking on the thumbnails takes you to the larger versions which, on my laptop at least, are sized to fit the screen. Clicking again will enlarge them to what looks to me to be 'actual pixels' size.

Kit

Hi Kit, I actually meant how to inline images in posts larger than thumbnails, other people seem to be able to do it.

Machined up the Z stepper plate this morning.

26878 26879 26880

[Kitwn]

Hi Kit, I actually meant how to inline images in posts larger than thumbnails, other people seem to be able to do it.

I kind of thought that, but those of us who care enough can see your full detail pictures and you look to be a bit busy with something more important than how big your forum images look!

Kit

[devmonkey]

Gantry extrusion has arrived from KJN, it is a real beast. KJN cutting appears very accurate.

26902 26903

So now I need to decide how to mount 15mm rail to the extrusion. I was considering 3 options:
1. Mount rail directly to extrusion between the two slots, drilling and tapping.
2. Mount rail on aluminium plate which is in turn bolted to the slots.
3. Mount rail directly to slot.

This shows that the profile is not flat across the width which I believe is by design so that the slots pull up to meet whatever you are clamping to them. The slot edges are 0.15mm lower than the centre, the centre and the 2 corners are perfectly aligned within the resolution I can measure with finest feeler gauge.

26905

Now I have the extrusion (1) is a non-starter as there is a very slight peak in the centre and I would need to mill it flat which is hard given the gantry is longer than my mill table, although not impossible.
(2) is what many people seem to do and so obviously works but is more work for me.
Can I get away with (3), see picture below, what do you think?

26904

I have a 50mm fly cutter, it would be possible given the extrusion is already flat and parallel to fly cut between the high points, would have to tram the mill carefully first.

[Kitwn]

That's serious looking profile! What total length did you get for your £340.24?

Kit

[devmonkey]

I wasn't that expensive, think it was about £130 plus vat for 1100mm.

[Voicecoil]

Looking at your pic with the straight edge it looks to me that the slope on the inside of the slot is slightly less steep than the slope from the slot to the outer edge. This would mean that your rail would be slightly canted if you go for fixing option 3).

[devmonkey]

You have very keen eyes! Yes at the slot the extrusion is 0.15mm low both sides, since the outside flat is shorter than the centre flat it must be at a slightly steeper angle. I have measured and calculated the cant angle the 15mm profile would have as 0.055 degrees or 198 arcseconds, does this matter (it will be mirrored on the second rail).

Or should I just man up and hit it with this:
26908

[Voicecoil]

That's not a lot of angle to be true, and when tightened up it might straighten it a bit more. On the other hand facing off the central flat and drilling holes would give a firmer fix I suspect as you're going into continuous metal, rather than a slot - and as Jazz pointed out with HGR15 there's not a lot of contact either side of an 8mm slot to give friction.

[devmonkey]

I temporarily bolted a length of rail to the slot, it seems to mount extremely firmly and doesn't move (maybe helped by the very slight V angle it is sitting in) but must be slightly canted as discussed above. Will decide whether to face mill it tomorrow, another option is just to mill a few mm each side of the slot so the rail can sit flat, this would also preserve the 3 planar points (both corners and centre) in case I mess up.

I initially wanted to fix the rail to the centre however this does increase the moment between the Y bearing and the Z plate by 20mm which may counter the increased stiffness from the firmer rail mounting.Then again the centre of this extrusion will be stiffer than either slot as it i supported equally whereas the slots are supported on one side by the unsupported corner. No idea really.

In the mean time I spent at least an hour tapping the ends:
26915 26916

[Kitwn]

If I was living in a densely populated part of the UK I'd be asking if any fellow forum member had a nice flat machine that could mill a shallow slot just wide enough for the rails to fit along the full length of the centre of the profile. Possibly in exchange for HobNobs. That looks to be by far the most rigid place to put them.

[routercnc]

One of my earlier machines used a heavy gauge extrusion with 2 slots and I used the slot closest to the spindle rather than go into the centre. I think it was slightly crowned across the width as well.
I used a full length steel insert inside the slot (a long thin bar with the edges chamfered) to thread into rather than lots of small t- nuts and this helped pull it flat at least locally.
There probably was an overall angle to the rail but I also had a similar rail on the underside surface which would slope the other way. When these were connected to Y axis assembly it must have preloaded the bearings slightly trying to make them parallel to each other. I didn’t notice any issues and the machine ran for years. You can check on the Hiwin site what misalignment is allowed and compare to your angle.

[devmonkey]

I just looked up the hiwin specs, they say 130um over 500mm max allowable surface height error between two rails in the same plane, so I think this equates to an angle of 0.015 degrees or 53 arcseconds, I would be around 4x that with the rails mounted directly to the first slot. It would probably be absolutely fine but anyway.

I've made up some feet from spare plate. These are bolted into the bottom of the extrusion and will then be clamped to my milling table.

26920 26921

I've just trammed the head to the table, and checked the table over the entire X travel. The table is within 0.01mm, i.e. I can't measure any error with a 0.01mm resolution DTI. Think i'm just going to flycut the top and bottom of the entire extrusion, then I'll have the option of mounting to a slot or tapping into the centre. No time to do it today taking the kids to see the new star wars film this afternoon.

[Voicecoil]

another option is just to mill a few mm each side of the slot so the rail can sit flat, this would also preserve the 3 planar points (both corners and centre) in case I mess up.
That's likely not such a bad idea as it could give some lateral stabilit - Hiwin suggest sitting the rails in such a groove in the manual. However you'll need to do it accurately as it removes the option for tweaking alignment to some extent.

.Then again the centre of this extrusion will be stiffer than either slot as it i supported equally whereas the slots are supported on one side by the unsupported corner. No idea really.
Routercnc's point about using a tapped strip rather than individual floating nuts might help this.

[devmonkey]

Well I've milled 4mm either side of the slot flat on each side of the extrusion. Setup took a long time making sure everything was flat with the dial indicator, nipping up the gibs to try and counter the sag from the long overhang. I pretty pleased with the result, seems at least as flat as the extrusion. I need to get some rails mounted to it and make the Y bearing plates so I can see if it is sufficiently parallel to be usable. DOC was about 0.25mm with a carbide endmill, could feel it was cutting deeper on the outside due to the increased slope as discussed in the previous posts.

I had to make the cut in two steps, moving the beam and refixing between cuts since the mill table wasn't long enough. I can see the transition but cannot feel it. Nice hatch marks show the head was in tram.

26926 26927 26928 26929 26930

[Voicecoil]

Nice work, hope it sorts it.

[Kitwn]

I agree with voicecoil. My only suggestion when you cut the other one would be to support that overhang with an elastic luggage strap from the roof of your shed.

Kit

[devmonkey]

Hi Kit, I cut them both in one sitting yesterday, i measured the effect of the sag over the saddle with a DTI 4mm offset from the spindle (I used an 8mm endmill for the facing operation) whilst lifting the weight of the overhanging part, didn't really have any effect with the gibs tight as you would expect given the table pivots on the saddle. Probably not an acceptable way to fly cut but with a small endmill it works. I thought about slinging the overhang to my engine hoist but the problem with that would have been as the table is moved to the other end you would need to reverse the force since it becomes overbalanced the other side.

I think the best possible way to balance a job like this is to put a sliding counter weight inside the extrusion and move it through as the work moves (in the opposite direction), but I couldn't be bothered in this instance, if I was using the large fly cutter I would have had to bother else it would have cut a tapered scallop in the top surface.

[ericks]

Interesting :)

[devmonkey]

Milled the Y bearing plates, just need to counter bore them. Lower plate wraps around the rails on the Z front plate to reduce coolant splash up inside the Z assembly, it is also recessed to accept the rear Z plate press fit so that it is easy to line it up with the upper Y plate. The row of 4 holes that will be used to bolt the lower Y to the rear Z plate are undersized as I want to use them as a drill guide for the tapped holes in the rear Z. The Z is too tall to stand on my mill so these will have to be drilled on the pillar drill or with the cordless.

Also the spindle mount arrived from Canada, very nice bit of kit.
26935 26937 26938 26936

[devmonkey]

Got the rails mounted on the milled gantry, worked really well. I've tightened them down but not yet aligned straight, for that I will mill up a couple of precision setting tools.

I then finished the Y bearing plates and bolted everything together. Everything slides very smoothly with bolts tight, very exciting! The Y bearings have the same smooth action across the entire length with no hint of any variation so I presume the gantry milling was sufficiently parallel.

26939 26940 26941

One thing I need to do before too long is hit the corners of these plates with a sander, they are razor sharp and I keep bleeding all over the lovely shiny parts.

[devmonkey]

Happy new year everyone!

Small progress update, I have finished the gantry end plates and X bearing plates.

26970 26971 26972 26973 26974

I have about 6 small plates to make for stepper and bk mounts then finished with the manual milling.

[Sterob]

Nice work Joe!
I see you are milling small slots to locate perpendicular sections to each other...... I enquired about doing just that in another forum with almost no reaction.
It seems to make sense to me ( no expert) What better way to ensure components are aligned correctly?

How have you found the process? A good idea or problematic?

I'm considering doing the same thing...when I eventually start my own build.

I see you have maximised the distance between your x rails, by placing them top and bottom of the Gantry.
Did that make things excessively more difficult?

Steve

[Kitwn]

Hi Steve,
Joe won't be out of bed yet:sleeping:

There's a fair amount of stuff on gantry design on the forum. Putting rails top and bottom has the advantage of pulling the Z axis closer to the gantry and reducing the overhang. It also gets the rails as far apart as possible for the chosen height of the gantry but makes the construction of the Z back-plate more complex. Alignment of the rails is not 'automatic' in any way and must be done with care.

The option of mounting rails directly on the front of the gantry pushes the Z axis out but can be easier to align, especially if you're using epoxy leveling or a machine-ground plate to ensure a perfectly plane surface. It also makes the Z axis back-pale into a single flat piece. It all comes down to what tools and measuring instruments you have access to.

As long as the first machine you build is adequate to cut out the pieces for a second, better version then you will be on an upward evolutionary slope.

Kit

[devmonkey]

Nice work Joe!
I see you are milling small slots to locate perpendicular sections to each other...... I enquired about doing just that in another forum with almost no reaction.
It seems to make sense to me ( no expert) What better way to ensure components are aligned correctly?

How have you found the process? A good idea or problematic?

I'm considering doing the same thing...when I eventually start my own build.

I see you have maximised the distance between your x rails, by placing them top and bottom of the Gantry.
Did that make things excessively more difficult?

Steve

Hi Steve,

I also thought twice before milling the locating slots on the different plates. On some joints it is not so critical, such as attaching a stepper mounting plate but when you are joining two plates that locate the axis at exactly 90 degrees such as where my gantry end plates join the X bearing plates it is super critical. In this example it was also critical the slots in the top of the X bearing plates were not only planar with the plate surface but also identical depth on both X bearing plates.

When I mill out shapes from plate I clamp the plate to MDF so the cutter clears the mill table. MDF (particularly when it has sat in the garage for a year) is not sufficiently uniform for accurate Z milling. Therefore I mill these slots after the outside profile has been finished. I set the Z height on the mill and lock the axis then I clamp each X bearing plate to the mill table directly and mill the slot. I do both plates without unlocking the Z axis so they should be identical.

Before doing any of this critical stuff (particularly milling the gantry extrusion flat where the Y rails are mounted) I checked the mill table was flat within 0.01mm over the entire travel with a DTI in the spindle. You don't want to do this if your saddle is sloppy or any long work really.

I'm an amateur machinist at best, only been playing around with the manual machine for a few months really so take this advice with a pinch of salt. The slots are within 0.01mm in depth over their length which is good enough and no worse than the milled plate itself. It is quite satisfying when you get a friction fit on the joint between two plates.

Also bear in mind that the milled plate isn't exactly whatever nominal dimension you purchase, mine in nominally 16mm but is infact 16.055mm thick, also your cutters are not exactly whatever dimension they are supposed to be. This means if you just mill the slot to the DRO readout it wont be the correct width.

What I do is designate one side of the slot as the reference since my cad drawing assumes 16mm plate, then mill that. Then open up the slot and use a piece of clean spare plate to test the slot until you get a nice fit, this is much more accurate than trying to measure with a vernier or micrometer as the slot is too shallow.

[devmonkey]

I see you have maximised the distance between your x rails, by placing them top and bottom of the Gantry.
Did that make things excessively more difficult?

Steve

Not really no. If you want the rails to sit flat then you have to mill the extrusion anyway particularly with 15mm rail on 8mm slot.

When you set the rail you will need a setting tool to offset the reference edge of the rail from the edge of the profile, I am making the assumption that the extrusion itself is straight. The setting tool with be a pair of L shapes milled together so they are identical with a hole so they can be bolted to the T slot. The short part of the L will then form a reference surface to locate the rail.

Note I'm referring to the rails on the gantry as the Y rails.

To get the separation of the Y bearing plates perfect is a little more involved since it is very hard to accurately measure the extrusion depth, again this is a nominal dimension when you purchase the profile, mine is nominally 160mm but is infact around 159.8mm, and now a bit less after I milled the flat for the rail.

So I slot joined the lower Y bearing plate to the rear Z plate but I didn't slot joint the upper Y bearing plate, I think this would have been impossible. Rather, with the Y bearings on the rails and bolted to their plates, I assembled the lower plate onto the Z axis and tightened. I then clamped the upper plate with a long clamp across the back of the gantry tensioning it against the lower plate and slightly pre-loading the bearings, then spot drilled through the holes in the Z plate into the upper Y bearing plate.

This means you only have to locate the holes in the rear Z plate to around 0.5mm accuracy...also should note that I'm using M6 bolts for these joints and I drilled 6mm holes, the 0.2 or so play of an M6 bolt in a 6mm hole is generally sufficient to align things if you are milling accurately, you don't need 6.5mm holes for M6 as is usual, although I could only do this as I counterbored with a 12mm endmill in the milling machine rather than an M6 counterbore which would have required 6.5mm holes.

I think it is much better with the rails top and bottom not just because of the increased distance between them but because of the reduced spindle overhang. My last machine which was much larger than this had the rails on the gantry face and although it was very simple to mount them it was less stiff due to the overhang.

[JAZZCNC]

Nice work Joe!
I see you are milling small slots to locate perpendicular sections to each other...... I enquired about doing just that in another forum with almost no reaction.
It seems to make sense to me ( no expert) What better way to ensure components are aligned correctly?

How have you found the process? A good idea or problematic?

I'm considering doing the same thing...when I eventually start my own build.

I see you have maximised the distance between your x rails, by placing them top and bottom of the Gantry.
Did that make things excessively more difficult?

Steve

Got to be careful here because it can and often does actually does make things harder for your self and not always more accurate by milling slots etc.? Reason is unless you can accurately machine all the major areas of the machine accurately then you will need some room for adjustment.

If you have read the forum you'll see I've said it many times that at DIY level the key to a great machine is to build in as much adjustment as possible. Nothing as changed in this statement.
Groves etc that are fixed will limit your ability to adjust out any errors coming from other areas that haven't been done to the same level. This is when you regret doing it.
The larger the machine the more this adjustability comes into play.

A better way I find at DIY level is not to limit adjustment with slots etc but use dowel pins that can be drilled and reamed after the machine adjustments have been made. This locks the machine in place but doesn't limit adjustment.

I would only consider milling slots etc if you can build the whole machine to the same level.! . . . . If not you will regret limiting the adjustments.

[devmonkey]

Got to be careful here because it can and often does actually does make things harder for your self and not always more accurate by milling slots etc.? Reason is unless you can accurately machine all the major areas of the machine accurately then you will need some room for adjustment.

If you have read the forum you'll see I've said it many times that at DIY level the key to a great machine is to build in as much adjustment as possible. Nothing as changed in this statement.
Groves etc that are fixed will limit your ability to adjust out any errors coming from other areas that haven't been done to the same level. This is when you regret doing it.
The larger the machine the more this adjustability comes into play.

A better way I find at DIY level is not to limit adjustment with slots etc but use dowel pins that can be drilled and reamed after the machine adjustments have been made. This locks the machine in place but doesn't limit adjustment.

I would only consider milling slots etc if you can build the whole machine to the same level.! . . . . If not you will regret limiting the adjustments.

Wise words about infinite adjustment.

I've built 4 machines in the last 20 years and milled slots in all of them for plate-plate, albeit these were all done on CNC not a manual mill. My approach has always been to make each sub-assembly locally square/aligned and only adjust between sub-assemblies (usuallly by shimming bearings). For example my gantry end plates are squared to the gantry and X bearing plates in 3 dimensions helped by milling the slots, this complete gantry/Y assembly will always be locally square assuming I can setup a milling machine properly to take a straight cut which I can. However the gantry itself could be twisted, unlikely but wont be known until final assembly, in which case X bearing interface with X plate will need shimming.

Sometimes you can't accurately determine a location for a joint until assembly in which case I estimate it and drill the pilot holes for the bolts then spot through to the joining plate with a centre drill once assembled, this was the case with the upper Y bearing plate on this machine. I don't use slots for this type of joint. However the slot in the bottom Y bearing plate guarantees that the two Y bearing plates (and therefore the Y bearings) are exactly the same distance behind the Z axis.

If I was using a hand router to cut aluminium as some brave chaps on here do then I definitely wouldn't be milling slots.

[devmonkey]

Once I fit the gantry end plate assemblies to the gantry I wont be able to mount the Z assembly without removing one of them, therefore I think this it is time to align and tighten the Y rails to the gantry in their final location.

For this I have just machined some precise setting tools, they were machined as a pair to ensure they are identical.

26988

Here they are attached to the gantry, they bolt into the T-slot. Plan is to snug them up, the milled faces should then act as a reference surface to match up the reference edge on the hiwin rail, they will be moved along for each rail bolt. Note that I made them long enough so they extend across two of the high points that are on the reference plane on the face of the gantry extrusion.

26989 26990

First gantry end assembly bolted in place, I used countersunk M8 bolts rather than cap head to reduce the amount of material removal for the counterbore, meant the countersinks had to be done on the mill as there is very little adjustment vs a cap head. As well as the 8 M8 bolts into the end of the extrusion there are also two M6 bolts bolting up through the X bearing plate into the two T-slots on the bottom of the gantry.

26983 26984

Bottom view showing bolts fixing up into extrusion:
26985

Plate-plate joint using the alignment slot:
26986

[JAZZCNC]

I've built 4 machines in the last 20 years and milled slots in all of them for plate-plate, albeit these were all done on CNC not a manual mill.

This is my point unless you have the tools and knowledge to use them then it's better that you don't limit adjustment and instead lock it down when setup.

As you probably know Mr. Enstein comes into play when setting up a CNC machine.? Any adjustment usually an equal and opposite reaction somewhere else and you soon end up chasing errors around the machine. This is when you appreciate the adjustment you built into the machine, esp on large machines because small errors get big very quickly as it gets wider and longer.!

[devmonkey]

As you probably know Mr. Enstein comes into play when setting up a CNC machine.? Any adjustment usually an equal and opposite reaction

Mr Newton I believe.

Yeh don't mill slots unless you have a mill.

[devmonkey]

Y rails are now fixed in their final position using the setting tools described above, worked very well and was super quick to do. Both gantry end assemblies are aligned and bolted to check final fit, everything looks good. Note rubber mallet, this is a precision build honest ;-)

26987

Other than ballscrews everything from the X plates upwards is now complete, will move back to the steel frame next, need to decide whether to pour epoxy or attempt to rig up a mobile surface grinding contraption to bring the frame into plane.

Jazz when you fettle beams holding the X rails into plane do you just scrape by hand checking with a precision level and blueing off a surface plate large enough to span the two X beams? Or do you just level each beam locally then shim one side to bring into plane? My last larger machine had X beams that could be adjusted independently up and down with the rail attached, this one has beams welded to the frame.

Cheers, Joe

[JAZZCNC]

Mr Newton I believe.

That's the bloke, another clever sod. Lol

[NordicCnc]

26987


Is there any pros/cons to mounting the x-axis linear rails on top and bottom, compared to mounting both on the front side? One thing that I can think of is:

- Mounted on front side is easier to get aligned but will not use table area as efficiently as top/bottom mounted rails. Any difference in stability?

[Voicecoil]

Both on the front tends to reduce spacing between them, hence more leverage for a given cutting force - but this leverage is in the plane of the screw fixing. It also pushes the spindle out from the gantry which could increase the twisting moment on the gantry. However it does increase clearance under the gantry and can allow a greater range of movement in Z. Top and bottom has greater spacing and hence less leverage on the rails, but it's then sideways which might not be so good unless you've milled slots as Joe has in this build. And you'll lose something like 30-45mm clearance under the gantry depending on the size of rails & thickness of mounting plate you're using. So there's pros and cons both ways, in the end it will all depend on your design as a whole.

[JAZZCNC]

Is there any pros/cons to mounting the x-axis linear rails on top and bottom, compared to mounting both on the front side? One thing that I can think of is:

- Mounted on front side is easier to get aligned but will not use table area as efficiently as top/bottom mounted rails. Any difference in stability?

Don't lose any sleep over either choice, choose whichever suits your requirements best because the difference is minimal and both will work fine provided you got a good strong gantry foundation.
I've built several machines using both methods and there is no obvious difference in cut quality or performance so build to suit your needs.

[NordicCnc]

Don't lose any sleep over either choice, choose whichever suits your requirements best because the difference is minimal and both will work fine provided you got a good strong gantry foundation.
I've built several machines using both methods and there is no obvious difference in cut quality or performance so build to suit your needs.The gantry foundation will be a 160x80 aluminium extrusion and the sides will be raised so that the x-axis is as close to the y-axis as possible. Right now my design ia with front mounted rails but I figured I am loosing quite a bit of work area because of it. Do you think it would be sufficient to change it to top and bottom mounted?

Skickat från min SM-A530F via Tapatalk

[NordicCnc]

Both on the front tends to reduce spacing between them, hence more leverage for a given cutting force - but this leverage is in the plane of the screw fixing. It also pushes the spindle out from the gantry which could increase the twisting moment on the gantry. However it does increase clearance under the gantry and can allow a greater range of movement in Z. Top and bottom has greater spacing and hence less leverage on the rails, but it's then sideways which might not be so good unless you've milled slots as Joe has in this build. And you'll lose something like 30-45mm clearance under the gantry depending on the size of rails & thickness of mounting plate you're using. So there's pros and cons both ways, in the end it will all depend on your design as a whole.Thanks for this very detailed answer. So to draw a conclusion for this is that if you are more concerned about work are that Z-axis clearance/travel, one should go for the top/bottom mounted rails, since it will most likely result in a little bit stiffer construction?

Skickat från min SM-A530F via Tapatalk

[JAZZCNC]

Do you think it would be sufficient to change it to top and bottom mounted?

My last post said it all really. Do which suits your needs best.

[Voicecoil]

You needn't lose too much work area, but there will always be a certain zone where the cutter cannot reach (unless you make your Y axis rails longer than the bed). This is because you want to have a decent distance between the carriages on your Y axis for stability (I went for 200mm IIRC), hence the machine will be at least this much longer than your cutting area.I arranged things on mine so the cutter reaches to within 30mm of the front of the bed, with the "dead space" behind.Decide on what X, Y and Z capability you want then as Jazz said choose whichever layout fits it most convieniently.

[Kitwn]

Something else to consider is the complexity of the Z-axis back plate. Rails on the face can allow this to be a simple flat plate whereas rails on top and bottom will require two accurate and rigid square joints.

[Sterob]

Good point..... 'Complexity and Less Overhang' verses 'Simplicity and more Overhang'...

[devmonkey]

Good point..... 'Complexity and Less Overhang' verses 'Simplicity and more Overhang'...

If you have access to a milling machine I would definitely put them top and bottom. If you want to mount them on the face and keep the same rail separation then you need a deeper gantry, if you use 160x80 profile for the gantry then the Y rail centres will *only* be 120mm apart (distance between outermost slots), the same profile with top-bottom mounts gets you ~180mm centres. On my last machine I used 160mm ali box section for the gantry and with face mounting close to the edges got around 140mm centres.

[NordicCnc]

If you have access to a milling machine I would definitely put them top and bottom. If you want to mount them on the face and keep the same rail separation then you need a deeper gantry, if you use 160x80 profile for the gantry then the Y rail centres will *only* be 120mm apart (distance between outermost slots), the same profile with top-bottom mounts gets you ~180mm centres. On my last machine I used 160mm ali box section for the gantry and with face mounting close to the edges got around 140mm centres.So do you think the difference with 40-50mm extra offset will make any notable difference to stiffness?

Skickat från min SM-A530F via Tapatalk

[devmonkey]

So do you think the difference with 40-50mm extra offset will make any notable difference to stiffness?

Skickat från min SM-A530F via Tapatalk

Say your overhang to spindle centre from gantry centre (assuming a rectangular gantry profile) reduces from 150mm to 100mm as a result of mounting rails top/bottom then you have decreased the torque applied to twist the gantry to 2/3rds for a given force applied to the spindle, so it really depends on the torsional stiffness of your gantry and the spacing of X bearings. Ideally the front X bearing wants to be at or forwards of the spindle so usually face mounting the Y rails increases the X bearing spacing, this is no bad thing but it does mean the machine needs to be longer to accommodate the same X travel.

For reference the difference in gantry twist for a given spindle force between when the Z is fully up and fully down will be greater than the difference in twist saved by shifting the spindle 50mm backwards.

As everyone says on this forum it depends on what you want to do with the machine as to how stiff it needs to be, having built previous machines that weren't stiff enough I now make every design choice around increasing stiffness. My choice to top/bottom mount the rails was both to increase stiffness and produce as compact a machine as possible as it is going into my garage so I don't have to traipse down to the large workshop in the woods in mid winter.

[NordicCnc]

Say your overhang to spindle centre from gantry centre (assuming a rectangular gantry profile) reduces from 150mm to 100mm as a result of mounting rails top/bottom then you have decreased the torque applied to twist the gantry to 2/3rds for a given force applied to the spindle, so it really depends on the torsional stiffness of your gantry and the spacing of X bearings. Ideally the front X bearing wants to be at or forwards of the spindle so usually face mounting the Y rails increases the X bearing spacing, this is no bad thing but it does mean the machine needs to be longer to accommodate the same X travel.

For reference the difference in gantry twist for a given spindle force between when the Z is fully up and fully down will be greater than the difference in twist saved by shifting the spindle 50mm backwards.

As everyone says on this forum it depends on what you want to do with the machine as to how stiff it needs to be, having built previous machines that weren't stiff enough I now make every design choice around increasing stiffness. My choice to top/bottom mount the rails was both to increase stiffness and produce as compact a machine as possible as it is going into my garage so I don't have to traipse down to the large workshop in the woods in mid winter.So what I plan to use the machine for is mostly wood and aluminium milling and occasionally very very light stel milling. Based on what you and everyone are saying I am leaning towards top/bottom mounted rails since stiffness is critical.

Have you built any machines with face mounted rails and 160x80 extrusion gantry that was not stiff enough to mill aluminium efficiently?

Skickat från min SM-A530F via Tapatalk

[devmonkey]

So what I plan to use the machine for is mostly wood and aluminium milling and occasionally very very light stel milling. Based on what you and everyone are saying I am leaning towards top/bottom mounted rails since stiffness is critical.

Have you built any machines with face mounted rails and 160x80 extrusion gantry that was not stiff enough to mill aluminium efficiently?

Skickat från min SM-A530F via Tapatalk

No every machine i've built, including one that was substantially made of MDF with an underslung gantry rail with only a single igus carriage on each X rail on which you could rock the gantry visibly by leaning on the spindle, were capable of machining aluminium. You just need to back off the speed and depth of cut if you lack stiffness, all the accuracy is delivered on the finishing cuts anyway, just as with a normal mill, tool deflection particularly with the thin tools we use in routers is usually as much or more of a problem. That said it is easier to produce accurate work on a stiffer machine, so depends what you are making. If it is machining a few plates out with accurate vertical edges and holes that is one thing, if you are running 3D paths to build molds etc then you need a stiffer machine, and a machine that is significantly stiff over the Z travel you need.

Note though that if you want the rails to sit flat on extrusion it does have to be milled irrespective of which face you bolt them to. I don't think many people do this on this forum but most still get completely acceptable results.

[JAZZCNC]

Have you built any machines with face-mounted rails and 160x80 extrusion gantry that was not stiff enough to mill aluminum efficiently?

Very few on this Forum can answer that question regards 160x80 profile because they won't have needed to change if done one before the other.!

I've built 5 machines that use face mounted and 14+ machines that use top/bot and all cut aluminum perfectly fine. These machine sizes range from 4 x 2 to 10 x 5 and the 6 machines with face-mounted are all 8 x 4 or 10 x 5 machines because saving space wasn't an issue. So in this case, I went for the easiest method of fixing and setup, also fewer parts to machine.

Honestly, it's not something you need to worry about on router that's mostly going to cut wood with occasional aluminum work.

One other method I use often, esp when needing to keep gantry low but with the highest clearance is to mix n match bottom rail on the front, top rail on the top. This gives the most clearance under the gantry and keeps gantry height low.

So like I say use the method that suits your needs because no one is better or worse than the other. You won't see any difference when using the machine.

[NordicCnc]

Very few on this Forum can answer that question regards 160x80 profile because they won't have needed to change if done one before the other.!

I've built 5 machines that use face mounted and 14+ machines that use top/bot and all cut aluminum perfectly fine. These machine sizes range from 4 x 2 to 10 x 5 and the 6 machines with face-mounted are all 8 x 4 or 10 x 5 machines because saving space wasn't an issue. So in this case, I went for the easiest method of fixing and setup, also fewer parts to machine.

Honestly, it's not something you need to worry about on router that's mostly going to cut wood with occasional aluminum work.

One other method I use often, esp when needing to keep gantry low but with the highest clearance is to mix n match bottom rail on the front, top rail on the top. This gives the most clearance under the gantry and keeps gantry height low.

So like I say use the method that suits your needs because no one is better or worse than the other. You won't see any difference when using the machine.Okay thanks for the information. I might just go with the top/bottom because to me it seems more rigid that way with larger spacing between the linear rails. I think I will be cutting mostly aluminium, some wood and occasionally steel with very light cuts.

Skickat från min SM-A530F via Tapatalk

[Kitwn]

As Joe said "if you have a milling machine...".

Perhaps the really clever trick is to design your gantry to take both front and top/bottom mounted rails. Front mounting the rails with a simple Z axis plate would then give you a machine which may be good enough to machine the parts needed to then convert it. Let the machine pull itself up by it's own boot laces.

Kit

[Sterob]

I have been thinking about this. I like the idea of placing the rails on the side of the extrusion( for simplicity), but I would like the wider spacing top and bottom mounting gives.( for ridgidity )

How about using 2 smaller extrusions, spaced apart so the ball screw will fit in between them, and use a say a 10mm Ali backing plate to stiffen the extrusions?
That could bolt to the Gantry arms as well?

I guess aligning 2 extrusions may be a bit of work?

Been thought of already?...lol

27039

[pippin88]

It's been done plenty of times. It's weaker than a large section.

[devmonkey]

Progress update!

I got back to the project after a break of a few months to work on other things. I ended up pouring epoxy to bring the X rails into plane, then set the master rail straight using a taut wire, microscope and some software as described here:
http://www.mycncuk.com/threads/12966-DIY-Laser-levelling-using-webcam-and-laser-level/page27

I now have both rails mounted and the gantry moves very smoothly.
28236

Next step is to fabricate the motor brackets and mount the ballscrews and steppers. I've ordered belts and pulleys and am waiting on these so in the meantime I started to fabricate a stand for the machine. Since the machine is itself a rigid structure this stand only serves to anchor it to the floor at a decent working height, hence the lack of bracing. All 4 legs will be bolted to the concrete floor then the machine will sit ontop on leveling bolts, finally the machine frame will be clamped hard to the stand so that it doesn't throw itself across the room (voice of experience...).

Cut and welded yesterday, painting in silver hammerite today. The more I use it the more I love this little MMA stick welder. welded the 3mm box section mitres at only 45amps! Almost as good as a MIG without any complications.
28237 28238 28239

[devmonkey]

I'm looking for some recommendations on bed materials, machine will mainly be used for aluminium and composites, so preferably something that can take coolant. My last machine used epoxy soaked MDF which was far from ideal. The plate would need to be 800x1300.

I don't have a large enough piece of aluminium plate. So options I'm looking at are:

1. 20mm Tufnol, can't find this in sheets longer than 1200 so would have to alter the frame, this isn't really a problem. I'm not sure whether this is rigid enough to span the cross beams that are on 400mm centres??

2. Purchase a piece of 20mm aluminium plate large enough, quiet expensive.

3. Locate some aluminium extrusion T-slot bed profile 'planks', I can't find these either.

Thoughts?

[devmonkey]

I'm looking for some recommendations on bed materials, machine will mainly be used for aluminium and composites, so preferably something that can take coolant. My last machine used epoxy soaked MDF which was far from ideal. The plate would need to be 800x1300.

I don't have a large enough piece of aluminium plate. So options I'm looking at are:

1. 20mm Tufnol, can't find this in sheets longer than 1200 so would have to alter the frame, this isn't really a problem. I'm not sure whether this is rigid enough to span the cross beams that are on 400mm centres??

2. Purchase a piece of 20mm aluminium plate large enough, quiet expensive.

3. Locate some aluminium extrusion T-slot bed profile 'planks', I can't find these either.

Thoughts?

A bit more research on different materials, Young's modulus (GPa):


MDF
4


SRBP (paxolin/tufnol)
6.5


Aluminium
69


Steel
200



The deflection of a rectangular section spanning two supports due to a downward force in the middle is inversely proportional to the Young's modulus * thickness^3.

So taking 16mm aluminium plate as a reference, thickness required for equivalent rigidity of those materials are:
MDF 41mm, SRBP 35mm.

Or another way, 25mm SRBP is equivalent to 12mm aluminium.

From this it is probably ok to conclude 25mm SRBP supported on 4 sides over rectangular bays in the machine frame that are 840x320mm is probably sufficient.

[devmonkey]

A bit more research on different materials, Young's modulus (GPa):


MDF
4


SRBP (paxolin/tufnol)
6.5


Aluminium
69


Steel
200



The deflection of a rectangular section spanning two supports due to a downward force in the middle is inversely proportional to the Young's modulus * thickness^3.

So taking 16mm aluminium plate as a reference, thickness required for equivalent rigidity of those materials are:
MDF 41mm, SRBP 35mm.

Or another way, 25mm SRBP is equivalent to 12mm aluminium.

From this it is probably ok to conclude 25mm SRBP supported on 4 sides over rectangular bays in the machine frame that are 840x320mm is probably sufficient.

I would probably use one large piece as the bed, then a second piece as the sacrificial board.

[JAZZCNC]

30mm HDPE makes a nice stable bed and you can cut a grid pattern to make it into a Vacuum bed. I bought a 10x5 sheet for £500 so if you get in touch with plastics supplier you might get a small piece or cutoff piece cheap enough.

[devmonkey]

30mm HDPE makes a nice stable bed and you can cut a grid pattern to make it into a Vacuum bed. I bought a 10x5 sheet for £500 so if you get in touch with plastics supplier you might get a small piece or cutoff piece cheap enough.

Ok thanks, that is interesting, YM of HDPE is only 0.8, how much support did you have under it, or what was the clear span? I guess we are only talking about supporting the mass of the part being machined and the downward force of the cutter, this is probably minimal, like to take a punt on what this force is?

Looks like an 8x4 sheet of 25mm SRBP is £380, this would give me enough for the base board and two spoil boards.

[devmonkey]

Stand levelled on floor and machine craned ontop to check final position, needs to be moved out from the wall a bit so I can fit/service the ballscrews motors. Starting to look like a CNC machine!
28249

[Kitwn]

Looks seriously serious! Make sure you have enough room to move around it while bending over and enough room to fit a screwdriver in all the places you didn't think you'd need to fit a screwdriver when you first thought "I'll put it here!".

[devmonkey]

Finished making the last set of plates, these are for the X and Y axis, ballscrew mounting. Hopefully this is the last of the aluminium work on the manual mill. I still need to make some stepper mounting plates but leaning towards doing those in 3mm steel.The white parts are 3D printed standoffs for the X axis ballnut holders, I didn't have any large enough ally stock so we will have to see how they fare. They get clamped between two aluminium lumps so should be ok, if not will replace when I have some bar stock.

28250 28251

[devmonkey]

Any they all fit:
28252 28253 28254

[devmonkey]

Was still waiting for the pulleys this morning so decided to align the Z axis assembly and fit the ballscrew. I needed a 0.25mm/10thou shim between the ballnut holder and Z-plate, but it was extremely difficult to get the shim into place as the holder is buried between the rails inside the assembly. A bit of lateral thinking and I used some plastic lamination film cut from a normal A4 laminating pouch.

This stuff has the advantage of a) being quite hard after lamination,b) having a heat activated adhesive. So using an old iron I laminated 4 layers of the stuff onto the back of the ballnut holder, this formed the perfect shim. I trimmed it with a scalpel and cut the bolt holes. Worked really well. Z axis is now complete apart from lubrication points. These are a pain with the nipples supplied with the hiwin carriages and the ballscrew so will need some thinking.

28265 28266 28267 28268

Just as I finished the Z the delivery guy turned up with my belts and pulleys, so I spent a rather boring couple of hours, boring, reaming, cross drilling and tapping 8 steel pulleys. All done now. I only installed one M5 grub screw per pulley, each shaft will have a flat on it, hopefully this will be sufficient. I could do with some longer grub screws, or maybe just use an M5 bolt given the low RPMs.

28269 28270 28271 28272 28273

[devmonkey]

Made the motor mounts today, 3mm steel cut from some scrap box section. I used a cheap screwfix holesaw, wasn't at all convinced it would work. Lots of cutting fluid and smoked like crazy but delivered a very clean hole, cutter seems like new still. Just cleaned them up on the linisher and ripped the end of my finger nail off, very messy.

28274 28275 28276

They are now painted.
28278

For the machine bed I got a few quotes for 30mm HDPE, various Phenolic paper boards, etc, all around £300. I called Smiths for a quote on 1300x880x20mm aluminium plate, quote was £274, then they found a pre-cut piece 1500x990x20 which I can have for the same price. Seems a no-brainer so am picking it up tomorrow.

I will get a 1200x600 piece of phenolic for the spoil boards, these are a standard size and just right for my machine and readily available from electrical wholesalers.

[devmonkey]

Picked up the bed today, 82kg of 20mm plate! Not sure how I'm going to cut it, too heavy for manoeuvring over the table saw. Will probably just use the circular saw very carefully...
28283

[devmonkey]

Duplicate post.

[devmonkey]

Bit of a set back today. I noticed that the Z screw was operating smoothly in one direction but lightly jammed and clicked when the direction was reversed. I've taken it off of the machine and now believing there must be a ball trapped between two circuits, however I've now re-packed it twice and the exact same problem persists. I think there is probably a small error where returns are machined leaving a tiny gap that allows one outer circuit and the middle circuit balls to touch. The plastic returns didn't look to clever to me.

Anyway I've contacted Fred from BST to see what he suggests.

[devmonkey]

Fred is sending a replacement, what an excellent after sales service. In the meantime I've ordered a ball nut from Cardiff for £11 that should arrive tomorrow so I can finish the machine.

[devmonkey]

My spoil board choice solved itself today. I had to take the car in for a firmware update (I kid you not) and whilst I as waiting I noticed a small plastic engineering shop over the road. Knocked on their door and the owner was absolutely flat out making perspex screens for covid protection in shops and offices, he said he was 10x busier than he'd ever been before although there was now a world shortage of perspex, well done him!

Anyway I asked about foamed PVC and he went into his warehouse and pulled out a 19mm offcut 1300x700 which he sold me for £30, perfect!
28303

[Nealieboyee]

Lovely machine. May I ask which aluminium you bought from Smiths? They have loads of different tooling plate names. Alplan, certal, contral etc.

Also, You have mounted your ball screw supports directly to the steel box section. How did you make sure they were aligned and in the same plane? Surely the box section wasn't that flat?

[devmonkey]

Lovely machine. May I ask which aluminium you bought from Smiths? They have loads of different tooling plate names. Alplan, certal, contral etc.

Also, You have mounted your ball screw supports directly to the steel box section. How did you make sure they were aligned and in the same plane? Surely the box section wasn't that flat?

This time I got a piece of KASTAL 300 from Smiths, but tbh all I do is phone my local branch and ask what they have ready to go, so long as it is cast milled plate.

The rails are set on epoxy what was cast in plane. The ballscrew supports are bolted to the steel and shimmed with the ballnut hard mounted to the gantry so it aligns perfectly. I. E. fit ball nut before end supports.

Edit:
This is what I use for shimming, £15 for a pack of different thicknesses from RS, works really well.
28306

[devmonkey]

I'm waiting on some proximity switches to arrive. In the meantime I've started to work on the control side, my enclosure with psu and drivers is finished so this is about the actual machine control

In the past I've always used MACH and an old PC, this time I'm going try using a standalone DDCS controller, no experience of these other than checking it powers up.
28307

The are 4 axis controllers (there is a thread on here somewhere) but have no capability to slave two axis together. My solution is to add a daughter board that will both slave two drivers and handle the gantry squaring. I don't think such a thing is available so I've started to build one. The basic system is an arduino nano with appropriate opto isolation that sits between the controller X outputs and X home input and the two X drivers and two X home switches.

The (untested) algo to deal with this is as follows. Under normal operation the arduino will mirror the X signals to drivers X1+X2. When a home switch is triggered AND the machine is moving in the homing direction the system will enter a homing state. It will suppress further steps to the motor whose home switch triggered whilst continuing to pass thru steps to the other motor. Once both home switches have triggered it will signal home to the controller. The controller will then back away from the home switches, the system will then perform the opposite logic and suppress steps to the motor whose switch releases first until both switches are released then clear home to the controller and return to the normal state.

The system never generates its own steps rather it chooses when to pass through steps from the controller, the hope is that it can sit passively and just do its job without the operator or controller having to do anything special. There will be a configurable max overdrive/underdrive number of steps that if exceeded will fault the estop. It will also be possible to set offsets for each motor home position so that the axis can be squared up without moving proximity switches or targets.

The DDCS has a sensible power domain isolation between the logic and stepper outputs on one side and home/limit inputs and outputs on the other side. My system will honor this isolation, on the switch side using opto isolators on the COM+/- domain. The stepper signal side will have a ground shared with the DDCS logic side, the stepper signals themselves are just 5v differential logic so can be connected directly to the arduino pins.

I've written the code, knocked up the schematic and just made a quick PCB so will assemble and test for when my home switches arrive.

28308 28309 28310

There is also a tiny oled screen and a couple of buttons for configuration.

Once I've validated it works as intended the code can be shared.

[Doddy]

Interesting idea, but I'd look at the pulse output frequency of the DDCS and your intended stepping rate, then the frequency response of the PC817s and the effective throughput you'll get on a 16MHz '328. You are limiting the performance of the machine. At the least I'd look to replace the opto's. Pass-through performance can be easily added with adding a 74-series logic device to allowing DDCS signalling to pass-through, or to augment from the Ardy.

Of course, your mileage might vary.

EDIT: not suggesting this, but have a look at the ESP32s - particularly with the onboard OLEDs, having a dual-core processor at 240MHz meant for a very easy protocol converter with fancy graphic display (was converting RS485 to RS232 at 1.2MHz, with one core dedicated to the graphics and one just copying from UART buffers)... similar but different kind of solution.

[devmonkey]

Clock of DDCS is 500khz, my code mirrors directly off the port so low overhead compared to arduino libs. I will verify propagation delay with the scope. PC817s are only on switches not step/dir, this is also the opto inside the DDCS, i've just had a look. Agree easy to build part of it with multiplexers but I'm not sure it is required. I've got a box full of ESPs here, again not required for this.

With 16x ustepping and 10mm pitch screws you are at 320 steps/mm so 500kHz gives 1500mm/s and a screw speed of 10k rpm, we only need 1/10th of this tops so I don't think frequency will be an issue.

[Doddy]

Interested to see how this goes. Glad to hear you're bit-banging, though as the 328 is inline with the axis pulses to the stepper drivers it has to keep up at all time (not just homing) and I'd be curious if you can hold the signalling whilst I2Cing the display - interesting challenge.

I did realise after writing that the optos were for homing switches/reflection into the DDCS, so, yeah, will be fine - I'll withdraw that.

[devmonkey]

Interested to see how this goes. Glad to hear you're bit-banging, though as the 328 is inline with the axis pulses to the stepper drivers it has to keep up at all time (not just homing) and I'd be curious if you can hold the signalling whilst I2Cing the display - interesting challenge.

I did realise after writing that the optos were for homing switches/reflection into the DDCS, so, yeah, will be fine - I'll withdraw that.

Yeh the I2C is the dubious part, albeit it is done in hardware, I haven't coded that yet, if it gets in the way it can be omitted. I just usually add them to all my boards these days as I have a drawer full and they only cost a few quid. In this case I had disable the uart to use the rx/tx pins for outputs so without I2C there is no way to get any debug out whilst testing. I did this so I could get a full 8 bit synchronous port write to the drivers rather that making multiple consecutive writes. Every true IO pin is in use ;-)

[Kitwn]

A very interesting project, especially as I was well on the way to completing something similar when I decided to put on my grown up trousers and upgrade my LinuxCNC version instead.
My version planned to have a board in the middle of the parallel cable from the PC to the BoB and used logic gates to select whether stepper control pulses were fed from the PC or the Arduino Nano (you just can't get away from them can you?!) and limit switch pulses fed back. When you selected gantry squaring the PC was left thinking the machine was idle and the Nano produced drive pulses which were gated to the two motor drivers by de-bounced and latched signals from the limit switches. Once that sequence was complete control was handed back to the PC for normal homing.

Under normal operation the stepper drive signals were only going through a couple of extra logic gates so no data speed problems were added and the timing of separately stopping the two motors was to be done by the discrete logic with no software delay to add timing errors. The nano was there to select who had control, generate step pulses at the two required speeds and set the direction of movement for the whole sequence. One of it's great weaknesses was that it relied on mechanically adjusting one of the end stops to set the squaring, though it would only have needed a few extra lines of code and maybe a gate or three to fine-adjust from the software.

I designed and built the whole thing including laboriously cutting and buzzing out both ends of a parallel cable but never actually put it into use having realised in the mean time that an upgrade of LinuxCNC was not as difficult as I'd first thought.

I look forward to seeing your video of the beast in action.

Kit

[devmonkey]

Hi Kit,

I still have a PC with a MACH license on it to revert to if this doesn't work. Just want to try and remove all the associated cruft (keyboard, monitor, desktop) I usually have around the machine to make it work. I've now drilled and assembled the board, tested connectivity then lacquered the copper side (just in case it is a keeper!).

28311 28312 28313

I need some more solder paste mine is about 5 years past its best and someone has bent the ends of my nice tweezers meaning putting those resistors on was a PITA.

Cheers, Joe

[Kitwn]

It's drilling the bloody holes that's always the pain isn't it? I use SMDs as much as possible and bought a vacuum placement thingy off eBay a few years back. I keep the solder paste in the fridge, especially as inside my shed can get up towards 50C around Christmas and new Year.

I hope it works as you want.

Kit

[Doddy]

Drilling?, if only we had a machine that could do all that for us :-)

And solder paste?, I've thrown more out than I've used, with it's damned stupid shelf-life (fridge or not...and a Northern UK climate). Right now 0.4mm rosin cored solder will do me and a decently "Large" iron tip. Oh, and flux... plenty of flux.

But, yeah, DevMonkey - hope it works and please report back. I need inspiration to continue my DRO/Pi controller for my mill.

[Kitwn]

Drilling?, if only we had a machine that could do all that for us :-)


You have. It's called a CNC router. The trick is making the machine and your ability to register a pre-etched circuit board on it accurate enough. Something I haven't even atempted yet.

I do like using the solder paste. That and a gas soldering iron with a hot-air head makes hand-soldering SOIC chips nice sand easy and the 1.27mm (thats 0.05 inch for our 19th century readers) pad spacing is reliably makeable using the toner-transfer method for DIY circuit boards....without holes!!!

Kit

[devmonkey]

You have. It's called a CNC router. The trick is making the machine and your ability to register a pre-etched circuit board on it accurate enough. Something I haven't even atempted yet.

I do like using the solder paste. That and a gas soldering iron with a hot-air head makes hand-soldering SOIC chips nice sand easy and the 1.27mm (thats 0.05 inch for our 19th century readers) pad spacing is reliably makeable using the toner-transfer method for DIY circuit boards....without holes!!!

Kit

For crude through hole boards with thick traces like this one I use the toner transfer method with a £50 Brother laser printer from Argos, good results and takes <15 mins to make a board including etch and drill, I drill by hand on a tiny toy CNC machine. This process works well for double sided boards as well.

28335

For higher resolution boards I use a CNC photo method on the same toy machine. Spray the copper with black paint then use a low power UV laser in the machine to isolation route it. Wash off the ablated paint then etch. If the board requires any drilling and/or is double sided I use a fixture so to re-locate it after etch. This process takes a bit longer but I can achieve 0.2mm features/tracks.

Anything more complicated with >2 layers gets sent to China, even if just a proto. All the SMD stuff gets soldered with hot air, agree very easy if your solder paste isn't like gritty chewing gum.

[devmonkey]

Interested to see how this goes. Glad to hear you're bit-banging, though as the 328 is inline with the axis pulses to the stepper drivers it has to keep up at all time (not just homing) and I'd be curious if you can hold the signalling whilst I2Cing the display - interesting challenge.

I did realise after writing that the optos were for homing switches/reflection into the DDCS, so, yeah, will be fine - I'll withdraw that.

Hi Doddy,

So it turns out if you ditch the standard I2C/TWI wire.h library and display driver, crack open the 328p datasheet, write your own crude framebuffer and I2C non-bocking state machine that operates directly on the TWI registers you can tick the state machine at a 2-3us per tick penalty. Will put it on the scope and see if this is sufficient.

Main loop becomes:
1. Mirror stepper/home signals with squaring logic.
2. Update framebuffer if needed (using a 5x7 font this is 5 memory reads/writes per character)
3. Tick I2C state machine.

I could remove the framebuffer to make it a bit faster if required, this will effectively amortise the font copy part at one byte/I2C tick.

Cheers, Joe

[Doddy]

Hi Doddy,
write your own crude framebuffer and I2C non-bocking state machine

"Crude"?, I prefer to think of this as "fit-for-purpose".

You know as well as I do the 328 is just one of a thousand different micro controllers that pollute the oceans and the Arduinio packaging is just a convenient way of throwing that chip into an accessible form for stupidly low cost - makes them a sensible solution for even the most simple of problems. There's no reason to use the Arduino libraries unless you're concerned about cross-platform compatibility or speed/ease of development (sacrificing code base, efficiency and real-time performance). I think you'd agree programming at the register level (I gave up at the instruction level many decades ago - I prefer to have a compiler at least) adds a level of complexity and care, but if that floats-your-boat (and it does, me) then hack away to your hearts content. At least you can be confident that you know exactly that the uC is behaving exactly as you want it to.

[devmonkey]

So I've just scoped the DDCS. An axis setup as per my machine, 1610 screw, 16x ustepping -> 320 steps/mm, max speed 10000mm/min.

As expected step frequency is ~53kHz, half period of 9.375us. Nyquist/Shannon tell us that this needs to be sampled at a minimum of 2f, i.e. 106kHz for perfect reconstruction. Puts an upper bound on my mirror code latency of ~9us, so to be safe say 5us. This is 80 cpu cycles, will it fit?? Should do, got to wire it in and test.

DDCS defaults set dir 7000ns (7us) ahead of step therefore code running at 5us will always see this. EM806 only requires 2.5us but configuring DDCS to deliver this will risk breaking the contract woth EM806.

[Doddy]

Interrupt on change of state? Your time-sensitive processing is in the discrete control and not the I2C if you're using the TWI registers.

[devmonkey]

Interrupt on change of state? Your time-sensitive processing is in the discrete control and not the I2C if you're using the TWI registers.

Yes it is possible you can register for toggle interrupts on any of the IO pins, trouble is my code is written in arduino/cpp at the moment, this means there is a stack to park so there is some overhead from servicing the interrupt. If the current polling approach takes too long will have a look.

[devmonkey]

So here is a baseline, the basic axis mirroring with auto squaring, but without I2C. Lag is 600ns after a lot of messing about I just stuck the logic in a lookup table and deleted all my code ;-) Yellow trace is step output from DDCS running at 10000mm/min, blue trace is the arduino step output, lag is shown in 'dt' bottom right.

28337

[Doddy]

You're not going to get better than that!

[AndyUK]

after a lot of messing about I just stuck the logic in a lookup table and deleted all my code ;-)

Hahaha love it. Sometimes 'ugly' solutions are just the best.

[devmonkey]

My dual motor homing / squaring board is now working with the DDCS. Video clip below. DDCS was told to 'home' the X axis:

1. Each motor stops when its respective home switch is activated,
2. Once both home switches are activated the board signals home to the DDCS,
3. DDCS now backs X away from home searching for switch release,
4. If one motor's home releases before the other this motor is stopped until both home switches are released,
5. Once both home switches are released the board clears home back the the DDCS and DDCS zero's X co-ordinate,
6. DDC now moves X a pre-configured distance away from home.

This logic only activates when a home switch is hit and the machine is travelling towards home. If travelling away from home either switch hit signals the DDCS (i.e. it behaves as a limit switch).

The only downside of this is that in normal operation driving the machine past home in the X- direction requires both switches to activate before limit- is signalled. This is not a problem unless one of the switches has fallen off.

28364
https://youtu.be/dyFAdsQIpxA

[Kitwn]

Call me stupid, but what exactly do you adjust to make the gantry square once the whole sequence has finished? What is the purpose of step 4? Do you have separate cotrol of the two X distances in 6?

[devmonkey]

Call me stupid, but what exactly do you adjust to make the gantry square once the whole sequence has finished? What is the purpose of step 4? Do you have separate cotrol of the two X distances in 6?

Hi Kit,

Quick answer, once the sequence is finished the gantry is square.

So the normal single motor home search (used by everything from gbrl on your 3d printer, mach3 and the DDCS) is:
1. Drive axis until home switch activates (at which point you have likely gone past it)
2. Retract axis very slowly until home switch releases (you are now at 'home').

In a setup with two motors one driving each side of the gantry adjust the switches/inductive targets so that when each respective release point is reached the axis is square.

Challenge now is to 'trick' the controller that is executing the single motor home search algo above into locating the point where both switches exactly release.

This is what the logic I described does. Step 4 is required during retraction as when a motor's home switch releases that motor is at 'home', whilst the remaining motor isn't yet there so requires more steps. My board never generates steps, it only suppresses steps if required. So in this case DDCS still believes it needs further retraction so continues to generate retraction steps, however my board only passes these onto the motor that has not yet reached 'home'. When this motor does finally activate it's switch both motors are at 'home' and my board signals this to the DDCS.

If your question was due to believing that because at step 2 both motors have activated their home switch and the DDCS has been signalled meant that the motors were at 'home' then this is not true. It wouldn't work as you may have overshot due to the momentum of the gantry and the angular momentum in the screw and it is highly unlikely the hysteresis of two switches is identical. This is why everyone does the very slow retraction until release which avoids both these problems.

This whole process will re-square a machine that has gone out of square. The machine only goes out of square if a driver misses some steps due to noise, a motor stalls or if someone turned a ballscrew by hand when the machine was off.

[Scustom]

Nice work devmonkey!!
Really interesting build!

[JAZZCNC]

Call me stupid, but what exactly do you adjust to make the gantry square once the whole sequence has finished? What is the purpose of step 4? Do you have separate cotrol of the two X distances in 6?

Ok your Stupid.:hysterical: . . .But so am I because I'm not seeing it either.?

To me it's just homing each axis separately, waiting until both inputs go high and backing off until both go low then moving both to a set distance.! It's not doing any individual axis correction and as the switches don't move it's not corrected anything.? Yes, you can square the gantry by offsetting switches using this method, which is how mach3 etc do it but so can any homing system really.!

If you want a true squaring system then it needs to drive each motor separately and adjust to a set distance. For instance, the CSlabs squaring system gives you options to measure the offset between the switch triggers and gives you the option to apply a correction to a set distance or let the system do it automatically. In which case it drives both motors until the first switch goes high then backs both off until it goes low, then it drives the opposite motor until switch goes high measures the difference and moves that axis only the measured difference. It's strange to watch as it does a little dance and the gantry is then aligned to the first switch.
If you want to apply an extra offset to twist either side then you can set a +/- distance for either side which it will apply after homing. It then backs both off a set distance if required.

[JAZZCNC]

Challenge now is to 'trick' the controller that is executing the single motor home search algo above into locating the point where both switches exactly release.

This is what the logic I described does. Step 4 is required during retraction as when a motor's home switch releases that motor is at 'home', whilst the remaining motor isn't yet there so requires more steps. My board never generates steps, it only suppresses steps if required.

I do a similar thing with the Cslabs IP-M controller that doesn't home each axis individual. I just do it using a modified macro which basicly homes twice and controls the drive enable using an output connected to a relay...! . . Simplizzzs

[devmonkey]

Ok your Stupid.:hysterical: . . .But so am I because I'm not seeing it either.?

To me it's just homing each axis separately, waiting until both inputs go high and backing off until both go low then moving both to a set distance.! It's not doing any individual axis correction and as the switches don't move it's not corrected anything.? Yes, you can square the gantry by offsetting switches using this method, which is how mach3 etc do it but so can any homing system really.!

If you want a true squaring system then it needs to drive each motor separately and adjust to a set distance. For instance, the CSlabs squaring system gives you options to measure the offset between the switch triggers and gives you the option to apply a correction to a set distance or let the system do it automatically. In which case it drives both motors until the first switch goes high then backs both off until it goes low, then it drives the opposite motor until switch goes high measures the difference and moves that axis only the measured difference. It's strange to watch as it does a little dance and the gantry is then aligned to the first switch.
If you want to apply an extra offset to twist either side then you can set a +/- distance for either side which it will apply after homing. It then backs both off a set distance if required.

Hi Jazz,

When you say any homing system can do this, the DDCS cannot as you proved yourself, infact no system can unless it is aware there are two motors on a single axis, i.e it has axis slaving, my board allows me to use the DDCS with two motors to square the gantry.

I also have the correction offset logic so you don't have to manually move the switches but currently no way to update these offsets without reloading the arduino as I haven't coded it yet. So in this test they were hard coded to zero, I thought this was the simplest way to explain how it works...LOL

Cheers, Joe

[JAZZCNC]

Hi Jazz,

When you say any homing system can do this, the DDCS cannot as you proved yourself, infact no system can unless it is aware there are two motors on a single axis, i.e it has axis slaving, my board allows me to use the DDCS with two motors to square the gantry.

Well I'm not sure how I have proved this my self.? but yes I agree if the system isn't designed to slave motors independently then it can't, but most systems like Mach3 or Linux CNC that use slaved motors independently can which is more what I was meaning. Your system appeared to just be doing what they do, but they don't Auto square the gantry to a set amount, they just move to switches that you set to square the gantry. (actually, I think Linux CNC can now.!).

Anyway good job all the same, I wasn't knocking your efforts just didn't see any auto squaring(offsetting) going on only hitting switches.!

[devmonkey]

Well I'm not sure how I have proved this my self.? but yes I agree if the system isn't designed to slave motors independently then it can't, but most systems like Mach3 or Linux CNC that use slaved motors independently can which is more what I was meaning. Your system appeared to just be doing what they do, but they don't Auto square the gantry to a set amount, they just move to switches that you set to square the gantry. (actually, I think Linux CNC can now.!).

Anyway good job all the same, I wasn't knocking your efforts just didn't see any auto squaring(offsetting) going on only hitting switches.!

I'm attempting to use the standalone DDCS controller for this machine Jazz rather than a PC. Before making this custom board I read Boyan's thread on these controllers here:
http://www.mycncuk.com/threads/10187-DDSCV1-1-3-and-4-axis-offline-motion-controller

I was just quoting your posts in that thread where you tested one of these controllers and proved that it was not possible to slave motors and this obviously being a problem for machines with >1 motors per axis. Here you go:
http://www.mycncuk.com/threads/10187-DDSCV1-1-3-and-4-axis-offline-motion-controller?p=85579#post85579


Maybe your miss understanding what was meant.? It was me who said doesn't work and I'm talking about Slaved axis. As in two motor/drives on separate outputs working as one axis.
That controller is exactly the same one I have and it DOESNT allow slaved motors. This video is just showing it working as 3 + 1 which the controller will do no problem. The fact they have twin screws means they are using dodgy practise of using one output to control two drives or even worse one drive to control two motors.?

IT DOESNT SLAVE TRUST ME.!!!!

I'm using the 4-axis version with much newer firmware but it still doesn't support slaving, hence the creation of this board to add the capability without the operator or the controller needing to know it is there, it just works passively in the background when required. The 4th axis is obviously then still available to use as a 4th axis as well.

For me the most important function is (re)squaring the gantry before every job, I think you are more concerned with setting up the initial squareness without moving targets. To be clear none of the methods implemented in any of the controllers perform auto squaring they all require some square reference setup on the machine to search for, whether that reference is the actual switch actuation point or some other position offset from it. Just like none of them support auto tramming or auto planaring (new word).

[JAZZCNC]

I'm using the 4-axis version with much newer firmware but it still doesn't support slaving, hence the creation of this board to add the capability without the operator or the controller needing to know it is there, it just works passively in the background when required. The 4th axis is obviously then still available to use as a 4th axis as well.

So if you still have 4th axis spare then you must be controlling 2 drives with 1 output.?

[devmonkey]

So if you still have 4th axis spare then you must be controlling 2 drives with 1 output.?


Yes, that is the whole point of this board I made, it enables the use of a single axis controller output and single home input to safely drive and also correctly home/resquare a dual motor axis with dual home switches. The controller executes its normal procedures without any knowledge that there are two motors and two switches.

[Kitwn]

Ok your Stupid.:hysterical: . . .But so am I because I'm not seeing it either.?


Actually that should be "Ok you're Stupid' but the grammar police will let you off just this once:excitement:

What confused me was the homing being set by the release of the switch. I only speak LinuxCNC which sets the homing on the (second) activation of the switch after the gantry has backed off and run in again slowly. An upgrade to version 2.8 includes the ability to have two separately controlled motors with their own switches to square the gantry. Then it runs off the switches by a set distance which can be slightly different for the two sides to fine-tune the squaring.


Yes, that is the whole point of this board I made, it enables the use of a single axis controller output and single home input to safely drive and also correctly home/resquare a dual motor axis with dual home switches. The controller executes its normal procedures without any knowledge that there are two motors and two switches.

That's basically what my board was designed to do but I opted to create all-new step pulses while the controller believed the machine was stationary. It also acted as the splitter for sending the pulses from one output axis of LinuxCNC to two motor drivers during normal operation. 'Great minds', eh?

[devmonkey]

That's basically what my board was designed to do but I opted to create all-new step pulses while the controller believed the machine was stationary. It also acted as the splitter for sending the pulses from one output axis of LinuxCNC to two motor drivers during normal operation. 'Great minds', eh?

Great minds! I just wanted the controller to stay in control at all times so I figured this approach was simpler and possibly a bit safer and it squares the axis as part of the controllers normal homing process. Not sure on the pros/cons of detection on approach vs retraction so long as it is done super slow I don't think it matters, may make more of a difference with mechanical switches.

I did notice a problem with my board yesterday, it was generating a spurious step every few seconds and you could hear them as faint ticks on the stepper whilst it was running. This was due to common mode noise on the step differential line, and is because I didn't bother to put a differential line receiver on the input lines, rather both the +ve and -ve signal are referenced to GND on the arduino. It was fixed by debouncing the inputs over 1 cpu cycle 62.5ns, i.e. reading them twice. However I think I will make a new board in a few days that properly terminates the differential signals, works well enough as it is to to get the machine working now. Any thoughts on the pros/cons of using a opto to terminate the differential signal like the stepper drivers do vs a proper line receiver?

The next step is work out how to bury these proximity sensors in the machine, the Z will be a bit tight and I may need to source a much smaller switch.

[Kitwn]

Not sure on the debouncing bit. Sounds like you have short duration spikes of noise but would need an o'scope to check. Some sort of filtering, ferrite beads or capacitors for example might sort it out but I haven't the experience with line receivers to advise you.

We've discussed opto-couplers at length in the past (wasn't Doddy involved?) and I have been experimenting with a Long Bendy Optical Coupler (LBOC) recently but that will be for another thread.

Kit

[Doddy]

Someone mention Optos? Great for killing line noise, but in this case my usual mantra of propagation delay through the Opto applies (but only because I've been bitten by Optos in the past). Arguable you should use a reverse-connected diode to avoid reverse-biasing the LED in the opto.

If you have noise on the line you really need to deal with it electrically, rather than through software, particularly with a finely crafted LUT solution as you have (I'm not taking the piss - it's a good solution here, mimicking one solution used by FPGAs for logic representation.... I'll be honest, when you first mentioned this project I was thinking it'd be a brilliant first-project for anyone looking to get into FPGAs, a Cyclone IV board will go cheap at around £20... but that sends the chat into a completely different direction and isn't really for this site) - in that case you could have easily configure for differential inputs. Anyway, as-it-is, I suppose my own view is clear from driving the spindle-encoder on the lathe - I use the MAX487 chips, on tuppeth-ha'peny Arduino RS485 interface boards to provide 120R differential drive/sinks as my weapon of choice.

[devmonkey]

Hi Doddy,

Yeh an FPGA would be an excellent choice although I don't have any non-BGA parts here and didn't want send out for 4 layer board to be made. Lattice have some TQFP parts that could be used but I haven't got any. I have some 485 line drivers though left over from running DMX lighting throughout my house, thanks for reminding me!

On the noise issue, it is being caused by the 328p sampling the differential pair wrt ground without removing the common mode noise (which will always be there). If I'm right it would be completely eliminated by properly terminating the pair with a line receiver.

The LUT does work nicely, I use an 8 bit address space at the moment, 6 input bits and 2 bits of state. I moved all my code into a loop that executes at startup, this iterates through the address space and writes the 8 bit output [2 bits of updated state as the FSM transitions, 6 bits of compressed output] into a 256byte block of memory aligned on 256byte memory boundaries. This removes the need to hand craft a LUT, all you are really doing is executing your logic for every possible input and caching the result.

Then the actual runtime programme loop written in assembler reads the input port directly into the lower byte of the indirect addressing register pair, debounces, merges in the 2 bits of state, immediately does the lookup, this is made very easy by keeping the LUT is aligned with a 256 byte memory boundary.

The assembler the decompresses the output bits, writes them out and updates the state bits. Repeat ad-nauseum.

[devmonkey]

All home switch fixings are now done, it is amazing how long the little jobs take, this was a whole afternoon. Milled up a little block to hold the Z switch, decided it would be better outside of the axis rather than inside regards adjustment. Then had to drill two very awkward 12mm holes in the X bearing plates with a portable drill, for this I had to mill up a drill guide, still easier than dismantling the gantry.

28370 28371 28372 28373 28374

[devmonkey]

Last of the tedious mechanical bits I think, energy chain carriers.

28375 28376 28377 28378

Next job soldering cables, lots of it, and bolting the machine stand to the floor.

[devmonkey]

Question for Doddy,

I've scoped the common mode noise and it isn't sufficient to cause a phantom step to be read at the 328p. I think the phantom steps I was detecting that I have 'fixed' by debouncing the read are actually being caused by reflections and/or induced voltage spikes as the differential pair switches. If this is correct placing a termination resistor across the pair as you do for rs485 should help do you concur?

Cheers, Joe

[devmonkey]

How do people usually wire the alarm outputs of the drivers?

My plan was to place these in series with the estop but this wont actually work for me. Reason being estop unlatches the relay that switches the main contactor that switches AC into the driver PSU, this means that the drivers are unpowered until reset is pressed latching the relay, however the relay wont latch as the drivers are unpowered therefore cannot switch on their open collector alarm outputs breaking the estop circuit.

The only thing I can think of is to add another relay in series with estop, via the normally closed contact and then have the driver alarms open this relay, this has the downside that you don't know if the alarm circuit is intact until it doesn't work.

[Doddy]

Question for Doddy,

I've scoped the common mode noise and it isn't sufficient to cause a phantom step to be read at the 328p. I think the phantom steps I was detecting that I have 'fixed' by debouncing the read are actually being caused by reflections and/or induced voltage spikes as the differential pair switches. If this is correct placing a termination resistor across the pair as you do for rs485 should help do you concur?

Cheers, Joe

I guess you know a bit about signal theory - you're trying to match the characteristic impedance of the transmission line at both ends - just randomly throwing resistors on won't be as effective as a properly matched terminator - and yes, as per 485. What's the driving system?, have a look at that and match it.

[devmonkey]

I guess you know a bit about signal theory - you're trying to match the characteristic impedance of the transmission line at both ends - just randomly throwing resistors on won't be as effective as a properly matched terminator - and yes, as per 485. What's the driving system?, have a look at that and match it.

Just taken the DDCS apart again, the differential outputs are all driven by a standard quad differential line drivers, AM26LV31, no resistors at the transmitting end. I'm driving this through a 2m db37 cable, not twisted pair, or if it is it is highly unlikely the right pairs are twisted. Still these are EIA-422 spec drivers same as max485, etc, so I reckon I wont go wrong with 120 ohms.

It DDCS has a really nice board, has an onboard FPGA, Altera cyclone3 that does all the realtime stuff. I've taken a few photos whilst it is open.

[Doddy]

Section 9.1 of the TI data sheet - recommends 100R, you're not far off. It's a fast switching device - most differential drivers (not design for high speed) will have deliberately slow slew-rates to limit EMI. But, it is what it is.

[devmonkey]

Progress update:

Stand is bolted to the floor and machine is clamped to the stand. I cut some little bits of angle and use M8 bolts for the clamps. So first the stand was levelled as best a possible with some composite slate tile shims and bolted to the floor. Then the machine was levelled on top of the stand (it has an M12 bolt threaded into the bottom of each corner) and checked that each bolt was bearing weight (attempting to avoid twisting the machine), then finally it was lightly clamped at 4 points to the stand.

28401 28402

I cut all the cabling to length and took it indoors to solder all sensors and motors, this took a few hours and rather a lot of heatshrink.

Next I fitted all the motors and sensors to the machine and started work on the enclosure.

A few trips to screwfix to further increase my holesaw collection, then an unpleasant afternoon drilling fan, exhaust, switch, indicators and gland ports in the steel enclosure. Now mounted to the wall and cabling fed through. Will do final wiring tomorrow and maybe the first machine moves!

28403 28404 28405

That ugly slot next to the glands is for the DB37 control cable to go to the DDCS control box.

Finally both drag chains were fixed.

28406 28407

[routercnc]

That machine is really starting to take shape. Nice work.

[Neale]

How do people usually wire the alarm outputs of the drivers?

My plan was to place these in series with the estop but this wont actually work for me. Reason being estop unlatches the relay that switches the main contactor that switches AC into the driver PSU, this means that the drivers are unpowered until reset is pressed latching the relay, however the relay wont latch as the drivers are unpowered therefore cannot switch on their open collector alarm outputs breaking the estop circuit.

The only thing I can think of is to add another relay in series with estop, via the normally closed contact and then have the driver alarms open this relay, this has the downside that you don't know if the alarm circuit is intact until it doesn't work.

I have wired fault outputs in series and taken that to a "servo fault" input on the motion controller. Partly for the reason you give - can't put power on the drivers via the safety relay until there is power on the drivers - and partly because I classify "fault" signals into two categories. Personal safety - when you hit estop you really want it to stop NOW so that is a safety relay task (mine cuts driver power, driver enable, and signals motion controller), and machine issue - driver fault, limit switch - and I'm happy that motion controller firmware can be trusted for that. Does DDSC have "driver/servo fault" input(s)?

[AndyUK]

My solution is similar to Neale's. The UB1 board I use has a simple safety-relay circuit, so the first 10 or so inputs can be linked to close a relay. My driver's fault signals feed into this, alongside my VFD's fault signal and a signal from my e-stop circuit. The driver's enable line is linked to the relay output, so if one drive faults the drives remain powered but not enabled. Same happens if the VFD drops out, or I hit the e-Stop button (but that also has the effect of killing the power to the drivers directly).

[devmonkey]

Thanks guys, lots of inspiration here.

The DDCS has two estop inputs but one is difficult to access on the MPG port. I have an idea how to sort this now. Currently I use the output of the safety relay to switch the low side of the contactor coil and also signal low to the DDCS estop input which is configured to go into emergency stop if this signal is high. I was looking to integrate the alarms on the estop switch side of the safety relay but this is wrong, instead I can put them on the output side. Take a resistor to +24v then series the NC alarm outputs via the same safety relay contact to ground. Take the DDCS estop input from the resistor.

With this configuration the alarms will only estop the DDCS not the contactor, but when the safety relay is tripped both DDCS and contactor will estop. Million dollar question, do I need a diode to prevent current flowing from emitter to collector through the alarm NPNs into the DDCS input when the safety relay is open? My analogue electronics is extremely rusty.


28416

[Kitwn]

Analogue is easy, just work out where the current will flow. Ohm's Law plus Superposition Theorem equals The Answer.

[devmonkey]

Good news machine is moving, very nicely indeed, I will post some video tomorrow.

I have a question about a potential issue I've discovered with the 24v dc smps I'm using for powering the controller and proximity switches. The unit is a high quality double insulated (no earth connection) TDK-lamda din rail SMPS and I've used them in automation projects before. However this is the first time I've noticed this issue.

Basically you get a slight tingle (barely noticeable electric shock) when touching both DC ground and mains earth.

I believe this is down to the 'Y' capacitor used in the SMPS to reduce EMI, it is a small capacity capacitor that couples input to output (when there is no earth connection) and is required by all SMPS to meet EMI regulations, however it has the nasty side effect of raising the DC output to around half mains voltage wrt mains earth. The capacitance is tiny so it isn't dangerous.

I noticed it whilst leaning on the earthed machine and holding the shield of my db37 cable which is connected to dc ground.

I've checked the unit with a megger to prove to myself it is properly isolated.

So do other people see this, does it matter? What do other people use for the 24v side of their systems? I might substitute it for a small power supply with an earth connection (class1??).

EDIT:
Just ordered a class1 meanwell unit MDR-60-24, this has an earth connection and the datasheet shows it's 'Y' capacitor is connected to it.
https://www.meanwell.co.uk/pub/media/productPDF/MDR-60-SPEC.PDF

Cheers, Joe

[Neale]

Good news machine is moving, very nicely indeed, I will post some video tomorrow.

I have a question about a potential issue I've discovered with the 24v dc smps I'm using for powering the controller and proximity switches. The unit is a high quality double insulated (no earth connection) TDK-lamda din rail SMPS and I've used them in automation projects before. However this is the first time I've noticed this issue.

Basically you get a slight tingle (barely noticeable electric shock) when touching both DC ground and mains earth.

I believe this is down to the 'Y' capacitor used in the SMPS to reduce EMI, it is a small capacity capacitor that couples input to output (when there is no earth connection) and is required by all SMPS to meet EMI regulations, however it has the nasty side effect of raising the DC output to around half mains voltage wrt mains earth. The capacitance is tiny so it isn't dangerous.

I noticed it whilst leaning on the earthed machine and holding the shield of my db37 cable which is connected to dc ground.

I've checked the unit with a megger to prove to myself it is properly isolated.

So do other people see this, does it matter? What do other people use for the 24v side of their systems? I might substitute it for a small power supply with an earth connection (class1??).

EDIT:
Just ordered a class1 meanwell unit MDR-60-24, this has an earth connection and the datasheet shows it's 'Y' capacitor is connected to it.
https://www.meanwell.co.uk/pub/media/productPDF/MDR-60-SPEC.PDF

Cheers, Joe

Why leave the 24V supply floating wrt earth? Take the 0V connection to the control box star point and all will be well. Can't see any reason not to ground it. That's what would normally be done - my control box has a 24V smps and a 5V+12V smps and all the 0V connections go straight to the star earth.

[devmonkey]

Should add the noise is coming from the stepper drivers as it only happens when the stepper is moving, so basically the EMI filter in the 24v smps is filtering what comes into it and coupling it to the 24v DC output which causes the DC output to float relative to mains earth because the neutral and earth are coupled at the electricity company's box in the street. I will fit an EMI filter in front of the toroidal stepper PSU to suppress noise returning back up the supply and into the 24v SMPS (and the rest of the house).

[devmonkey]

Why leave the 24V supply floating wrt earth? Take the 0V connection to the control box star point and all will be well. Can't see any reason not to ground it. That's what would normally be done - my control box has a 24V smps and a 5V+12V smps and all the 0V connections go straight to the star earth.

Is that really recommended? Doesn't it defeat the purpose of the galvanic isolation?

[Neale]

Not quite sure what the issue is here. If the output of a psu - any psu - is left floating them you can guarantee what volts you will see between the terminals but volts wrt ground is undetermined. It will be somewhere between earth and full mains depending on leakage, transformer internal coupling, EMI filler leakage and so on. Take one terminal to ground and everything is now properly referenced to that - no more tingles!

I don't see that the EMI filter effectiveness or anything else will be affected. In fact, electrically, you are in a much better position because at the moment I would be worrying about the effect of stray voltages on sensitive components. If that "tingle" happens to occur around any high-impedance inputs...

The reason for leaving the outputs floating is that someone might want a -24V supply so in that case they would ground the +'ve output pin. There might be other reasons but that's one good one.

[devmonkey]

I was thinking that the galvanic isolation (floating DC) is there to prevent a current path between AC live and DC, and that when using a probe you temporarily defeat it. I think also galvanic isolation prevents ground loops should a secondary piece of equipment grounded equipment like a PC be connected, not a consideration here.

But as you say you might get a tingle.

[Neale]

There are a few reasons for the galvanic isolation. That's actually just a posh name for saying that there are no connections from the live/neutral terminals to the output terminals. I mentioned one reason for this earlier - so that the same smps could be used as a +24V source or a -24V source with one output terminal grounded. Another reason (and someone else mentioned this in a post a little while back) for industrial equipment is that if the 24V supply rails are floating with respect to ground, then you can put in fault detection mechanisms that check if either rail does get shorted to ground under fault conditions; the machine can then be stopped before any magic smoke is emitted. Or, maybe, you have a motor speed controller that needs a 24V supply, but is connected to the mains supply to the motor with no isolation. However, if neither smps output terminal is grounded, you really need to make sure that none of the control electronics wiring or terminals are touchable (due to the whole thing floating at some undetermined but possibly dangerous voltage with respect to ground).

On a practical note, this also means that a simple electrical touch-off system for tool height setting is not going to be easy as this depends on the touchplate being connected to one controller input and being shorted to ground by the tool tip via spindle/machine frame, etc. If the controller is not referenced to ground, then this ain't going to work, or you will need to isolate the spindle from the machine frame and provide an additional connection to it - which will all be floating at "it tingles a bit" voltages.

All these issues disappear if you just take the SMPS 0V connection to ground. I cannot see any downside to this.It is what just about any other home-built machine does.

External, e.g. PC, connections? Valid point, and might be an issue in an industrial environment where different bits of kit are connected to different supplies. In a domestic situation where all the equipment is probably connected to the same ring main, this is a non-issue. In addition, an RJ45 ethernet connection will be galvanically isolated and I believe that the USB connection spec calls for galvanic isolation as well.

[devmonkey]

There are a few reasons for the galvanic isolation. That's actually just a posh name for saying that there are no connections from the live/neutral terminals to the output terminals. I mentioned one reason for this earlier - so that the same smps could be used as a +24V source or a -24V source with one output terminal grounded. Another reason (and someone else mentioned this in a post a little while back) for industrial equipment is that if the 24V supply rails are floating with respect to ground, then you can put in fault detection mechanisms that check if either rail does get shorted to ground under fault conditions; the machine can then be stopped before any magic smoke is emitted. Or, maybe, you have a motor speed controller that needs a 24V supply, but is connected to the mains supply to the motor with no isolation. However, if neither smps output terminal is grounded, you really need to make sure that none of the control electronics wiring or terminals are touchable (due to the whole thing floating at some undetermined but possibly dangerous voltage with respect to ground).

On a practical note, this also means that a simple electrical touch-off system for tool height setting is not going to be easy as this depends on the touchplate being connected to one controller input and being shorted to ground by the tool tip via spindle/machine frame, etc. If the controller is not referenced to ground, then this ain't going to work, or you will need to isolate the spindle from the machine frame and provide an additional connection to it - which will all be floating at "it tingles a bit" voltages.

All these issues disappear if you just take the SMPS 0V connection to ground. I cannot see any downside to this.It is what just about any other home-built machine does.

External, e.g. PC, connections? Valid point, and might be an issue in an industrial environment where different bits of kit are connected to different supplies. In a domestic situation where all the equipment is probably connected to the same ring main, this is a non-issue. In addition, an RJ45 ethernet connection will be galvanically isolated and I believe that the USB connection spec calls for galvanic isolation as well.

I get your point on the convenience of probing Neale, however if you don't connect DC ground to AC ground and you are using an SMPS there is no current path to AC ground at all, there is no shock risk. You get the tingle from the Y capacitor in the SMPS EMI filter there are usually two, one coupling AC live to DC+ and the other AC neutral to DC-, the capacitors are not allowed to be big enough to carry enough charge to hurt you and they are required to fail to open circuit. If you take any 5v USB charger and hold the shield whilst touching AC ground you will get the same tingle.

I'm not against connecting AC ground with DC ground but it does come with the risk that an AC live fault in the machine enclosure will fry all your electronics as there is a current path between AC live and DC ground. My last machine didn't have this connection so when probing a temporary DC ground connection was placed on the spindle then removed afterwards. I'm not sure which is more correct.

[devmonkey]

Progress update. Today was spent putting some controls in a box and attaching it to the machine. I used an old TV mount and made some crude alterations to it with the welder.

28423 28424 28425 28426

I also soldering up the MPG connections which was a massive pain in the rear.

First moves:

https://youtu.be/u7VPziPBhRg

[Neale]

Great when it reaches that point - gives you the drive to get it finished!

Seems a little slow when jogging with the MPG - is that a "settings" thing? Mine winds back and forth a bit faster, but there's also a "step size" switch on the MPG that effectively changes jogging speed.

[devmonkey]

Yep the DDCS lets you configure the step size per MPG tick for each of the three speeds on the MPG but I haven't configured it yet, so it is defaulted to 4 microns or something ridiculous. Sounds rough on the lowest setting as it is below the precision of the machine.

Still got to rig up the home switches and solder all the cable shields on, then I can tune the config. Also need to dig out my old VFD and see it it still works, it may have been exposed to the elements.

Then cut the plate for the bed and get the machine to drill itself....

Before going much further I will run some complex toolpaths in the air so see if the DDCS is any good, I have no reason to believe that it isn't but want to know sooner rather than later if I need to switch to a PC, hopefully it will be fine.

[devmonkey]

I ended up fitting the tiny hiwin grease nipples to the HG15 carriages, these are a pain. Today I knocked up an adaptor from a piece of aluminium, it works but only about 50%.

28429 28430

I've read the hiwin instructions for moving the grease point to the side of the carriage by piercing one of the blind holes in the green plastic molding, has anyone tried this? It would greatly simplify greasing my Z axis.

[devmonkey]

VFD is working, quite surprised by this as it has been 'outside' for a couple of years. DDCS is controlling the spindle on/off and speed.

Today I cut the 20mm plate for the bed. The blank weighed 80kg which is too much to safely manoeuvre around the table saw so I just my very old and rough circular saw, lots of fluid, lots of smoke, but it did a pretty good job.
28447 28448

I took the edge off with a chamfer bit in a wood router, the bed sitting in place:
28449

I took the a rough height map with a DTI clamped to the Z. As expected from bringing the rails into plane with laser and epoxy I have around 1-1.5mm total error (this was the frame error before pouring epoxy). The cross members of the frame are also not perfectly in plane and would require a little shimming under the bed.

I'd rather have the bed closer to planar that this, although the eventual spoil board skimming will fix it, so I'm contemplating attaching some shimming material to the cross members and surfacing this then bolting the bed plate down on top.

Another thought is to float the bed plate on some rubber washers and compress them until the DTI reads zero everywhere.

[Nealieboyee]

I would either bolt down some thinnish alu plate on each member and then skim them all level, or pour more epoxy on the members so they self level. But that means making a framework for the epoxy to travel between members.

[Kitwn]

Epoxy sounds a bit too complicated for something so simple and you'd have to confirm the frame was adjusted to make the plane of the rails orthogonal to local gravity or it becomes pointless.
I'd go with the shim and skim. That cannot fail to put the plane of the base parallel in all directions to the plane which the spindle moves which is what you really want and will minimise the thickness of material you need to skim off the final, sacrificial surface.

[Clive S]

I took the a rough height map with a DTI clamped to the Z. As expected from bringing the rails into plane with laser and epoxy I have around 1-1.5mm total error (this was the frame error before pouring epoxy). The cross members of the frame are also not perfectly in plane and would require a little shimming under the bed.

I'd rather have the bed closer to planar that this, although the eventual spoil board skimming will fix it, so I'm contemplating attaching some shimming material to the cross members and surfacing this then bolting the bed plate down on top.

Out of interest did you put a moat across the two rails when you poured it?

[JAZZCNC]

I would use epoxy metal paste. Use levelling grub screws in the bed and adjust until planer, then remove and skim all the surfaces with a lite coating of epoxy putty then put plate back on and tweak it out any error from lifting etc.
Tape the bed where it meets the epoxy and when dry you will be able to remove bed.

[JAZZCNC]

I'd go with the shim and skim. That cannot fail to put the plane of the base parallel in all directions to the plane which the spindle moves which is what you really want and will minimise the thickness of material you need to skim off the final, sacrificial surface.

It's not possible to skim the whole surface using the machine because plate is the full size of the base so cannot skim the portion under the gantry.!! Only way this could work is if the plate was the actual cutting size.

[Kitwn]

It's not possible to skim the whole surface using the machine because plate is the full size of the base so cannot skim the portion under the gantry.!! Only way this could work is if the plate was the actual cutting size.

Obvious when you point it out!:stupid:

I like the grub screws and epoxy putty idea, but might it be better to use ordinary bolts instead? It would mean a slightly thicker layer of putty but it would allow you to adjust each bolt with an open spanner while the gauge is on it's head rather than have to measure, move the gantry away to adjust, move it back to measure, move it away to adjust again etc.

[JAZZCNC]

I like the grub screws and epoxy putty idea, but might it be better to use ordinary bolts instead? It would mean a slightly thicker layer of putty but it would allow you to adjust each bolt with an open spanner while the gauge is on it's head rather than have to measure, move the gantry away to adjust, move it back to measure, move it away to adjust again etc.

No point really and you want to be measuring off the bed not the bolt head. A M3 or M4 socket cap grub screw is easily adjusted thru the hole and is small enough not affect the bed too much.

There are several ways to do the same thing and it depends on how your planning on using the bed regards fixtures etc. The main point was Epoxy putty works good for this application, it could easily be car body filler(Bondo to you upside downers) if your on a budget...Lol

[Kitwn]

I was thinking the grub screws would be in the frame, not holes drilled in the base board. Slowly but surely the lights come on.

[devmonkey]

Thanks for the ideas guys. I went ahead and epoxied some 2" x 1/8th aluminium strip to the 3 cross members that are machinable from the gantry. There is around 60mm outside each of these which will be shimmed manually. The rear cross member cannot be machined as it is under/behind the gantry as Jazz says, so this will get manually shimmed as well. Quite easy to do with a DTI n the Z as 90% of the bed plate is supported by the milled strips,

28459

So the exciting bit was getting the machine to machine itself, first time it has cut anything:

https://youtu.be/mnwhJR-eVZg

Finish was superb for chewing gum grade aluminium, thickness of thinnest final shim is ~1mm.

28460 28461 28462

I used an 8mm 2 flute carbide bit, there is no scalloping which means the head is in pretty good tram (I can't tram it until the bed is on). I ran the tool path twice, initially to bring all 3 strips down to height then dropped the Z 0.1mm for a finish pass.

[devmonkey]

Out of interest did you put a moat across the two rails when you poured it?

Hi Clive, yes there was a moat, I put it outside of the frame at one end. The rails are planar within 10um, I checked them with my laser software. Or at least they were when I fitted them, they have probably drifted a bit as I've hoisted the machine around the shop a couple of times since then. I will recheck them at some point.

I didn't post anything about how I did the pour but there are two important points I'd like to pass on to get the best result,

1. As per Boyan's findings the moat should be outside the frame, close to the same cross section as the rail support epoxy and the pour should extend beyond both ends.
2. You must use non-absorbent material for the moat and dams, anything else and non-linear capillary action into the material will mess things up. I used 10mm pvc angle for the dams and pvc trunking for the moat. The dams were stuck to the frame with thin toffee tape which worked well (didn't leak) but was a pain to get off.

28463

[devmonkey]

Time to fix the bed down. I want to finalise the work holding at the same time. I have a few choices,

1. Drill and tap a grid of holes in the plate and bring these holes up through the spoil board,
2. Just drill and tap the spoil board (pvc foam) creating blind holes.
3. Cut t-slots into the plate and create narrow planks of spoil board between them (I would need to get a special bit to cut the t-slot),
4. Create t-slots from planks of spoil board, not sure how strong they will be.

T-slots are probably more work but less tapping, what do people think?

[JAZZCNC]

I'd go with a mixture of T-Slot Track in the Spoil board around the outer edges and a matrix of Tapped holes.
The T-slot allows quick and easy clamping of large pieces.
The Matrix of holes allows you to clamp small parts easily and just about any where on the bed and nearly always means you can find a place on your part to clamp.

The pain of tapping all those holes is worth the effort IME.!! . . .

[devmonkey]

Cool, any suggestion on the pitch of the matrix? I was thinking 150mm to limit the tapping...

Also would you tap the bed or the spoil board?

[JAZZCNC]

Cool, any suggestion on the pitch of the matrix? I was thinking 150mm to limit the tapping...

Also would you tap the bed or the spoil board?

No 150mm is way too big, 75mm would be max I'd go with, 50mm better. Don't be put off by the tapping, the thought of it is worse than the doing. Plus you don't have to tap every hole straight away, could even just tap every other hole and use the others for dowel pins for fixture guides etc. This is where a matrix really comes into it's own because it's very flexible.

Tap the Bed not the spoil board. You'll be cutting into the spoil board and slowly surfacing it away.

[Kitwn]

I have M8 tee nuts in the underside of my plywood baseboard with a drilled MDF spoilboard on top but I doubt you will allow anything so porous or flexible onto your machine. My spacing is 100mm and it's definitely too large.

One thing to consider is making the effort to get the holes on exact multiples of 50 or 75mm from your zero reference in both X and Y. You probably will drill them here anyway but this makes it easier to plan the layout of a job if you want to drill fixing holes in the workpiece, dedicated spoilboard or a jig. I say this because I made the mistake of drilling the holes, fitting the tee-nuts and then turning the board over to screw it down which means the grid is not as perfectly aligned as I would wish. We live and learn, it'll get fixed one day.

[Clive S]

1. As per Boyan's findings the moat should be outside the frame, close to the same cross section as the rail support epoxy and the pour should extend beyond both ends.

I actually don't agree with the above. I used two moats about a third from each end. But what is a must, the epoxy must be wide enough to all for the mucus (or whatever you call it) on each side and use the very slow epoxy.

[devmonkey]

I have M8 tee nuts in the underside of my plywood baseboard with a drilled MDF spoilboard on top but I doubt you will allow anything so porous or flexible onto your machine. My spacing is 100mm and it's definitely too large.

One thing to consider is making the effort to get the holes on exact multiples of 50 or 75mm from your zero reference in both X and Y. You probably will drill them here anyway but this makes it easier to plan the layout of a job if you want to drill fixing holes in the workpiece, dedicated spoilboard or a jig. I say this because I made the mistake of drilling the holes, fitting the tee-nuts and then turning the board over to screw it down which means the grid is not as perfectly aligned as I would wish. We live and learn, it'll get fixed one day.

I was going to drill the fixture holes before finally fitting the bed, but I agree it is better to do it afterwards once the machine is properly aligned. I haven't squared the gantry yet, nor properly defined zero. Will get the bed fixed to the frame in its final location, clamp on a sheet of MDF on top to square the gantry, define zero then finally drill the fixture holes.

I'm also toying with the idea of redefining my long axis as Y and the gantry axis as X so it lines up with the natural way of looking at it from the front and the axis orientations match the DDCS buttons, otherwise I'm going to make a mistake. I would define zero as front left.

I have drilled and tapped the bed and frame, and bolted to the locations where I had fixed and milled the aluminium strip. Reads within +/- 0.01mm everywhere in this region and there is a 0.02mm droop towards the as yet unsupported rear end. I think rather than try and shim the unsupported bits with shim material I'm just going to squirt some structural 2 component polyester resin in there, it is close enough and I will just make it worse trying to shim normally.

28464 28465

[devmonkey]

I actually don't agree with the above. I used two moats about a third from each end. But what is a must, the epoxy must be wide enough to all for the mucus (or whatever you call it) on each side and use the very slow epoxy.

I guess whatever works for you.

I did some tests under the microscope before I did the actual pour, the epoxy shrinkage phase occurs after it has setup too much to freely flow and find level, this (as Boyan determined) is what causes dips around dam ends and bridge joins. I used West's 105/209 and the shrinkage phase starts after about 1 hour by which time it is getting too viscous to re-level itself. The way I saw it is the bridge is only effective before the shrinkage starts and after it starts you want the area x-section of the epoxy you want to keep to be as consistent as possible meaning you have to extend it past the ends of the machine and you want the bridge to be connected to a part you don't want to keep as it either it will pull from the sides or the sides will pull from it. Either way it doesn't matter as you are going to cut this bit off.

The microscope method was stolen from the "Dam Busters" use of angled lights on the wing tips to determine they were the correct height above the water to drop their bombs. You point a laser pointer at an angle to the surface, focus the microscope directly down onto the point where the laser hits the surface and as the surface height changes the dot will move, magic ;-)

[JAZZCNC]

I have drilled and tapped the bed and frame, and bolted to the locations where I had fixed and milled the aluminium strip. Reads within +/- 0.01mm everywhere in this region and there is a 0.02mm droop towards the as yet unsupported rear end. I think rather than try and shim the unsupported bits with shim material I'm just going to squirt some structural 2 component polyester resin in there, it is close enough and I will just make it worse trying to shim normally.

In the picture you still have the protective coating on the plate so you can't trust those measurements if you did it with that still on.?

Regards the Epoxy and bridge etc then I've probably done more pours than anyone here, I've lost count of how many but it ranges from Medium and V-large, my experience is between both you and Clive. I agree the moats are better outside the frame but you need them both ends of the machine. Also, make the moats lower than rail surface as this seems to help the flow. However the most important I find is the temperature, keep the temp on the low side of what's recommended as it slows the curing process and gives it more time to level. The only tricky bit is it's a fine line between just right and too cold and it not setting correctly.

[devmonkey]

In the picture you still have the protective coating on the plate so you can't trust those measurements if you did it with that still on.?


The covering is tight and uniform so I think I can trust them enough, I DTI'd the milled shims first which was zero everywhere, happy that my +/- 0.01mm on the top is a mixture of error from the covering and the plate itself. Anyway doesn't need to be perfect as the spoil board will take up any error, if I need a very flat part for bolting a vice to I will skim the aluminium bed, I left the bolt heads low enough to do this.

[devmonkey]

Bed is now fully fixed. I reckon it is within 0.02mm total error, not that I really trust this digital DTI.

The rear cross member and the outer edges of the other cross members that did not have the milled shim were done as follows. I laid some narrow strips of closed cell self-adhesive glazing rubber either side of each bolt hole. Then I re-placed the bed and tightened the central set of bolts down hard onto their milled shims. I then indicated the entire bed adjusting the outer bolts to zero, the glazing rubber provided a bit of resistance to bias the bed plate high so the bolts had something to tension against. Finally I got under the machine and injected some construction resin into the gap, this set hard in a few minutes as it is so hot today, it is basically bondo in a cartridge.

The bed plate had a very thin film of WD40 on it to act as a release so hopefully it will be possible to get it off again should I ever need to., not sure the resin sticks well to the shiny aluminium plate anyway.

28469 28470 28471

This is the very last time that bed will look so nice!

[devmonkey]

EM806 question.

All the documents state that the alarm output is 'active high' by default and that 'active high' means

"Active High
means high output impedance for drive error and Active Low
means low output impedance for driver error. "

When I connect to the drive over RS232 it indeed states it that the alarm is active high. However it is behaving opposite to the manuals, rather it is high impedance normally and low impedance in error, I verified this by stalling a motor. I have now configured it to active low which allowed me to daisy chain all the alarms together - they are conducting to ground until a drive goes into error or the fault wiring is broken.

So my question is why is the behaviour opposite to the manuals and should I be concerned that things like the step edge setting may also be reversed? Is the manual just wrong and 'active high' means their logic output is high in error driving the base current on the NPN which then conducts, i.e. has low impedance?

[JAZZCNC]

I've had this many times with the AM882 and the EM806 and it's a random thing, it's like they must put something in backward...Lol

Regards the Step edge then I'm not so sure because while I've had many drives be on the wrong side of the edge I never actually checked which was at fault, the controller, or the drive. I just do a quick back n forth check with g-code and if it drops steps then flip it over. I do this on every machine regardless if the drive and controller say they match or not.!

[devmonkey]

I've had this many times with the AM882 and the EM806 and it's a random thing, it's like they must put something in backward...Lol

Regards the Step edge then I'm not so sure because while I've had many drives be on the wrong side of the edge I never actually checked which was at fault, the controller, or the drive. I just do a quick back n forth check with g-code and if it drops steps then flip it over. I do this on every machine regardless if the drive and controller say they match or not.!

Thanks for clarifying Jazz. All 4 drives I have here behave the same way, opposite to the manual. Possibly there are multiple versions of the firmware floating around.

Will do the test for step edge, does it lose a step everytime the axis reverses, i.e. I should be able to test it as creep on a DTI at either end of the stroke?

[JAZZCNC]

Thanks for clarifying Jazz. All 4 drives I have here behave the same way, opposite to the manual. Possibly there are multiple versions of the firmware floating around.

Will do the test for step edge, does it lose a step everytime the axis reverses, i.e. I should be able to test it as creep on a DTI at either end of the stroke?

Yes it drops a step every direction change. I just zero the axis put DTI on it and do 200-300 small G0 moves, 5mm or so, with the last move back to zero.

[devmonkey]

I squared the gantry today. I used the method of equalising diagonals. First I had to make a measuring stick:

28479 28480

Then I had the machine drill 4 holes in a rectangle and inserted dowel pins, zero'd the DTI on one diagonal then measure the error on the other. The arduino board I made for axis squaring only had 8 bits of step offset. This machine uses 1610 screws and 16x microstepping so 8 bits represents 255*10/(200*16) = 0.8mm of software correction to the limit switch. Therefore I had to get it close first by moving the proximity sensor targets first, I got the diagonal error down to 0.32mm.

Fine tuning squareness was then done by changing the offset in the arduino code. After lots of messing around trying to calculate the gantry error angle then the actual number of steps to offset one switch I gave up. I just drilled holes with offset at zero, then with offset at 255, then interpolated the two errors to get the 'correct' offset (note this is not the correct way to do this but it works well enough over these very short distances). Plugged this into the software and low and behold:
28481

0.01mm error over ~1000mm diagonals. Square enough for now, it is repeatable. Can't do better than that without a much better DTI and some measurement of backlash.

Annoying sources of error whilst doing this were 1) a dowel pin that got scratched up pulling it out with pliers, 2) not having the DDCS home twice, this made it a little more consistent, don't know why.

I also noticed that the EM806 stall detection is not working for a stalled start, it works fine if you stall a spinning motor. Is this something I can tune or is it just a limitation of sensorless stall detection?

[Neale]

I squared the gantry today. I used the method of equalising diagonals. First I had to make a measuring stick:

28479 28480

Then I had the machine drill 4 holes in a rectangle and inserted dowel pins, zero'd the DTI on one diagonal then measure the error on the other. The arduino board I made for axis squaring only had 8 bits of step offset. This machine uses 1610 screws and 16x microstepping so 8 bits represents 255*10/(200*16) = 0.8mm of software correction to the limit switch. Therefore I had to get it close first by moving the proximity sensor targets first, I got the diagonal error down to 0.32mm.

Fine tuning squareness was then done by changing the offset in the arduino code. After lots of messing around trying to calculate the gantry error angle then the actual number of steps to offset one switch I gave up. I just drilled holes with offset at zero, then with offset at 255, then interpolated the two errors to get the 'correct' offset (note this is not the correct way to do this but it works well enough over these very short distances). Plugged this into the software and low and behold:
28481

0.01mm error over ~1000mm diagonals. Square enough for now, it is repeatable. Can't do better than that without a much better DTI and some measurement of backlash.

Annoying sources of error whilst doing this were 1) a dowel pin that got scratched up pulling it out with pliers, 2) not having the DDCS home twice, this made it a little more consistent, don't know why.

I also noticed that the EM806 stall detection is not working for a stalled start, it works fine if you stall a spinning motor. Is this something I can tune or is it just a limitation of sensorless stall detection?

I think that stall detection only works if the motor speed is greater than 300 rpm - so won't detect a stationary, stalled, motor. Unfortunately...

[JAZZCNC]

I think that stall detection only works if the motor speed is greater than 300 rpm - so won't detect a stationary, stalled, motor. Unfortunately...

Yep Stall detect only works above 300Rpm and yes it's limited in its accuracy and reliability.

[devmonkey]

Yep Stall detect only works above 300Rpm and yes it's limited in its accuracy and reliability.

Fair enough, it was a rather artificial scenario caused by me not plugging in the signals to one of the motors on the dual axis whilst still powering it (so the driver had it locked), the other motor stalled obviously.

[Kitwn]

0.01mm error over ~1000mm diagonals. Square enough for now, it is repeatable.

Nice job Joe. I think most of your readers would consider that 'square enough for now' :thumsup:

[devmonkey]

Just in the process of rigging up some dust extraction which is going well, but I also need air assist for aluminium chip removal. I have a very loud compressor I don't want to use, can anyone point me towards something quiet that has the required flow rate and pressure (apparently some people use pond air pumps??) for blasting chips out of deep pockets?

I've also seen these small plastic fans you can slide onto the cutter but I can't imagine they can blow chips out of deep narrow pockets.

[Voicecoil]

I found a little side channel blower in fleaBay for about £60 - with a little nozzle it blew chips everywhere, so much so that I had to throttle it back a bit. If you're tempted to go the same route just be aware that those things come in different flavours of pressure and volume delivery, the one I found did quite a reasonable pressure, ~280mBar IIRC.
I find the noise of the air jet far louder than that of the blower itself.

[devmonkey]

I found a little side channel blower in fleaBay for about £60 - with a little nozzle it blew chips everywhere, so much so that I had to throttle it back a bit. If you're tempted to go the same route just be aware that those things come in different flavours of pressure and volume delivery, the one I found did quite a reasonable pressure, ~280mBar IIRC.
I find the noise of the air jet far louder than that of the blower itself.

Thanks for the tip I will keep a look out for one, what power rating is yours? Could probably machine up a really small one on a brushless RC motor and mount it on the Z. Another project...

[Voicecoil]

Thanks for the tip I will keep a look out for one, what power rating is yours? Could probably machine up a really small one on a brushless RC motor and mount it on the Z. Another project...

Mine's about 500W @ 50Hz mains, will do a bit more at 60Hz.

[devmonkey]

Machine is going really well, I've made a few parts for the local flying club.

I'm now in need of a rotary axis, I have a couple of options:
1. Use the headstock and tail from an old rusty taig lathe I have with a belt drive off of a stepper or servo,
2. Convert a rotary table,
3. Buy one.

I need it quite quickly, can anyone recommend anything that is known to work, even if it is one of those chinese harmonic drives? I need something with a 100mm chuck and minimal backlash as it will be used for continuous processing not stop/lock.

I was looking at this, seems extremely cheap for a harmonic drive but they have them in europe so I could get it fast:
https://www.aliexpress.com/item/33025578546.html

Cheers, Joe

[devmonkey]

I needed an HTD5 pulley for a project, this was a bit odd as it needed to be fixed to a spinning bearing housing rather than to a shaft so I decided to make one on the machine. I used the cambam trochoidal plugin for the roughing allowing the entire part to be milled with a 3mm endmill, took a bit longer but meant no tool change. Part was cut dry with air to clear chips, worked out really well.

28713 28714

Trochoidal clearing is excellent, no gummy chips sticking to the endmill. This was cut from some 5000 series (not sure exact spec) scrap plate, it is quite gummy, would have been better in 6061 but I didn't have any to hand. Still the part is more than serviceable for its intended purpose which is to drive a multi-turn encoder (near zero load). Was cut at 3mm DOC, 15% stepover, 24k rpm, 3000mm/s for clearing (although the machine never gets anywhere near this on this part due to the tiny radius of the spirals).


https://youtu.be/skiw7pmDoQk

I will probably recut it with a high speed profile to finish, the finish pass was way too slow (600mm/s) which led to rubbing and the surface finish not being perfect. This was entirely my fault for rushing the CAM and not setting the feed.

[Clive S]

Very nice. I happen to make two the other day for an encoder 28 tooth .
28715
.
28716

[devmonkey]

Very nice. I happen to make two the other day for an encoder 28 tooth .
28715
.
28716

They are nice, what are they made out of and how did you cut that thread on the inside of the hollow one?

I just made it a baby brother, this one is going on an 8mm shaft with a 2mm shaft stuck in the end to go onto the encoder gearbox.

28717 28718 28719

[Clive S]

The thread was done on a Myford S7 converted to cnc with an encoder to do threading.
They are made out or 6082 I think

Didn't do a vid of the thread but had to make a male thread first. The hollow one is being fitted on the end of a spindle lathe to drive the encoder. The pin nut was off the lathe and used that for the fit 35x1.5mm

28720

https://www.youtube.com/watch?v=MyLwFqa02Bc

https://www.youtube.com/watch?v=ZR_W3w_rews

https://www.youtube.com/watch?v=IiMME9BcBHw

[Kitwn]

'Great minds' and all that.

Yesterday I used the CamBam trochoidal pocket plugin to cut out the teeth for a ratchet made from Jarrah hardwood. Conventional shallow cuts tend to take the corners off the teeth where the grain is tangential to the wheel so being able to do a 12mm depth of cut with a 2mm diameter tool is a bonus. I still had to redraw the original ratchet design (produced in Gearotic which will also draw HTD and many other pulleys for you) with rounded teeth to avoid some chipping. This cut used a two-flute straight cutter with a speed of 1000mm/min and a stepover of 0.15 (0.3mm) and the maximum 24000 rpm. This took about 25 minutes for the 110mm diameter wheel.

28721

[Sterob]

Very nice gentlemen....I want to be able to do that when I grow up.

[Voicecoil]

I needed an HTD5 pulley for a project, this was a bit odd as it needed to be fixed to a spinning bearing housing rather than to a shaft so I decided to make one on the machine. I used the cambam trochoidal plugin for the roughing allowing the entire part to be milled with a 3mm endmill, took a bit longer but meant no tool change. Part was cut dry with air to clear chips, worked out really well.
Well done, nice bit of work that. BTW, how did you hold it whilst machining? Pieces like that where the finished part doesn't have a lot of thickness I always find a bit of a challenge to hold down rigid.

[devmonkey]

'Great minds' and all that.

Yesterday I used the CamBam trochoidal pocket plugin to cut out the teeth for a ratchet made from Jarrah hardwood. Conventional shallow cuts tend to take the corners off the teeth where the grain is tangential to the wheel so being able to do a 12mm depth of cut with a 2mm diameter tool is a bonus. I still had to redraw the original ratchet design (produced in Gearotic which will also draw HTD and many other pulleys for you) with rounded teeth to avoid some chipping. This cut used a two-flute straight cutter with a speed of 1000mm/min and a stepover of 0.15 (0.3mm) and the maximum 24000 rpm. This took about 25 minutes for the 110mm diameter wheel.

28721

Very nice Kit! This was the first time I'd used trochoidal clearing, definitely a keeper for thin endmills in metal.

Now I have a question for you pulley experts. I need to make a very large 157 tooth 8m timing pulley ring that will be fitted to the perimeter of a winch drum. The pulley needs to be made of a ring of material ~25mm thick with an outer diameter of 400mm and inner of 340mm. It could be made from aluminium, SRBP, delrin or even hdpe.

Trouble is I don't want to waste a large piece of material just to make the ring, any ideas? I've even considered casting a blank in aluminium but I don't have a foundry. Are there any castable machineable urethanes that are strong enough for timing belt pulley teeth?

[Kitwn]

I'm well pleased with the trocho pocket plug-in, it's a major step forward in being able to make hardwood clock wheels sucessfully.

How are you designing the ring of teeth for the pulley? I have the Gearotic software which can create accurate timing pulleys in all sorts of formats. If you need help with that I can send you a PDF for importing into CamBam (that's another plug-in).

Well if you want to get into a new hobby that involves playing with fire, melting aluminium is easy. The YT video below gives you the idea. Use previously cast aluminium alloys (old water pumps from your friendly neighbourhood motor mender etc.) rather than cans or other scrap. Sawing off the bottom of an old fire extinguisher makes a good crucible. Getting a sucessful cast with it is another story however.

Not sure on the use of 2-part plastics for something this robust or how well they machine but one option to consider is to make a master in MDF with your CNC machine and then a silicone mold to cast the final pulley ring, or a blank for final machining. I've done a bit of this for arty stuff but not for engineering. You aren't going to buy from Australia but the Barnes site linked below has info on materials and how to use them. Similar stuff will be available in Pommieland.

https://www.barnes.com.au/

https://www.youtube.com/watch?v=lSoWxG30rb0

Good luck

Kit

[devmonkey]

I'm well pleased with the trocho pocket plug-in, it's a major step forward in being able to make hardwood clock wheels sucessfully.

How are you designing the ring of teeth for the pulley? I have the Gearotic software which can create accurate timing pulleys in all sorts of formats. If you need help with that I can send you a PDF for importing into CamBam (that's another plug-in).

Well if you want to get into a new hobby that involves playing with fire, melting aluminium is easy. The YT video below gives you the idea. Use previously cast aluminium alloys (old water pumps from your friendly neighbourhood motor mender etc.) rather than cans or other scrap. Sawing off the bottom of an old fire extinguisher makes a good crucible. Getting a sucessful cast with it is another story however.

Not sure on the use of 2-part plastics for something this robust or how well they machine but one option to consider is to make a master in MDF with your CNC machine and then a silicone mold to cast the final pulley ring, or a blank for final machining. I've done a bit of this for arty stuff but not for engineering. You aren't going to buy from Australia but the Barnes site linked below has info on materials and how to use them. Similar stuff will be available in Pommieland.

https://www.barnes.com.au/

https://www.youtube.com/watch?v=lSoWxG30rb0

Good luck

Kit

I use a free command line dxf generator that can generate all of the main timing pulley profiles, super simple.

I just cut an 8m test pulley in 10mm SRBP P1, if came out very nice and extremely strong teeth. I think I will use this material and laminate a few layers together then bolt through, it is pretty cheap stuff and locally available to me.

28722 28723

[Clive S]

I use a free command line dxf generator that can generate all of the main timing pulley profiles, super simple.

I just cut an 8m test pulley in 10mm SRBP P1, if came out very nice and extremely strong teeth. I think I will use this material and laminate a few layers together then bolt through, it is pretty cheap stuff and locally available to me.

28722 28723

That's cheating. Do you have a link for it. :cool:

[devmonkey]

That's cheating. Do you have a link for it. :cool:

Here you go, https://github.com/paulius-zubavicius/pulleys-profile-generator

You need java to run it.

[devmonkey]

I made the large timing pulley I was proposing today, it is a 157 tooth HTD 8M (20mm belt) for a project, pulley is ~400mm diameter. It was made from 10mm SBRP with 3 laminations at the rim. Turned out fantastic. I used screwfix cyano to laminate, this makes a very strong joint which you cannot get apart. It will never the less be reinforced with a ring of M5 bolts.

28735 28736 28737 28738

[Voicecoil]

I made the large timing pulley I was proposing today, it is a 157 tooth HTD 8M (20mm belt) for a project, pulley is ~400mm diameter. It was made from 10mm SBRP with 3 laminations at the rim. Turned out fantastic. I used screwfix cyano to laminate, this makes a very strong joint which you cannot get apart. It will never the less be reinforced with a ring of M5 bolts.

28735 28736 28737 28738

What a monster? what's it for may I ask?

[devmonkey]

What a monster? what's it for may I ask?

It is for winching humans into the air attached to paragliders, this pulley forms one side of the winch drum and is driven at around 10kW by a 25kW BLDC motor under FOC control.

[Kitwn]

Nice! It's very satisfying when the machine you have put so much time and effort into designing and building starts to produce that kind of output.:victorious:

[Evengravy]

Seconded, that's a lovely piece of work. :cool:

[devmonkey]

Tool deflection chaps, tool deflection!

So I received a bunch of single flute endmills from China, these things cut aluminium and SRBP nicely. However I'm having some problems, and would like your diagnosis,

So I've just cut a bunch of parts that need a hole that is a reasonable sliding fit on a 25mm shaft. I used one of my new 4mm single flute endmills, 2mm DOC, 1600mm/min into 10mm SRBP. It cuts like butter. However when I measure the hole afterwards it is larger at the top than the bottom by about 0.15mm, effectively the hole is tapered. If I cut the same hole with my usual 6mm 2 flute it is straight.

I assumed this was the single flute deflecting on subsequent deeper passes but it could also be that the endmills are themselves tapered?? What do you think?

I guess the solution is to rough with the single flute and finish with the 2 flute, just a bit of a hassle.

Also could it be that i'm spinning them too fast a 24krpm and they are bowing out due to centrifugal forces?

[Voicecoil]

What happens if you take the feed rate down? And how long is the flute on the single? greater than the depth of the material?

[devmonkey]

It improves somewhat with slower feed, the flute is longer than the material is deep. I just found one source of the problem the bolts holding the Y axis plate that ties the Z to the Y ballnut were not cranked down, they had worked themselves to just loose, nust have forgotton to loctite those one...

Anyway this tidied up the top diameters which are now spot on again, but with the 4mm bit the hole bottom is still smaller than the top, error is much less, maybe 0.05mm. This is enough to be annoying as you get a decent H8 fit to the shaft on one side and but tight on the other, opening up the hole doesn't solve the problem as it becomes loose on the top. Doesn't matter at all for normal profile work just shaft and bearing fits. I will stick with the 6mm known quality cutters for doing these.

I also just ordered a hopefully decent set of collets from Arc, not convinced by the much cheaper set I have based on the damage they take when you break an endmill.

[devmonkey]

I've done some more experiments to try and determine the source of the non-vertical walls I get cutting SRBP with single flute endmills. I surfaced the spoil board and trammed the spindle before running these tests.

In all cases the toolpath was a 25mm square in 10mm material, outer roughing profile to 9mm deep leaving 0.2mm for a single finish path at full depth.

#1 4mm Chinese SF, 0.2mm taper top to bottom,
#2 3.175 Chinese SF, 0.1mm taper top to bottom,
#3 6mm Euro SF, 0.03mm taper top to bottom,
#4 6mm Chinese 2Flute, 0.00 taper top to bottom.

The taper is not constant, rather it bulges towards the bottom where the skirt was left to hold the part for the finish pass.

I think that with this slightly flexible material the single flute cutters when slotting are leaving a slot smaller than the tool diameter, this is then cleaned up on the finish pass but not where the tool is cutting the skirt. The same material squeezing does not occur with 2 flute cutters. Has anyone else experience this with single flutes, I haven't noticed this happening in aluminium.

[JAZZCNC]

I always find single flutes cut undersize. Never been sure it's a deflection or the tool. I always finish with a 2 flute and I also find HSS gives a better surface finish than carbide in aluminum for the finish passes.

I hardly use Single flute cutters anymore. My preferred method is 3 flute carbide rougher leaving 0.4mm then 1x semi-finish pass @0.3mm and a final 0.1mm final finish pass. For work that doesn't need the best surface finish and accuracy I just take the full 0.4mm as a finish pass.

[devmonkey]

I always find single flutes cut undersize. Never been sure it's a deflection or the tool. I always finish with a 2 flute and I also find HSS gives a better surface finish than carbide in aluminum for the finish passes.

I hardly use Single flute cutters anymore. My preferred method is 3 flute carbide rougher leaving 0.4mm then 1x semi-finish pass @0.3mm and a final 0.1mm final finish pass. For work that doesn't need the best surface finish and accuracy I just take the full 0.4mm as a finish pass.



Hi Jazz, what 3 flute rougher do you recommend? I was quite looking forward to making less chips with these 1/8 single flutes but no point if you have to go around with a 6mm 2 flute to clean up,

[JAZZCNC]

Hi Jazz, what 3 flute rougher do you recommend? I was quite looking forward to making less chips with these 1/8 single flutes but no point if you have to go around with a 6mm 2 flute to clean up,

I mostly cut aluminum and use 8mm reduced neck Alu power cutters from cutwell tools, thou recently I've tried the APT version and they seem ok plus good price.

https://www.shop-apt.co.uk/economy-3-flute-roughing-carbide-end-mills-for-aluminium/3-flute-carbide-roughing-end-mill-for-aluminium-8mm-diameter.html

I also tried these and they give a very nice finish, but I've not got enough time on them yet to see how well they last.
https://www.shop-apt.co.uk/3-flute-veriable-unequal-helix-carbide-end-mills-for-aluminium-dlc-coated.html

[devmonkey]

Those look excellent value thanks. How deep DOC do you go with the rougher in ali?

[JAZZCNC]

Those look excellent value thanks. How deep DOC do you go with the rougher in ali?

It depends on the grade. I mostly use Cast tooling plate and I can take 4mm DOC slot milling with coolant but I often keep it to half that with a bump in feed to save the cutter but run it dry with air. to keep the mess down. Feeds n speeds I tend to play around with depending on the job but they range from 1000mm/min to 1600mm/min. RPM 12K to 20K often around 15k.

I did play around with deeper DOC and the cutters will happily take 100% Dia but my 2.2Kw spindle isn't very keen, there's nothing left for a safety margin and a slight soft spot will stall it.
Side cutting it will happily take 1.5xD and I often cut 20mm plate full pass just dibbling it away 0.30mm, which is roughly the serration depth, hence my 0.4 finish passes. When using i-machining (adaptive) I can really ramp it up and it's crazy fast at full DOC.

[devmonkey]

It depends on the grade. I mostly use Cast tooling plate and I can take 4mm DOC slot milling with coolant but I often keep it to half that with a bump in feed to save the cutter but run it dry with air. to keep the mess down. Feeds n speeds I tend to play around with depending on the job but they range from 1000mm/min to 1600mm/min. RPM 12K to 20K often around 15k.

I did play around with deeper DOC and the cutters will happily take 100% Dia but my 2.2Kw spindle isn't very keen, there's nothing left for a safety margin and a slight soft spot will stall it.
Side cutting it will happily take 1.5xD and I often cut 20mm plate full pass just dibbling it away 0.30mm, which is roughly the serration depth, hence my 0.4 finish passes. When using i-machining (adaptive) I can really ramp it up and it's crazy fast at full DOC.

Right i'll order a couple, sounds excellent. Are you saying you can cut cast plate full depth (20mm) at 1000mm/min using trochoidal/adaptive? I haven't tried anything like that aggressive, I usually cut at 6mm DOC with a 6mm cutter, ~1600mm/min and 1mm stepover into plate with trochoidal.

[JAZZCNC]

Right i'll order a couple, sounds excellent. Are you saying you can cut cast plate full depth (20mm) at 1000mm/min using trochoidal/adaptive? I haven't tried anything like that aggressive, I usually cut at 6mm DOC with a 6mm cutter, ~1600mm/min and 1mm stepover into plate with trochoidal.

No Using i-machining which is Solid-Cam's version of trochoidal I can cut much more aggressively than 1000mm/min at 20mm DOC. i-machining adjusts the feed based on geometry while it's cutting and usually, it's anywhere between 2000-4000mm/min and 14-18000rpm. And that's on the medium setting 4, if I bump up the level to 7 it really rips chips, but it knocks the hell out the machine and spindle so I keep it between 3-4. This usually spits out feeds between 2-3000mm/min, 12000rpm and varies the step over between 0.1 and 0.9mm. See Pic

28841

[szecho]

I have read this thread and I have a question.
I saw you have plates welded on the frame and a washers under the supports.
How did you position the ball screw support (bk/bf) on the frame? By the thickness of the washers you set the screw to be parallel with the rail?

[Tom J]

Hi
Is it SKU IR26526 profile?

No delete option for duplicate post - sorry:(

[Tom J]

Gantry extrusion has arrived from KJN, it is a real beast. KJN cutting appears very accurate.

26902 26903

So now I need to decide how to mount 15mm rail to the extrusion. I was considering 3 options:
1. Mount rail directly to extrusion between the two slots, drilling and tapping.
2. Mount rail on aluminium plate which is in turn bolted to the slots.
3. Mount rail directly to slot.

This shows that the profile is not flat across the width which I believe is by design so that the slots pull up to meet whatever you are clamping to them. The slot edges are 0.15mm lower than the centre, the centre and the 2 corners are perfectly aligned within the resolution I can measure with finest feeler gauge.

26905

Now I have the extrusion (1) is a non-starter as there is a very slight peak in the centre and I would need to mill it flat which is hard given the gantry is longer than my mill table, although not impossible.
(2) is what many people seem to do and so obviously works but is more work for me.
Can I get away with (3), see picture below, what do you think?

26904

I have a 50mm fly cutter, it would be possible given the extrusion is already flat and parallel to fly cut between the high points, would have to tram the mill carefully first.

Hi
Is it SKU IR26526 profile?

[devmonkey]

Making some minor improvements to my machine so thought I'd report them here.

I've been using an mdf spoil board with t-nuts, I have hated this solution from the beginning but have been avoiding tapping 250 M6 holes in the alu bed plate.

I decided to just do it, I wasn't going to tap them by hand so I purchased a thread cutter for M6. The bed is made of 20mm cast aluminium plate, since there was only one chance to get this right I tested all the feeds and speeds on some scrap identical plate first. This was the first time I have used a thread cutter and you must dial in the toolpath first to get the type of fit you want on the bolt (basically change the diameter by 5 microns until happy). I'll give some details of the feeds and speeds to help anyone else processing similar material with a 2.2kw chinese spindle on a job that cannot be messed up.

1. Drilling was the main concern with the chinese spindle as it has not much torque at the rpm drill bits need. I found the sweet spot to be the spindle slowed down to 5300 rpm, feed rate 700mm/min, 2 pecks each half depth to total depth of 19mm. Less rpm or more feed and spindle would stall, more rpm or less feed and chips would melt on the tip.

Sprayed some WD40 on it for each hole. This created nice shiny round holes, bit was still like new afterwards, not bad for something that probably cost a few pence.
32355
32354

2. Chamfered with a 6mm 45 degree bit.

3. I blew out the holes, then sprayed WD40 over each.

4. Thread cut bottom up with this bit at 16000rpm 250mm/min:
https://www.shop-apt.co.uk/metric-single-tooth-thread-mills-for-titanium-and-hrsa-materials-internal-60-full-form/HRST060480N1.0.html

32352


https://www.youtube.com/watch?v=FriMvIZroH4

Result was a perfect grid of blind tapped holes on 50mm pitch, very pleased.

32351
32353

[devmonkey]

32356

New spoil board machined with thru holes to reach the threaded bed holes. Missing corner is for vice and 4th axis which sit on their own registration plates. Sometimes I need to cut jobs that use the whole bed, when I next need to do one i'll make another spoil board without the missing corner, they only take a few minutes to swap and surface.

Next improvement will be replacing the standalone ddnc 3.1 controller with UCCNC and their ethernet motion controller. I need the ability to write custom plugins / macros which is missing / hard with the ddnc.

Does anyone know if there is stock of these in the UK, I vaguely remember someone, Jazz?, may have been a distributor?

[Clive S]

32356

New spoil board machined with thru holes to reach the threaded bed holes. Missing corner is for vice and 4th axis which sit on their own registration plates. Sometimes I need to cut jobs that use the whole bed, when I next need to do one i'll make another spoil board without the missing corner, they only take a few minutes to swap and surface.

Next improvement will be replacing the standalone ddnc 3.1 controller with UCCNC and their ethernet motion controller. I need the ability to write custom plugins / macros which is missing / hard with the ddnc.

Does anyone know if there is stock of these in the UK, I vaguely remember someone, Jazz?, may have been a distributor?

Yes. Jazzcnc is a member of this forum. https://www.jazzcnc.co.uk/