PDA

View Full Version : BUILD LOG: Wobblycogs CNC Mk2



Wobblycogs
15-08-2013, 02:18 PM
Hello again everyone,

My first CNC (http://www.mycncuk.com/forums/router-build-logs/5299-wobblycogs-cnc-mk1.html) design was, in hindsight, perhaps a little bit ambitious for a first build. After building one of the sides I realized that welding was the only realistic option and that my current workshop just wasn't big enough to fit the machine (and it would weigh too much for the floor).

Rather than wait for the workshop to be built I thought I'd come up with a new design for a more modest sized CNC machine that will fit in the space I currently have. There's not much to see yet but I was hoping to get a bit of feedback on the pulley arrangement for the x-axis as this is an area that I'm not very familiar with. Firstly some pictures...

9461
9460
9457
9458
9459

The frame is going to be all extruded aluminium if possible. The blue sheet forming the bed is 600x600mm to (hopefully) give a cutting area of 400x400mm. Z axis travel will be fairly limited, perhaps as little as 50mm, as this machine is being designed to mainly cut aluminium.

The x-axis rails are SBR16 although I could probably fit SBR20 if necessary. The rails are sitting on 60x90 heavy extrusion as sold by KJN. There's 180mm between bearing block centres.

The ball screws are currently spec'ed as 1605 with 16 tooth HTD5 pulleys on the end (with extra long end machining). The stepper will be a nema23 but I've not picked one out yet. The stepper is fitted with a 32 tooth HTD5 pulley to give a 2:1 ratio.

The stepper and the two idler pulleys are mounted on a 10mm aluminium plate. For build simplicity the idlers are used to take up the slack on the belt rather than the stepper. The stepper is mounted using stand off legs. The belt length (at central idler adjustment) calculates as 1736mm

The plan is to have this machine standing vertically, or perhaps wall mounted, so that the x-axis belt is at the top.

My questions so far...


Does the belt make enough contact with stepper motor pulley? I'd guess about 12 to 15 teeth are in contact with the belt.
How is the 32 tooth pulley attached to the stepper shaft? There's only 15mm of shaft on the stepper so I'm guessing it will have to be a taper lock (also Belting Online don't bore 32 tooth pulleys at 6.35mm).
Where can I get idler wheels from? Neither Belting Online nor Bearing Station seem to do them. I'm tempted to follow Jazz and use bearings and penny washers (http://www.mycncuk.com/forums/router-build-logs/4513-3-axis-cnc-router.html#post30518)...


Cheers :-D

JAZZCNC
15-08-2013, 04:29 PM
I'm tempted to follow Jazz and use bearings and penny washers

Nothing wrong with Bearing and penny washers they've been working fine for 6yrs now.!! . . . . That bit of string for a belt won't work thou. .:joker:

Ok so first don't go with 1605 and 2:1 ratio it's not need and you'll want more speed than that.
Direct 1:1 will be enough for cutting aluminium. If you want to go 1:2 then buy 1610 then you'll get same resolution and 1:1 speed has 1605 but double the torque of motors so lots of torque while cutting at low speeds. You'll rarely cut above 1500mm/min in aluminium.
With 1:2 ratio on 1610 then it's a simple switch of pulley if you want speed for routing, plastics etc.

Next the frame is no where near strong enough for cutting aluminium. Beef up the lower portion, I'd suggest an outer frame with inner cross braces. Really those corner angle brackets are rubbish so I'd think about some good strong plates with plenty of bolts.

If your going vertical then you'll need to think about the gantry, Z axis has depending on which way you orientate it will depend on how chips collect on surfaces etc. You'll certainly be wanting to put covers on Z axis and belts to help stop chips collecting if your putting belts at top.
It's like this Aluminium chips fly every where and they go up before coming down so just think about what they will land on and or were could collect.
In general it's not a problem more of an annoyance because on my machine the Z axis front plate faces down and if doing again I'd face up and put cover on because the chips collect on rear of front plate. Basicly Z axis is on wrong side of gantry facing down, I'd put on top side facing up.

Hope this helps.

Wobblycogs
15-08-2013, 06:28 PM
Thanks, as always Jazz your reasoning makes perfect sense on the ballscrews so I'll change then to 1610's. Hopefully this new design will be stiff enough. As for the bely I was thinking I could save a bit of money if I just used a bit of string pulled tight ;-)

Everywhere you see a hole that looks like it should have a bolt there will be one.

This shot is of the base with the new base plate removed showing the new end pieces on the bed rails - these will be bolted on. The corner brackets are still there but they are more to affix the x-axis to the bed rather than hold the whole thing rigid.
9472

The back plate has grown "ears" and bolt holes. Long bolts will run down into the x-axis and the rails at the end of the bed.
9469

There's a new front plate (10mm) which again has bolt holes for the x-axis and the bed.
9471

Finally the base has been covered in a 5mm sheet of aluminium and bolted all over the place for bracing.
9470

I forgot to mention in my first post but the blue bed piece will also be bolted to the bed for bracing.

The only downside I can see of this is chips will accumulate on the front plate, a small price to pay if it's cutting aluminium accurately! :chuncky:

Iwant1
15-08-2013, 07:05 PM
Jazz,

sorry if I'm getting this completely wrong, but the op wrote a gearing ratio of 2:1, but also has written and drawn a 32 tooth pulley on the motor and 16 tooth pulley on the screw. Doesn't this setup actually give him a 1:2 ratio, so 1 turn motor gives him 2 turns screw. So theres plenty of speed, which isn't needed as you say 1:1 is ample for aluminium, the op's main working material.

Adil

Wobblycogs
15-08-2013, 07:08 PM
I think I've got the ratio the wrong way around, it should probably read 1:2. To clarify there is a 32 tooth pulley on the stepper and 16 tooth pulleys on the ballscrews.

JAZZCNC
15-08-2013, 08:35 PM
Jazz,

sorry if I'm getting this completely wrong, but the op wrote a gearing ratio of 2:1, but also has written and drawn a 32 tooth pulley on the motor and 16 tooth pulley on the screw. Doesn't this setup actually give him a 1:2 ratio, so 1 turn motor gives him 2 turns screw. So theres plenty of speed, which isn't needed as you say 1:1 is ample for aluminium, the op's main working material.

Yes but I knew what he meant because of the pulley sizes but I copy and pasted so didn't correct it sorry.!!

Edit: Corrected now..!:thumsup:

Wobblycogs
17-08-2013, 10:07 AM
The design has progressed a bit so I thought it was time to get a bit of feedback...

9523 9524 9525 9526 9527 9529

I've decided to try to increase the maximum material thickness to more like 100mm (up from 50mm before). To achieve this though I seem to have ended up with an unusual bearing block placement on the z-axis, see image 2. The idea is to be able to lift the z-axis high enough so that the bottom of the gantry is the limiting factor for the material depth. As drawn I have 180mm top to bottom between the z-axis bearing blocks but I could easily increase this - I was thinking 250mm would compensate for the additional over hang of the z-axis.

The gantry has changed design several times :chuncky:. It started out as two pieces of 45x45 heavy (a bit like this (http://www.mycncuk.com/forums/router-build-logs/4513-3-axis-cnc-router.html#post30777)) then got changed to two pieces of 60x90 heavy , then to a single piece of 90x90 heavy.

I finally settled on two pieces of 45x60 heavy with a backing plate of 10mm aluminium, the unsupported span is 610mm. The deflection calculator on the Metallin website (using the closest match) seems to think the z-axis deflection should be ok. Using 60x90 heavy would reduce deflection, I'm sure, but it was expensive and reduced the x-axis cutting. I think this is probably a fair compromise.

The belt arrangement of the x-axis has been improved. The ratio is now 1:1 and there's better belt contact with the stepper pulley.

I've not added the y or z-axis steppers into the diagram yet (should be fun). Also still to be drawn is a chip guard for the y-axis screw. There will be a thin plate with a long slot (and maybe brushes) for the y-axis connection to the z-axis assembly.

JAZZCNC
17-08-2013, 01:24 PM
Ok it's getting there now.!! . . . Few more things I'll comment on.

I can see what you trying to achieve with Z axis but doing this your compromising it's strength where it will probably spend most of it's time cutting and that is Extended further out. Because the bearings are high up the extension distance is long when cutting thinner material and I'd estimate you'll rarely cut thicker than 50mm material.
Much better way would be to put bearings lower down but not flush with bottom, leave clearance for bolt access to Y axis bearing plate. Then make the spindle mount adjustable on front plate if you ever need to cut tall material.
Looks like your planning on using WC spindle so that alone can slide in the mount easy enough so you'll find a intermediate location for spindle mount that will allow you to adjust for all materials.


Next the material thickness of the Front and Rear Z plates are not thick enough and they could resonate and flex. Wouldn't go less than 20mm here.
The plate on the rear of moving Y axis carriage isn't needed or adding much strength so drop that It's only adding weight and cost. Good way to help gain strength in this area is to make a cover for the Z axis from folded 3mm ali plate and bolt into sides of Z rear plate and Y axis top bearing plate. This way it braces this area and protects the Z axis ballscrew belts etc see pic of cad 3d model and unfolded sheet, Next week I'll show real thing cause I've got to make one.
95319530

Also Got a way for you to increase cutting area and increase Y axis strength but will be hard to explain so may end up drawing it to best show.? I'll try explaining first.!

Drop the profile spacers on bearing plates this will allow you to use longer rails that extend right up to gantry sides. This will allow the Y axis bearings to travel over the X axis bearings. Now because you can still have the same Y axis bearing spacing(or more) you can drop the width of Z axis front plate and therefore gain some cutting width with no loss strength.
You'll need to brace the X axis bearing/gantry plate area but that's easy enough done with triangle plates.!

Hope this makes sense.??

Robin Hewitt
17-08-2013, 01:50 PM
I can see what you trying to achieve with Z axis but doing this your compromising it's strength where it will probably spend most of it's time cutting and that is Extended further out.

What he said, plus, unless there is something in the way that I can't see, might you move that spindle motor backwards, sink it through it's mounting plate so it sits between the Z slides rather than way out in front of them?

JAZZCNC
17-08-2013, 02:04 PM
What he said, plus, unless there is something in the way that I can't see, might you move that spindle motor backwards, sink it through it's mounting plate so it sits between the Z slides rather than way out in front of them?

Hey.??? . . .Even I'm not with you here Robin how's he going to do that without a completely different Z axis design.?

I know what you mean if re-design but to accomplish this he'd have to turn rails/bearings 90Deg and that causes other weakness and issues which IMO would be worse than spindle extending out.? . . . If you have another way on the same theme he's taking now that I'm not seeing then get doodling.!!

Wobblycogs
17-08-2013, 03:16 PM
Really like the z-axis idea Jazz, thanks for that. As suggested I've...


Moved the bearing z-axis blocks down to almost the bottom of the plate.
Increased the spacing between the z-axis bearing blocks to 250mm.
Beefed the z-axis up to 20mm plate all round (except the bottom spacer plate, will this be ok?).
Upgraded the motor mounting brackets to 15mm as they looked a bit weedy up against the 20mm back plate.
Put holes up the z-axis plate so the brackets can be adjusted up and down, in the lowest setting shown I have about 45mm of cutting depth with a 50mm bit. I'm sure this will be enough for the vast majority of the time.
Added the chip plate for the y-axis ballscrew (still need to actually attach the z-axis to the screw)


95329533

I've got to get back to fixing the house (plastering, one of my least favourite jobs) but I'll try to draw up the other ideas tonight.

P.S. I was thinking the gantry back plate would have to stay because the screw is mounted to it but now I think about it I could mount the screw on one of the gantry beams. I'm tempted to still lock the two beams together vertically but with a small plate in the middle of the span. I'll have to have a think about the cover idea as I'm not clear how it would work (I think I've seen what you're talking about in another thread though).

Cheers.

JAZZCNC
17-08-2013, 03:55 PM
P.S. I was thinking the gantry back plate would have to stay because the screw is mounted to it but now I think about it I could mount the screw on one of the gantry beams. I'm tempted to still lock the two beams together vertically but with a small plate in the middle of the span. I'll have to have a think about the cover idea as I'm not clear how it would work (I think I've seen what you're talking about in another thread though).

No you misunderstand here I think.? The plate on the Rear of the moving Y/Z axis carriage I meant. It does very little in adding strength, think of box with no sides.!!
Often this way causes more hassle than it's worth because if the edges are not machined perfectly 90deg then when tightened it twists the assembly on the bearings and locks up solid.!!

The slotted front way works great, you've probably seen I have design that uses it so know it works fine but it does cost more. It does offer some nice advantages thou like ball-screws are protected and every thing is neatly out the way. The way I do it the motor and belt are all inside the Gantry and covered up.

The other design which would suit your machine nice and is cheap and easy to build is the L shape gantry using 45x90. I've made few machines with this design and it's proven design which is very strong and reliable. Has some of the same benefits the Slotted design does in that the screw and motor is placed behind the Gantry out of the way. It does put the screw back a bit further than is considered ideal but again It's proven to be no issue in real world use.
If this design interests you then tomorrow I will have hopefully put the final touches to a gantry I'm building so can show the real thing with cable chain, Z axis covers etc instead of 3D model.

Wobblycogs
17-08-2013, 06:18 PM
Ah, I see what you mean now and it makes perfect sense. I don't have an abundance of metal working tools (yet) so making two perfectly parallel faces would have been a challenge, presumably that would normally be done by milling them parallel?

I considered the L shaped gantry design but I'm hoping to mount this machine vertically at some point and it seemed simpler to just use a piece of plate at the back and benefit from a better / simpler connection with the ball screw.

Right, on with the design...

Wobblycogs
17-08-2013, 08:15 PM
Jazz, was this what you meant when you were talking about having the y-axis bearings travel over the x-axis and make the front z-axis plate narrower? I like this design as I've gained about 30mm of cutting width and I've got a wider spacing on the y-axis bearings (in fact it looks like I could still go wider). The z-axis front plate is now just wide enough to take the rails and the ballscrew.

9537

I wasn't happy with the x-axis cutting length after putting the thicker 20mm plates on the z-axis so the x-axis has grown 100mm :-). I've also been working on getting the stepper in for the y-axis. You can see the hole in the gantry back plate for the belt. The stepper will be mounted on a gantry side plate.

JAZZCNC
17-08-2013, 08:18 PM
I considered the L shaped gantry design but I'm hoping to mount this machine vertically at some point and it seemed simpler to just use a piece of plate at the back and benefit from a better / simpler connection with the ball screw.

Both would work fine for your needs. The ball-screw connection is no more difficult or worse really and actually it's easier to align ballnut mount. The plus with this design is it's cheaper because you won't need the rear plate and the screw and motor belts etc are out of firing line of chips etc. . . It's neat and works well with cable management, which is something you should be aware of and design for has it's often overlooked and regretted when finished.!

JAZZCNC
17-08-2013, 08:41 PM
Jazz, was this what you meant when you were talking about having the y-axis bearings travel over the x-axis and make the front z-axis plate narrower?

Yes that's it. Just to help drop weight a bit and look a bit better then think about making the back plate thinner in middle but with ears at top and bottom.? Like this.9538

I've made a couple of Z axis for folks like this and works good and looks better.

JAZZCNC
17-08-2013, 08:46 PM
Another thing.? . . Regards the spindle mount make it so can de fastened from front and not from behind front plate. May seem like a small detail but you'll thank me for it if did it other way round.!

JAZZCNC
17-08-2013, 08:52 PM
Another thing I'd like to say is that 45x45 profile is weak so if that bed frame is all 45 then I'd invest in beefing up to 45x90 you'll get a much stronger machine for not much more money.! . . . No point going to all this trouble and spoil it for an extra £50.!

Wobblycogs
20-08-2013, 07:34 PM
Thanks for the suggestions Jazz most are drawn up already. Anyway, here are some shots of the project so far.

958495859586958795889589

I've added in a triangle brace for the gantry side plate but I rather like the one here (http://www.mycncuk.com/forums/router-build-logs/6382-steel-frame-cnc-router-design-build-2.html#post48181) so I might switch it for that. Depends on how much of an improvement it would make though.

As suggested the spindle mounts have been redesigned so that they fix from the front and the z-axis back plate now has a waist to reduce weight.

I initially had the z-axis stepper using stand-off legs like the x and y steppers but I felt this would weaken the top plate too much so it got a proper frame. The benefit of this is I can lock the z-axis back plate into the stepper support for additional rigidity. The stepper and pulleys will also be enclosed in a bent metal box.

The y-axis stepper has been mounted on an extension of the right hand gantry plate. Again this will be enclosed in a bent metal box to stop chips getting in. I've split the front chip guard in two so that the left section can be removed for accessing the belt and pulley for the y-axis without taking the machine completely apart (more cunning than a fox)!

I've had a think about where cables will run (very good suggestion). I think a cable drag on the right hand side and across the back should work fairly well. I've not even considered drawing it yet but I don't see anything obvious in the way to stop that idea.

I've not upgraded the bed to 45x90 yet because I wasn't sure if it should have the 90 edge vertical or horizontal. I assume vertical to provide resistance to twisting.

Now for the problems...

I've been reading loads of threads and I'm getting concerned that 45x60 profile in the gantry isn't strong enough and that I should probably go with 60x90. It's a damn sight more expensive and I don't want to upgrade if it won't help but I'd rather have a machine that was a bit over built than a bit under built.

The plan currently has SBR16 rail all over but I (and and a lot of other people it seems) have noticed there's not a lot of clearance (e.g. none) on the z-axis ballnut. The million dollar question is do I switch to SBR20 rail or try and rout out a pocket for the ballnut on the z-axis if needed?

And the big one... At the moment the rail is mounted straight onto the profile. Ooops, school boy error. I think I'd get away with it on the x-axis as there's a good 6mm of aluminium to bolt into. On the y-axis though it's clearly not enough as drawn. Presumably a 10mm plate bolted to the profile will fix this? Does it need to be toolplate?

Some other, probably stupid, ideas have come to mind though, would this machine be capable of engraving and or cutting brass? I've cut a couple of small bits of brass with a hacksaw and it seemed similar to aluminium in hardness. I've also considered making the x-axis significantly longer, say 1000mm of cutting length, I'm trying to resist that though as I still want to build my Mk1 machine which will be designed to handle large pieces.

EDIT: As well as considering using SBR20 rails what about moving to profiled rails instead? They would be a budget buster but I believe I could mount 15mm rails straight onto the extrusion and from what I've read they would lead to a better machine over all. A couple of lucky buys from the far east might make it possible.

JAZZCNC
20-08-2013, 10:26 PM
Now for the problems...

I've been reading loads of threads and I'm getting concerned that 45x60 profile in the gantry isn't strong enough and that I should probably go with 60x90. It's a damn sight more expensive and I don't want to upgrade if it won't help but I'd rather have a machine that was a bit over built than a bit under built.

Absolutely would go bigger and even at 60x90 you won't be over building believe me.!! . . like the Famous film words.? "Don't under estimate the force" Wobbycogs.!!


The plan currently has SBR16 rail all over but I (and and a lot of other people it seems) have noticed there's not a lot of clearance (e.g. none) on the z-axis ballnut. The million dollar question is do I switch to SBR20 rail or try and rout out a pocket for the ballnut on the z-axis if needed?

Either has you can always pack for clearance and it keeps Z axis compact and light. Bigger will give clearance at the cost of weight and extra standoff.


And the big one... At the moment the rail is mounted straight onto the profile. Ooops, school boy error. I think I'd get away with it on the x-axis as there's a good 6mm of aluminium to bolt into. On the y-axis though it's clearly not enough as drawn. Presumably a 10mm plate bolted to the profile will fix this? Does it need to be toolplate?

Yes use a 10mm plate but wouldn't bother with tooling plate has the rail base will have more error than the standard plate so waste of money if your not going to skim the rail base.!


Some other, probably stupid, ideas have come to mind though, would this machine be capable of engraving and or cutting brass? I've cut a couple of small bits of brass with a hacksaw and it seemed similar to aluminium in hardness. I've also considered making the x-axis significantly longer, say 1000mm of cutting length, I'm trying to resist that though as I still want to build my Mk1 machine which will be designed to handle large pieces.

If it will cut Ali properly then it will cut brass Ok.

njhussey
21-08-2013, 09:05 AM
Graham, liking the new design! When you want some 10mm and 20mm Ali plate for your Z & Y axis give me a shout as I've got loads stashed away (a 2200 x 950mm piece of 10mm 6082 (albeit cut up into smaller pieces now...) and a 1065 x 850 piece of 20mm) and won't be using anywhere near all of it. I'm sure we can come to some arrangement...

Wobblycogs
21-08-2013, 09:10 AM
Thanks Neil, this CNC building lark is expensive but every little saving helps. I saw a massive piece of 25mm aluminium plate go on eBay over the weekend. I was wondering if you'd grabbed it. I took a punt but I wasn't trying too hard as it's thicker than I need.

njhussey
21-08-2013, 09:58 AM
Well when you have settled on the sizes you want give me a shout...free Ali in exchange for machining services could be good?

Wobblycogs
21-08-2013, 10:05 AM
Sounds like a plan but I think you might be over estimating how quickly I can get this machine built :chuncky:

I'm going hell for leather at it presently but I'd guess it's still going to be a two or three months at least before it's built.

njhussey
21-08-2013, 11:30 AM
Quicker than mine at the rate I'm going!! I'll probably look at it again when I get back off holiday and blinking re-design it again!!

Wobblycogs
21-08-2013, 10:29 PM
Well I've broken rule number one and bought the first pieces of the machine before the design is properly finished. I decided to take the plunge and get some second hand THK profiled rails. They are a bit longer than the supported rails shown in the design above so I'm beefing up the gantry. The X and Y rails are SHS15R@1040mm and the Z is SSR15XW@410mm. I've since read that SSR are better suited to horizontal alignment because of the bearing arrangement but to be honest they are working so far below their rated maximum load I'm not worried.

I have a quick question about mountings the rails on extrusion though.

For the Y-axis I was thinking of doing this:

9678

Does this look like a good idea? There's 4mm of aluminium and the bolt is an M5 so that would mean 5 threads. I'm concerned that if there's the slightest bit of warp in the extrusion it'll throw the rail out because it's fairly snug in that central channel.

For the X-axis I've got a couple of options:

96839684

Neither seems to be a great option but the one with the rail over the T-slot looks precarious to me.

I'm seriously considering ditching extrusion and going with rectangular section. Any one know where I can get some with decent wall thickness?

EDIT: Scratch that last question, I must be going blind, Aluminium Warehouse do 2500mm 100x50x6 box section for £52. That makes it significantly cheaper than the 60x90 extrusion I was going to use.

Wobblycogs
22-08-2013, 11:43 AM
I couldn't get off to sleep last night for thinking about the CNC so I decided to get up an do a bit to the design. I wanted to address a couple of issues:

1) the rails didn't mount very cleanly onto the aluminium extrusion
2) it felt like too many parts were becoming sufficiently complicated that I would need access to a milling machine. I know I could get some / all of parts made for me but part of the fun of this project is coming up with a design I can make myself.

With those issues in mind I switched over to using aluminium rectangular section 100x50x6. The other major change was to mount the rails on the front of the gantry rather than top and bottom. I'm not sure if this is the greatest of ideas but I think it'll make the build quite a bit easier.

The shots below show the Y and Z axis together. There's a bit of tidying up to do moving bolt holes but the basic structure is in place.

973097319732

If vibration is an issue with the box section on the gantry I'll pack one or both beams with fine gravel / sand. Anything shown in red will be tooling plate.

JAZZCNC
22-08-2013, 07:19 PM
I couldn't get off to sleep last night for thinking about the CNC so I decided to get up an do a bit to the design. I wanted to address a couple of issues:

Ah ah your officially Addicted . . . .At your first meeting the words to use are " Hello my names Wobbly and I'm an addict" . . .Welcome to the club.!!

Now regards the changes then your making it worse not better.?? . . . Distance from Ballnut to Z axis rear plate is too much and that long bracket will flex and resonate.
I'd just flip the plate to front and cut a slot in it and mount ball-screw on back side so the distance is less, your also protecting the screw and motor. Stick another thin plate on rear and it's fully protected and helps stiffen gantry.

Don't go with your current design you'll regret it in this form.!!

Wobblycogs
22-08-2013, 09:32 PM
Yep, I'm completely addicted now. Work is suffering, I'm getting no sleep, the bank manager thinks there's been a robbery...

Lol, when I was drawing that long connection bracket I could hear you in my head saying "that's too long" :-D

As an alternative to mounting the 10mm plate on the front of the gantry could I instead mount the ball screw on the gantry supports like this and put a thin (5mm?) plate with a slot on the front? That gets the screw closer to the z-axis back plate and the gantry gets the full support of a 10mm plate on the back.

97349737

I've had to significantly increase the size of the base to accommodate the new rails. I can't really afford a full sheet across the whole base but I've added corner supports all around the bottom and the top has 200mm wide plate running the full length. All plate is 10mm thick and will be thoroughly bolted to the 45x90 heavy extrusion.

97359736

I actually feel like I didn't do to badly with the redesign if I only needed to move the screw, perhaps there's hope for this machine yet :joyous:

JAZZCNC
23-08-2013, 12:25 AM
As an alternative to mounting the 10mm plate on the front of the gantry could I instead mount the ball screw on the gantry supports like this and put a thin (5mm?) plate with a slot on the front? That gets the screw closer to the z-axis back plate and the gantry gets the full support of a 10mm plate on the back.

Yes you could no problem but you'll get more support for gantry from that plate if it was on front has the rails will be fastened to it and can't flex in the middle. Has it is now the Box in the middle could flex has the only support is on the back side and the rails applying all the pressure are on the front side.!
You'd be better with it on front and either another thinner plate across back adding bit of support or 2-3 10mm plate braces across back.

JAZZCNC
23-08-2013, 12:28 AM
Hold on.!!! . . . . You are having gantry sides aren't you.?? . . . .Your not planning on having it open like those pics show are you.?

I just thought you where showing it like that to save drawing them.!! . . . If not having sides then your in trouble and this won't work very well.!

Edit:

Ok on looking again it's not that bad but think it would be benefit from gantry sides has it would help stiffen things up. Also it would make a better stronger connection to the ballscrew.
Sides and the plate on front would make it quite bit stiffer with less flex.!

The way you have it now this would need beefing up anyway.

Wobblycogs
23-08-2013, 07:43 AM
One beefed up gantry with added sides. Now I draw it I can see this is significantly stiffer than what I had, I was relying on the 10mm plate not bending along it's weakest axis, doh!

97409741

As you suggest there will be 3*10mm thick plates on the back with the rest covered with a thin sheet. The front will be a full 10mm sheet with a slot for the ball screw. Feel free to imagine bolt holes all over the place ;-)

Thanks again for all the help.:encouragement:

EDIT:

There you go it's drawn up

9743

JAZZCNC
23-08-2013, 02:02 PM
It's close very close but not quite Carling.!!. . .Lol

Gantry's fine but those drop brackets for X axis ballnut mounts are too flimsy so make bit wider at top and beef them up a bit and your there.!

Wobblycogs
23-08-2013, 02:31 PM
Gantry's fine...

:yahoo:

I've bunked off work for the afternoon to do some CNC design so I'll get those x-axis brackets beefed up a bit. I'm also swapping the ballscrew and stepper around inside the gantry (so the screw is at the top) that way the slot is further from the chips.

I've been reading a few posts of yours regarding aluminium machining. I'm going to have a crack at using my circular saw to cut most of the pieces and then clean up with a router bit. I've tried routing aluminium in the past but only 5mm sheet. It went well but I'd agree with you it was scary at first.

I was also having a read up on pulleys. I've got 9mm 16 tooth on there at the minute but I think I'll upgrade them to 15mm 20 tooth (all HTD5). Will a 15mm pulley (like this (http://www.beltingonline.com/20-tooth-htd5-pulley-20-5m-15f-7776)) mount cleanly onto the shaft of a typically 3Nm stepper? I'm just a little concerned because the L dimension on the pulley is 26mm and the shaft of the stepper seems to only be 16 to 24mm long depending where I look.

P.S. Bloomin' love this (http://www.mycncuk.com/forums/router-build-logs/6274-1rst-build-time-do-next-step.html#post47101) design by the way. Once I have my workshop built I think there might have to be a bit of creative copying for CNC number 2 :whistle:

Wobblycogs
27-08-2013, 09:52 AM
Another weekend comes and goes with much work on the house and little on the CNC, sigh. Anyway, after getting a tick for the gantry :biggrin: I've been going around the design putting in all the bolts. I figure it's better to find bolt hole collisions at this stage rather than when I come to build the machine. It's a really slow process even (even using the hole wizard) but I think it's worth it. For example I thought most holes would be using M5 bolts with some M6. In reality it's turned out most will have to be M4 with some M5. Also the thread depth on the linear bearings is surprisingly short so some careful counterboring will be required. The Z-axis stepper has grown a bent metal box cover. I'll probably end up adding a hinged flap on the front at some point - there was a front originally but I realized would wouldn't be able to fit the stepper through the hole I'd left!

97819782978397849785

The X-Axis is still has quite a bit of work to be done on it. As you can see there's no stepper, pulleys etc. The rails and screws are in the final location though as this is dictated by the rails. After reading this (http://www.mycncuk.com/forums/router-build-logs/5144-cncdudez-muscle-small-foot-print-machine.html) thread I'm toying with the idea of designing in the ability to use liquid coolant when cutting aluminium, I suspect that might be biting off more than I can chew at this stage though. I think I'll keep it in mind as a potential future improvement.

The rails, fingers crossed, should be with me fairly soon. Parcel Farce reports that they are in the country awaiting customs clearance. So far my first purchase from the far east seems to be going smoothly. Phew.

I got to thinking over the weekend about whether I would manage to finish this project (actually when because I'm going to finish it if it kills me) and it occurred to me that a lot of projects don't seem to make it past the planning stage and a good few seem to fail in the building stage. I wonder what percentage of people who want to build a CNC actually finish the task and end up with a machine that works (at least reasonably well).

JAZZCNC
27-08-2013, 07:35 PM
The Z-axis stepper has grown a bent metal box cover. I'll probably end up adding a hinged flap on the front at some point - there was a front originally but I realized would wouldn't be able to fit the stepper through the hole I'd left!

I would make the Z front plate just bit narrower than Rear plate and put a full cover over the z axis so it protects rails and everything.
Easy done with 1.5mm Ali sheet, just make a cardboard template and then cut around. Don't need fancy bender or saws just score profile edges with stanley knife, trim with tin snips into straight length then bend to break off.
Easy folded with simple scores on fold lines (back side.!!). The pics below show cover which was folded by hand without clamping to make a permanent template then unfolded, (see pic.). . . If you clamp the sheet on top surface before folding and fold on edge of board then makes a better fold.

Takes about 30mins to make.!!

Wobblycogs
31-08-2013, 08:36 PM
That looks more like a tank than a cnc machine :-). I bet it effective at keeping the chips out though.

Well, I've been hard at it putting bolts in the design and I think I've got to the point were there's not much point adding any more as they clearly wont clash. So with that done here are some shots of what will get built unless anyone comes up with a show stopper. my estimate for maximum cutting sizes are X: 840mm, Y: 840mm and Z: 125mm assuming a 50mm bit. I cab squeeze a bit more out of Z by moving the spindle up a series of holes.

992299239924992599269927992899299930

CraftyGeek
01-09-2013, 06:56 AM
Is the plan still to mount it vertically?...how is it going to be attached to the wall?

Wobblycogs
01-09-2013, 07:26 AM
The original plan was to go vertical but the design has grown quite a bit since then to the point where it's not practical to vertically mount it, at least not in my current workshop anyway. When I move to a bigger workshop I might look into mounting it vertically again but to be honest I think what will probably happen is I build a steel base for it and use underneath the machine for storage.

CraftyGeek
01-09-2013, 09:01 AM
Ah, ok - my next question was going to be how will you level it if its going to be horizontal...but I guess you'd put adjustable feet on the steel frame in that case?

Also - I got my wrist firmly slapped by Jazz for my design with the stepper motor mounted on the Z axis front plate...he said it should be on the Y axis plate instead to help keep it out of the way & reduce the weight on the z axis.

Wobblycogs
01-09-2013, 10:14 AM
For now I'm going to put feet on what you see as the machine will either be on the floor or a bench somewhere. Once it's in it's final home on a proper frame there will be "something better".

I agree having the stepper on the front of the Z-axis isn't ideal but I don't see it being enough of a problem for me. I moved it from the back plate to the front plate because I had to get power and water to the spindle anyway. Having the stepper on the back plate made getting the cables over more difficult so the front plate seemed like a better choice. As I see it there are a couple of other advantages as well, it's allowed me to lift the z-axis screw up away from the chips a bit and the weight of the stepper will act as a pre-load on that axis.

CraftyGeek
01-09-2013, 10:18 AM
I thought you'd have your reasons...thought I ought to mention it though ;-)

Looking forward to seeing the actual build!

JAZZCNC
01-09-2013, 09:37 PM
Also - I got my wrist firmly slapped by Jazz for my design with the stepper motor mounted on the Z axis front plate...he said it should be on the Y axis plate instead to help keep it out of the way & reduce the weight on the z axis.

Nah nah not having that Geeky boy I only suggest or recommend if you choose to do it different then I don't give a toss, it's your loss. . Lol


I agree having the stepper on the front of the Z-axis isn't ideal but I don't see it being enough of a problem for me.

It's not a problem(my machines like that) just from experience the other way round is better, allows higher accelerations and Velocity and reduces chances of missed steps on Z axis when working hard doing 3D work.


As I see it there are a couple of other advantages as well, it's allowed me to lift the z-axis screw up away from the chips a bit and the weight of the stepper will act as a pre-load on that axis.

No advantages here in real use only Disadvantages.?
Chips will get in there no matter unless you cover, hence why I fully cover, May not be so pretty but it's the only way to keep chips out properly and increases component life too.
The extra weight of Ballscrew, 2 x bearing mounts, stepper motor, Stepper mount, belts & pulleys makes a considerable difference to the acceleration you can achieve and extra weight will cause missed steps if motors pushed remotely close to it's corner speed.

The extra weight acting has preload is hardly worth the costs in performance and extra stress on motor/drive.

Again not saying it won't work because it does, just works much better other way.!!

Wobblycogs
02-09-2013, 02:05 PM
Thanks Dean. Given what you've just said I'll stick with the stepper on the front of the Z and see how it goes. It wouldn't be the end of the world to move it to the back to get better performance if it's needed.

I've started pricing the parts up, it's looking like the build will come in around the £2000 mark. I'll put the pricing spreadsheet up somewhere once I've finished filling it in as it might be useful to people trying to decide if they can afford a build.

JAZZCNC
02-09-2013, 02:53 PM
Thanks Dean. Given what you've just said I'll stick with the stepper on the front of the Z and see how it goes.

Ok but I will say " I TOLD YOU SO" when you start asking why Z axis is doing strange things.!!!. . . . . (Oh Did I forget to say My Z axis done like this uses Nema 34 motor.!!)

Wobblycogs
02-09-2013, 03:10 PM
LOL, perhaps I'll have a bit of rethink about that then. I had hoped to save a bit of metal on the Z-axis by moving the stepper to the front but I ended up with basically the same amount, the price difference was only a couple of pounds so it wasn't really worth it - and it took an age to re-align everything and put the bolts back in :-). I've come to the conclusion that SolidWorks + Smart Fastners + Flexible Assemblies is not a good combination.

Wobblycogs
02-09-2013, 10:28 PM
Well I couldn't leave it alone :-)

One quick re-design of the z-axis stepper placement and I can't believe I didn't just go with this all along. Trying to save a few pounds on the z-axis I think was a false economy. This design has the ease of cabling of the previous but with lower z-axis weight as well. Double win, now I've just got to plug all the numbers for the z-axis parts into the spreadsheet again...

99529953

JAZZCNC
02-09-2013, 10:38 PM
Great. . :yahoo:. . . . Now put a proper bloody cover over the Z axis and give those bearings and screws a break. . .Lol

JAZZCNC
02-09-2013, 10:51 PM
What width is your Z axis.? If you can make use of it I have a Z axis motor mount I made too narrow. It looks like pic below but with larger centre slot so 20T pulley goes thru. . . . It's 160mm wide and 19mm plate made for Nema 23/24 (60mm) motor.

Just cover postage and it's yours.

CraftyGeek
03-09-2013, 06:04 AM
Well I couldn't leave it alone :-)

One quick re-design of the z-axis stepper placement and I can't believe I didn't just go with this all along. Trying to save a few pounds on the z-axis I think was a false economy. This design has the ease of cabling of the previous but with lower z-axis weight as well.

99529953

I had a similar thought when I moved mine around as well :-p

Wobblycogs
03-09-2013, 07:03 AM
Now put a proper bloody cover over the Z axis and give those bearings and screws a break.

LOL, I've got to leave myself something to regret not designing in from the start!

That stepper motor mount would be really useful, thanks (my back plate is 190mm wide so it would fit perfectly). The stepper mounting plates are one of the bits I wasn't looking forward to making by hand.

Wobblycogs
03-09-2013, 04:07 PM
Possibly this is a silly thought but is it worth trying to getting the X and Y axis ball screws machined so that they can be used with a fixed BK bearing on both ends? The price difference of the bearings doesn't seem to be much and it would increase the critical speed nicely. The only problem I can see is that I'd have to be quite accurate with the positioning of bearings.

Any recommendations on stepper motors? I'm currently thinking the 4Nm steppers (http://www.cnc4you.co.uk/index.php?route=product/product&path=83_85&product_id=70) from CNC4YOU look quite nice.

JAZZCNC
03-09-2013, 04:30 PM
Possibly this is a silly thought but is it worth trying to getting the X and Y axis ball screws machined so that they can be used with a fixed BK bearing on both ends?

Not a silly thought and won't make it worse that's for sure but hardly worth the hassle has you'll have no problem with critical speed or whip at this length screw using steppers.

Regards the 4Nm then I've found the 3Nm motors spin faster and see very little to any difference in torque.!

Wobblycogs
10-09-2013, 07:38 PM
I've been doing some reading about the electronics side of the project recently hence the lack of updates. I'm currently favouring a Mach based set up using an Ethernet SmoothStepper and the PMDX-126 board. A fairly expensive set up admittedly but one that I hope will be fairly trouble free. I admit I'm tempted to try out LinuxCNC first as I have a pile of old computers hanging around and a fair bit of experience with Linux (I run half a dozen Linux servers) but as an old friend once said to me "Linux is only free if your time is worth nothing" and I've found that to almost always be true.

My rails finally arrived from South Korea (48 hours to get from South Korea to Cardiff, 3 weeks nearly to get from Cardiff to Gloucester!). I got caught for import duty but not a lot so can't really grumble and they were still much cheaper than locally sourced stuff. I've given them a little slide about and everything seems fine. If I press down on the block it slides more easily, I'm guessing this is pre-loading? Is there anything in particular I should look for to check if they are ok?

10054

njhussey
11-09-2013, 10:16 AM
I'm just about to press the button on the ballscrews and round rails so would be interested to hear where you got your linear rails from as I might bite the bullet, pay the extra, and get these from the outset. also I'm going to get a PDMX-126 BOB so give me a shout and I'll order one along with you and perhaps we can save some carriage by getting 2?

I keep meaning to have a play with Linux but I've been out of playing with PC's for years now so probably know enough to get me into trouble but not out of it!!

Wobblycogs
11-09-2013, 10:28 AM
The rails came from eBay a seller called fa-system (http://myworld.ebay.co.uk/fa-system/). All told they cost £484, the same kit from Zapp would be £683 (not sure if that Zapp price includes postage). As far as I can tell the rails are still in tip top condition. The Z-axis have clearly seen a harder life than the X and Y which look almost brand new, I'd guess they all still have a ton of life left in them though.

I like the idea of buying the BOB's together, I'm pretty sure that's the model I'll end up going with but I want to have a bit more of a read first. I'd guess I would be looking to order late next week if that's ok?

Have you had any thoughts about spindles? There's an eBay seller "happykisssoul" that's shipping 2.2kW from the UK for a good price (even including insane shipping costs). I've seen a couple of guys say they have bought from him without problems.

njhussey
11-09-2013, 11:04 AM
End of next week is good with me for the BOB. I've looked at that seller and even bookmarked some of the rails but they would be miss matched sets and the NSK S20 rails that are 1390mm long look like they have a bit of damage to one rail so not 100% happy about getting it. Might just have a look about and see what else is available.

I've made my own spindle which should put out 1.7kW and I've got a 25V 47A PSU to power it....

10060

10059

Wobblycogs
11-09-2013, 11:12 AM
I think you are right not to touch those NSK rails, the damage is way to close to the bearings for my liking. There are some HiWin and THK clones over on AliExpress that looked ok and the reviews I've found are generally positive.

Does the power supply really have that written on the top? I can't quite tell if it's been "photoshopped" on. You certainly wouldn't mis-read the spec.

njhussey
11-09-2013, 11:16 AM
I got it made from a chap on the heli forum I'm a member of, it was going to power my LIPO battery charger but I've got a bigger supply for that now :biggrin: The writing is vinyl stickers so could be peeled off if so wished...

njhussey
11-09-2013, 12:21 PM
Just looked at the rails again and the damage is right at the end of the rail, as it's 1390mm long and I only want 1300mm I could cut that off....

Wobblycogs
18-09-2013, 12:25 PM
I've been having a look at the control / motion side of the machine for the last week or so hence the lack of updates. I think I've now settled on a fairly uncontroversial set of components but I'd be grateful if someone more experienced could have a quick look over the shopping list...


Steppers - 4 * Nema23 3.1Nm, CNC4YOU (http://www.cnc4you.co.uk/index.php?route=product/product&path=83_84&product_id=65)
Drivers - 4* AM882, Zapp (http://www.zappautomation.co.uk/en/digital-stepper-drivers/304-am882.html) - Is there a reliable cheaper source?
BOB - PMDX-126 Rev C (http://www.pmdx.com/PMDX-126) - I plan on using the Rev B slave feature for the X-axis.
Spindle Speed Control - PMDX-107 (http://www.pmdx.com/PMDX-107) - Unlikely I'll set this up straight away but might as well get it now
Step Generator - Ethernet SmoothStepper (http://www.warp9td.com/)
Software - Mach3 (http://www.machsupport.com/) - Might wait till Mach4 as it looks like there won't be an upgrade option.
Stepper Cable - 4 Core CY Cable 1.5mm^2 (http://www.csecables.com/acatalog/CY-Cable-4-Core-1-5mm2.html)
Spindle - Spindle 2.2kW with VFD (http://www.ebay.co.uk/itm/WATER-COOLE-MOTOR-SPINDLE-2-2KW-VFD-MATCHING-INVERTER-ENGRAVING-GRINDING-l7-/190752376454)
Computer - One of several spares kicking around the house, running Win XP probably.
Power Supply - Currently planning a self build...


Total cost: :sorrow:

I plan on adding limit switches, water cooling pump, radiator, etc but I think this will get me started. I could do with a recommendation on cable for the spindle though :smile:

Most of the machine will be bolted together with M4 cap head screws, will this be strong enough? I settled on M4 rather than M5 to get additional threads in to the aluminium box section. My gut feel is that with the number of bolts I'm using either would be fine but I don't want to go buying a million M4's only to find they are the wrong size :biggrin:

njhussey
18-09-2013, 12:54 PM
Snap to most of that apart from the slaved x axis...oh and the spindle.

I saw this Aluminium 120mm Radiator for PC or CNC Cooling. G1/4" Fitting | eBay (http://www.ebay.co.uk/itm/Aluminium-120mm-Radiator-for-PC-or-CNC-Cooling-G1-4-Fitting-/300959741115?pt=UK_Computing_Water_Cooling&hash=item46129934bb) cute little radiator on eBay and was thinking of getting it as I might put a water cooling mount on my machine for my spindle to cool the bearing housing down as I can see it getting quite hot. I've got a spare 120mm 24V fan to mount to it and would "borrow" some tubing and fittings from work to connect it all up.

Wobblycogs
18-09-2013, 12:59 PM
Yeah, that's the sort of thing I was thinking off. I'm going to mount the radiator, fan and a small pump on the gantry somewhere. I've got a load of computer bits and some fish keeping bits so I might just be able to bodge something together from bits I can find around the house. I've read a few places that the water cooled spindles can run for quite a while without cooling so I think that'll get me started.

Jonathan
18-09-2013, 02:47 PM
Drivers are, as usual, cheaper from aliexpress:
New Leadshine AM882 Stepper Drive Stepping Motor Driver 80V 8.2A with Sensorless Detection #SM414 @SD-in Motor Driver from Industry & Business on Aliexpress.com (http://www.aliexpress.com/item/New-Leadshine-AM882-Stepper-Drive-Stepping-Motor-Driver-80V-8-2A-with-Sensorless-Detection-SM414-SD/570168083.html)

4 seems a bit excessive to me. You can maybe justify two for the slaved axis, but I doubt you'll see much benefit using them compared to DQ860MA on the other axes.
BOB/Spindle control/ESS/Mach3... well if you must.

The cable you've chosen for the stepper motors is fine and also good for the spindle, or you could instead use 4-core 1mm^2 for the spindle. I'd stick with the 1.5mm^2 for everything as it just makes estimating the length required easier.

njhussey
18-09-2013, 04:32 PM
For the sake of saving approx £20 per drive I think I'll probably just go for all AM882's, in the grand scheme of things I don't think it will make that much difference to the overall cost as I've now thrown caution to the wind and given up on my "cheap" route....

Wobblycogs
18-09-2013, 05:26 PM
I'm with you on this Neil. My spreadsheet currently shows it's going to cost in the region of £3200 although I think there are some savings to be made on materials. A £40 (maybe £80) saving would be nice and I'm thankful for the suggestion but at the same time I have to weigh that up against the fact I'm very time poor and spending a little extra to walk a well trodden path feels like a reasonable investment.

It's basically the same reason I've chosen to go the Mach3/ESS route. I'm sure I could get LinuxCNC to work with a cheap BOB but the days of me being able to spend 40 hours a week in the shop are, sadly, a distant memory. I'm also interested in trying to write applications to directly control the machine and doing that via the ESS looks like it would be significantly easier than trying to do it through the parallel port and dealing with the exacting timing issues (this could be very naive of me though).

Neil, would you'd be interested in a combined purchase of the AM882's Jonathan linked to?

JAZZCNC
18-09-2013, 05:58 PM
Hang fire with the ESS/PMDX?spindle control because I've just received the Csmio-IP-M Today form Zapp and so far it's looking like a fantastic bit of kit that I feel is going to change my view on ESS/PMDX preference.!! . . Esp has I can get from Gary @ Zapp without all the crap of buyng from US.

It's very very neat and you can see it's been design with industry in mind has it's all DIN rail mounted with good quality terminal connectors and it uses 24v inputs/outputs.
Comes with spindle control built in and it's giving a super stable accurate RPM straight from the box, no messing with de-bounce or any thing like that like you can have with cheap spindle boards. Two Relays on board for controlling For/Rev.

Loads of Digital Input's/outputs all 24V signals.

So far I've only hooked it up to some AM882 and 3Nm motors on the bench but I can tell from way motors sound and the speed and smoothness I'm getting from them it's giving super smooth pulses. You can always tell nice clean pulse's when the motors are turning slow or spinning very fast and these sound sweet.!!

I'll be throwing all this lot on a machine shortly but before that I'm going to hook it up to my machine so can give it a good thrashing and compare it to ESS and the Russian PLCM-E3P(which it's currently running) also the old USB-SS. Will probably throw a thread up when I've played more, Rest assured I'll do my best break it.! . Lol

Wobblycogs
18-09-2013, 06:10 PM
I saw you mention you'd got a Csmio on another thread. I have to admit it had me interested but I couldn't find much else about it so went with the standard setup.

It certainly sounds like a top quality solution. If it's as good as you think and comes in at roughly the same sort of price I'll be more than happy to switch over, I await your review with bated breath :)

JAZZCNC
18-09-2013, 06:20 PM
It certainly sounds like a top quality solution. If it's as good as you think and comes in at roughly the same sort of price I'll be more than happy to switch over, I await your review with bated breath :)

It's very high quality and looking brilliant so far. It will come in much cheaper than ESS/PMDX combo has you won't need the Pmdx126 or 107.

njhussey
18-09-2013, 06:21 PM
I'm all ears.......

Clive S
18-09-2013, 06:27 PM
Dean Will it slave the x motor. I thought only the S version did. But could be wrong. ..Clive

JAZZCNC
18-09-2013, 06:46 PM
Dean Will it slave the x motor. I thought only the S version did. But could be wrong. ..Clive

Not got that far yet Clive but I think it does but just not within the unit with advanced features like -S model has and just leaves it down to Mach to control in normal way. Not sure how it will handle Homing etc but I'll try to test it out when get time.

For my use on intended machine then it's not required so I'm not concerned about that and has you know I try to avoid slaving anyway.

Wobblycogs
19-09-2013, 01:21 PM
I'm must be missing some understanding because I've never seen anyone set up a machine in the way I'm about to suggest but I can't immediately see anything wrong with it...

Imagine a machine with dual x-axis steppers, why can't you just wire both stepper driver control lines in parallel to the x-axis connector on the BOB?

I can see that the signal strength (current) may be a problem as it will be halved but that doesn't feel like it would be insurmountable. If the machine is set up correctly wiring in parallel looks like the same end result as, for example, the slave feature of the PMDX-126.

Am I right in thinking that the step line indicates that the stepper should take a step by going high and then the direction line is high or low depending on the direction of rotation? Is there any other signalling taking place along these lines?

Jonathan
19-09-2013, 01:30 PM
You could do it that way, but in Mach3 people like to home each X-axis motor separately to square the gantry, which clearly can't be done with them both connected to the same output. I'm not keen on that method as it bends the gantry, albeit only slightly, every time you home the axis.


Am I right in thinking that the step line indicates that the stepper should take a step by going high and then the direction line is high or low depending on the direction of rotation?

Yes, that's correct although it can be inverted.

Robin Hewitt
19-09-2013, 01:41 PM
Is there any other signalling taking place along these lines?

What pattern does the driver put on it's outputs when you switch on?

It may go to a set start position, it may go to the position it was at when you switched off.

You have to figure out if this can blow your start location.

Wobblycogs
19-09-2013, 01:42 PM
Thanks Jonathan.

I can see that you couldn't home each motor separately with a parallel connection but I'm a little surprised that you would need or want to (at least after the initial machine set-up). Surely the two motors should be taking identical steps and therefore remain square on it's own?

Obviously no two steppers will be perfectly identical but the difference in step size between two motors must be tiny for the dual motor set up to work at all. Even if the two steppers did have a non-trivial difference in step size surely the worst case would be wherever is furthest from the home position getting worse in a linear fashion the further from the home position.

Wobblycogs
19-09-2013, 01:47 PM
What pattern does the driver put on it's outputs when you switch on?

It may go to a set start position, it may go to the position it was at when you switched off.

You have to figure out if this can blow your start location.

This is all hypothetical at the moment. I've been reading the manual for the CSMIO/IP-M which doesn't support slaving and I was wondering why I've not seen people just wiring drivers in parallel. I came to the conclusion there must be a good reason for it or BOB makers wouldn't advertise support for slaved drives as a selling point.

JAZZCNC
19-09-2013, 04:24 PM
Think it's mostly to do with timing issues and resonance handling. The BOB's that handle slaving do some internal trickery and split/buffer and boost the signals so they arrive at the same time with enough juice.!

Slaved motors can remain sync'd but they need to be tuned correctly and run well below there limit. If you run near the limit and have half decent weight gantry and cutting hard or fast then it's easy to start dropping steps over the job period.

Most troubles with missed steps come from over tuning and running too fast losing the steps from gantry pushing on de-acceleration or sharp directional changes.
This happens more than you realise when cutting materials that need higher feeds like woods and plastics and jobs which take several hours.
Now over the course of a job these lost steps don't often add up enough to cause trouble with woods etc because of the lower resolution and tolerances required but if not homed regularly then they would accumulate and cause troubles so home switches are needed and because you can't instantly see which motor lost what with out using dial gauge etc so then easiest solution is use Home switches and square back up.

Like Jonathan I'm not a fan of twisting gantry back into square but my solution is to not bother with slaved motors and connect screws with belts so Eliminating these hassles.!!

Run the motors within there comfort zone when at high feeds and slaved screws work fine.. . .Push it and you'll encounter troubles guaranteed.!!

Jonathan
19-09-2013, 04:46 PM
Occaisionally, I check the squareness of my machine by drilling 3 holes to prescribe a right angled triangle and accurately measure their center distance to determine skewness, and proceed to compensate back to square. I've taken to putting marks on the rotating ballnuts to aid the correction process.

I've never seen the machine loose or gain even just one microstep on either motor, except when I've crashed it. That's including on a job where the machine was left running for the best part of 47 hours.

Wobblycogs
19-09-2013, 06:01 PM
Thanks both.

So, I really like the look of the CSMIO/IP-M but the lack of built in support for slaving makes me nervous. From the manual it looks like it could be done via Mach3 but it's not at all clear what the consequences of doing this would be e.g. would homing / squaring work correctly.

I would happily just ditch one of the x-axis steppers (less hassle, cheaper, etc) but I can't see a way to combine an open back to the machine and a belt drive system without ending up with belts all over the place or the set up being impossible to tension. Any ideas would be greatly appreciated.

1017710178

I should say I'm not totally fixated on having an open back to the machine, it's nice to have and may make it more versatile but it's not the end of the world if it goes.

Jonathan
19-09-2013, 06:05 PM
If the output from the breakout board can't source enough current for two driver inputs, then you can buffer it with just one transistor and two resistors and that's guaranteed to work.

JAZZCNC
19-09-2013, 06:36 PM
Occaisionally, I check the squareness of my machine by drilling 3 holes to prescribe a right angled triangle and accurately measure their center distance to determine skewness, and proceed to compensate back to square. I've taken to putting marks on the rotating ballnuts to aid the correction process.

Ok but why would you need marks on rotating nuts to aid correction if it never loses a step or position.?? Should be nothing to correct.!!



I've never seen the machine loose or gain even just one microstep on either motor, except when I've crashed it. That's including on a job where the machine was left running for the best part of 47 hours.

But what was you cutting and at what feeds.? . . Aluminium I presume.! . . . Try cutting wood with inertia of rotating screws not rotating nuts around 7-8mtr/min for 47hrs and check it.!!
Also your machine is setup correctly and presumably with plenty of torque left, I'm talking higher feeds and motors tuned close to edge which often happens causing lost steps and much sooner than 47hrs.!!

Again I'll say it " Tune the motors properly with plenty of safety margin and there's no problem with slaved motors" but still it can happen and that's not good and healthy for screws or machine if one motor stalls racking gantry at rapid speeds.!!!!. . . . With belts running out of sync is not even in the frame and stays set just like the day it was set regardless of crashing or motor tuning.!! . . . .And trust me I've crashed hard and plenty so it's well tested. . .Lol

Jonathan
19-09-2013, 07:57 PM
Ok but why would you need marks on rotating nuts to aid correction if it never loses a step or position.?? Should be nothing to correct.!!

The marks are there so I can check that it's square when I switch the motors on, as clearly they can move when turned off. I could do the same with switches, but a permanent marker is cheaper!


But what was you cutting and at what feeds.? . . Aluminium I presume.! . . . Try cutting wood with inertia of rotating screws not rotating nuts around 7-8mtr/min for 47hrs and check it.!!

Not aluminium. The feedrates were low and the acceleration set quite high. The result would be the same if I ran my machine for that amount of time with the maximum feedrates, because the motors are tuned properly. Yes, I could set the rapid feedrate slightly higher and loose the odd step here and there, but there's no reason to as I chose the correct size motors for the machine, so the safe feedrates are adequate.


Also your machine is setup correctly and presumably with plenty of torque left,I'm talking higher feeds and motors tuned close to edge

If that's the case then they should buy the right motors and set up the machine correctly in the first place, instead of linking the screws with a belt so it doesn't matter if there's a missed step.

How do you accurately set the machine square when a belt is linking the screws?

JAZZCNC
19-09-2013, 09:20 PM
If that's the case then they should buy the right motors and set up the machine correctly in the first place, instead of linking the screws with a belt so it doesn't matter if there's a missed step.

Yep agreed but that's not always the case and some don't have knowledge to setup correctly from beginning and damage can be done before they learn.!! Plus there's the other added benefits using belts give like piece of mind a stall isn't going to turn gantry into a Twizzler.!! . . . Anyway it's bit Linux vs Mach thing we are obviously going to have to agrees to disagree.!


How do you accurately set the machine square when a belt is linking the screws?

Loosen bolts on gantry etc and pulleys then turn ball screws . . . . Simplizz.!! . . . Bit fiddly and time consuming at first but one time deal so no problem.

Wobblycogs
21-09-2013, 10:42 PM
Well it's been a busy day for the credit card. The drivers, steppers, ballscrews and nuts, bearings and mounts, spindle and vfd have all been bought mostly from China. I think I might have accidentally bought a collet set as well - I asked for a spindle with a ER20 collet and I think he's read that as "collet set". Not a problem though as I was going to buy a set anyway, I hadn't realized Chai had started sell them. If anyone needs drag chain he seems to be doing some nice looking stuff at a reasonable price.

I've also been tempted over to the dark side, I installed LinuxCNC on one of the old machines I have kicking about last night and I have to say I like what I see. There were a couple of minor complications but over all it seems to be a nicely put together system. I ran the latency tests for a couple of hours with a fully stressed machine and achieved ~7,000ns max jitter so I think it should work quite well. The parallel port supports SPP, EPP, ECP and ECP + EPP modes. I've seen someone mention that one of EPP or ECP is better but I can't find the thread again.

I was looking closely at the electronics in Jonathans "sufficiently strong (http://www.mycncuk.com/forums/router-build-logs/6484-sufficiently-strong-machine.html)" build and I can't see a BOB anywhere in the control box. In fact it looks like the drivers are connected directly to the parallel port with the cable being wrapped in a bit of foil. Is that correct? I've not been able to find all that much mention of people running steppers direct from the parallel port, just the odd post here and there but I wouldn't mind giving it a go. As has been mentioned elsewhere I can always move onto using an ESS + BOB if necessary but I might as well give it a shot.

njhussey
21-09-2013, 11:08 PM
I ordered my ballscrews off Chai last week, I also bought a set of collets off him but got 2.5m of energy chain off fleabay...going to be like Christmas soon with all this equipment arriving. Best order some steel Monday and get welding...

Jonathan
21-09-2013, 11:08 PM
7000ns is exceptionally good latency, what motherboard etc are you using?


I was looking closely at the electronics in Jonathans "sufficiently strong (http://www.mycncuk.com/forums/router-build-logs/6484-sufficiently-strong-machine.html)" build and I can't see a BOB anywhere in the control box. In fact it looks like the drivers are connected directly to the parallel port with the cable being wrapped in a bit of foil. Is that correct? [...] but I might as well give it a shot.

I left Sasha to wire up the control box and he had problems with the breakoutboard, so for now the stepper drivers are indeed connected directly to the port. The output voltage of the port is sufficient so switch the driver inputs reliably and the driver inputs are opto-isolated, so there's no chance of damaging the port. To get isolated inputs it's easiest to just get a breakout board. In short, it works... but that doesn't mean it the best idea. So by all means give it a shot, but I wouldn't advise using this setup permanently.

Not sure why the foil is there, maybe he ran out of shielded cable but I thought I lent him a 100m reel!

Wobblycogs
22-09-2013, 08:54 AM
Yes, I was quite pleased with the latency. Looking at the machine spec I'd forgotten quite how good it was, massive overkill for a control box I know but it was sitting around not doing anything. In fact it's been switched off for so long the BIOS battery needs replacing. I'll replace that and re-run the latency test to confirm the result.

The machine is:

Motherboard: GA-K8NXP-SLI (http://www.gigabyte.com/products/product-page.aspx?pid=1859#ov)
Processor: Athlon 64 3000+ (not sure which revision, I think it's a 130nm ClawHammer as I bought it quite soon after they were released)
Memory: 3GB DDR400
Video Card: some old nVidia card, can look it up if you're interested.

Initially I was getting a latency of about 50,000ns but I noticed that was caused by a spike when I first played sound through the on-board sound card. Disabling the on-board sound in the BIOS gave me a 7000ns. After performing all the package updates the system required I re-ran the tests and got around 6500ns. I also tried the nVidia binary drivers to see if that would improve video performance - bad idea, whenever anything using OpenGL started it caused a 250,000ns spike in jitter!

I hear what you are saying regarding running direct from the parallel port. Would you also recommend the PMDX-126 as a nice BOB?

Wobblycogs
02-11-2013, 07:01 PM
I've been a bit busy with work for the last couple of months hence the lack of posts but the build has moved forward a bit. I've now got the steppers, drivers, screws, nuts, bob, nut brackets, spindle, VFD and a few other bits and bobs as well.

I met up with Neil the other night to swap parts from a joint buy and he kindly gave me a bit of 10mm plate to have a practice with. When I first saw it my heart sank - it looked distinctly too thin to build a machine from. I thought I'd give it a crack anyway and build one of the 8 brackets that I plan on using to support the X-axis and here's the results:
105711057210573

The angle from the base to the back is 89.7 deg which I'm fairly pleased with considering this is my first attempt at metal work. It's surprisingly strong but is it strong enough? The the parts we first roughly jigsawed (or hand sawn once the jigsaw blade gave up) and then cut to exact size on the mitre saw. The accuracy is ok but not exactly the perfection I had envisioned. I'll have to decide now whether to get some kind soul to mill the parts for me or to press on with trying to cut the pieces myself.

Wobblycogs
16-11-2013, 12:43 PM
After my bracket making endeavours the other week I've been thinking that it's probably better if I just get the majority of the pieces cut professionally. I think I probably could just about cut the pieces myself and end up with a working machine but I'd be old and grey by the time I finished (or I'd end up buying a milling machine).

A benefit of getting the parts made is I don't have to worry about cuts that would be difficult for me to make. With that in mind I've changed the design slightly. The parts in eye watering green have been upgraded from 10 to 15mm. The brace for the gantry is gone in favour of a single stronger side plate (there will be M5 bolts through the side plate into the front and back gantry plates and the base plate).

1077410775

I was thinking about maybe getting the parts water jet cut by a local company that could also supply the aluminium for the job. Unfortunately I have no idea how much water jet cutting costs, I'm guessing it's not cheap though. The other option would be to find some kind soul around these parts that would cut them for me.

JAZZCNC
16-11-2013, 03:17 PM
I was thinking about maybe getting the parts water jet cut by a local company that could also supply the aluminium for the job. Unfortunately I have no idea how much water jet cutting costs, I'm guessing it's not cheap though. The other option would be to find some kind soul around these parts that would cut them for me.

Only do this for non critical parts has water jetting can leave slight angle on the edges so any plates the mate to each other can end up at and angle.

Wobblycogs
21-11-2013, 05:07 PM
I've been reading up on the system electronics recently and I'm a bit confused about the current settings for the drivers. I'm using AM882 drivers and 60BYG301B steppers from CNC4YOU. The steppers will be wired in parallel and the stepper data sheet lists the current for that configuration as 4.2 Amps - all well and good so far.

The problem is that the manual (I have) for the AM882 under "Connections to 8-lead motors > Parallel Connections" it says:

...Multiply the per phase (or unipolar) current rating by 1.96, or the bipolar current rating but 1.4, to determine the peak output current.

That seems to imply that I should set the driver to 4.2 * 1.4 = 5.8A (dip switch settings SW1 = On, SW2 = On, SW3 = Off gives 5.5A or set exactly using ProTuner). That feels wrong to me though. I think the AM882 manual is assuming that the stepper data sheet will give the RMS current but it's not clear from the data sheet if that is the case.

What I think I should do is use ProTuner to set the peak current to 4.2A as the nearest dip switch setting is SW1 = On, SW2 = Off, SW3 = On which only gives 3.6A. Sound right?

EddyCurrent
21-11-2013, 08:37 PM
From your data sheet the unipolar current is 3A so 1.96*3=5.88
the bipolar current is 4A so 1.4*4.2=5.88
the nearest setting is 5.5A with switches 1,2,3 set to ON,ON,OFF, this setting also supplies 3.93A RMS according to the AM882 manual.

That's how I would set them anyway but seeing as my setup is very similar AM882 + 60BYGH401 I'll be interested to see other replies.

Wobblycogs
21-11-2013, 08:46 PM
Thanks Eddy, last thing I want to do is fry a stepper as soon as I switch it on.

The AM882 drivers can be connected to a PC for set up. This lets you set an exact current limit (0.1A resolution it looks like from the screenshots) amongst other things.

Jonathan
21-11-2013, 09:27 PM
The correct current is the current at which the motors don't get too hot, i.e. above 80°C. Generally the easy way to find this, as you've done, is to use the datasheet, but if you're not sure then you can just start with a lower current and work up.

EddyCurrent
21-11-2013, 11:40 PM
There's also this to set first.

"Auto configuration by SW4
Change SW4 two times in one second to identify the motor parameter after
power-up if it is the first time installation. Please note that the rotation
switch position must be 0 when using this function. "

Also to set the microstep resolution and current using Pro Tuner then switches 1,2,3,5,6,7 all need to be set to ON and switch 4 OFF

Wobblycogs
22-11-2013, 08:51 AM
Thanks Jonathan, that was basically the approach I was going to take if necessary.

Eddy, I don't know why but I was under the impression that the auto-detect would only work for Leadshine steppers. I suppose if it's measuring the steppers response to given input signals though it could work for any make of stepper. I'll give it a go to see what it comes up with, I think the auto-detect can be triggered from software as well which is probably the way I'd go.

Wobblycogs
04-01-2014, 08:25 PM
I had some time off over Christmas so got cracking with putting together the electrical cabinet. I've literally just achieved "first rotation" as it shall forever be known in our house.

11183

My main inspiration was Eddy's recent build but there are some of my own ideas in there too. The case is a 600x600 and I've left space for an additional driver if / when I add one. The power supply was proving difficult to fit in until I had the burst of inspiration to go vertical. The clear perspex cover (I call it the idiot guard) was a secondary addition after I got a first hand experience of quite how much power the capacitor bank holds:

11184

The capacitors were almost completely discharged when I probed the top to see what voltage they still held. One slip with the positive probe and a bright flash and bang later and I need a new probe.

The builds not quite finished yet but as you can see I've got a stepper attached and had it rotating using the test mode of the PMDX. Some how, despite measuring everything ten times I've still managed to make it a bit tight around the BOB. I must have been having an off day or something but fortunately I've got just enough space.

I'm going to look at building a soft start for the transformer. As I mentioned in another thread I'm getting some nuisance trips off a 6A Type D breaker. I might give a slow blow fuse a go if I remember to buy one next time I need some parts but I can't imagine it's good for the transformer to be getting whacked with 100A+ at start up. What I can't decide is whether to build a timer circuit to disconnect the soft start thermistor or to just leave it in all the time. The power supply is delivering a steady 71V, many thanks for the help with that so far guys :D.

Anyway, the one thing that surprised me about the set up was that the stepper wasn't silent when it was stationary. It makes a sort of kettle just going off the boil sound if that makes any sense. Is this normal? When it's turning it's quite quiet although I've only had it turning at one speed. I should probably say I've not fine tuned the current for the steppers yet, I just quickly selected a setting that was well below the maximum the steppers can handle.

EddyCurrent
04-01-2014, 08:45 PM
"First Rotation", sounds like a maiden flight.

Looking great, 600x600 seems a good size as there's plenty of room in there. Perspex cover is a good idea, it's always best to aim for IP2X with a panel.
I also found stepper a bit noise upon first fire up, used Pro Tuner to set the current and now very quite indeed.
I don't know what the requirement are for that inverter but mine specified 75mm free space both top and bottom for cooling, must be installed upright, no requirement for free space at the sides.

Clive S
04-01-2014, 08:52 PM
I don't know what the requirement are for that inverter but mine specified 75mm free space both top and bottom for cooling, must be installed upright, no requirement for free space at the sides.

Eddy Is that to stop the electric from falling out.:beer:

Wobblycogs
04-01-2014, 09:03 PM
Cheers Eddy. Once the case is finished it will be dust tight (I think your case is the same design). I figure anyone inside the case should have a pretty damn good idea of what they are doing or they shouldn't be in there! The cover over the power supply is stop stupid accidents, it wouldn't stop a concerted effort to electrocute yourself.

I don't remember seeing any spec's for space around the rectifier although I'm guessing there will have been some. There's a good bit of space around it though and the aluminium it's bolted too will act as a great heat sink. I'll keep an eye on it but I don't envisage a problem. I don't like the way the bleed resistors are connected at the moment, it feels like a bodge although it works fine. I was asking about the power supply over on another forum and someone pointed me in the direction of active rectifiers, apparently they dissipate almost no power - I bet they cost a fortune though.

Something else I meant to ask, how are you connecting the various bits inside your cabinet to the various bits on the outside e.g. steppers, e-stop / limits etc. I'd like to be able to just unplug everything and take the case away but I'm concerned plugs and sockets will introduce noise.

EddyCurrent
04-01-2014, 09:05 PM
Eddy Is that to stop the electric from falling out.:beer:

No that's what insulation is for :nevreness: also when I first read it was an 'inverter' I mounted it upside down :stupid:

EddyCurrent
04-01-2014, 09:14 PM
I figure anyone inside the case should have a pretty damn good idea of what they are doing or they shouldn't be in there!

You can't use that as an excuse though.


I don't remember seeing any spec's for space around the rectifier although I'm guessing there will have been some.

I meant the VFD, sorry I never usually call them VFD's


Something else I meant to ask, how are you connecting the various bits inside your cabinet to the various bits on the outside e.g. steppers, e-stop / limits etc. I'd like to be able to just unplug everything and take the case away but I'm concerned plugs and sockets will introduce noise.

I'm going straight into terminals via compression glands in the bottom gland-plate, I can't see me wanting to move it much but I'm prepared to disconnect if I do. From my experience, albeit on much larger VFD's, the plugs & sockets did not like the high frequencies coupled with the current, although they were sized for the job the insulation used to break down over time.

Washout
04-01-2014, 10:22 PM
That's a very nice box setup/layout.

Can I ask where you get your Din rail and connectors etc. from - I may have to look at something similar for my control box and also I have a 1970's horror of a lighting circuit junction box in my loft that needs replacing with something more elegant.

Regarding the VFD, the mounting orientation advice in the manual is likely because of the cooling fan, so its not blowing hot air over something else that might be susceptible to over heating (the number of times front to back and side to side cooling for servers and networking kit has been overlooked when buying kit for some data centres I have had to look into is scary).

Wobblycogs
04-01-2014, 11:02 PM
Almost all the DIN mounted stuff comes from Chalon Components. Their prices are about as good as I found anywhere. The enclosure is particularly nice. I think the only major item I didn't get from them was the main door locking switch.

Now that I realise Eddy meant the VFD I can see why he's mentioned it. There's not 70mm below but there's not much air restriction. There will be a hole above to let hot air out and it's at the top of the case specifically so it doesn't heat anything else.

Wobblycogs
15-03-2014, 09:17 PM
I've been working on the control case on and off for the last couple of months, I got a good few hours in today so I thought I'd share an update and a couple of pictures.

I got the soft start parts ordered, built and fitted. I'm really pleased with the out come. There's certainly no more tripping breakers / blowing fuses. I don't know if it's psychological but it actually sounds quieter when the main power is engaged. I've set the timer to two seconds to give the capacitors a chance to fully charge while the system is still being current limited. I also re-jigged the bleed resistors so that they dissipate a lot less power.

After that I set about making the top and bottom panel. My plan was to have all the connections coming out the bottom of the control case with just a couple of vents at the top. The whole case can then be mounted on a wall somewhere up out of the way. I spent an inordinate amount of time researching connectors and eventually settled on some from ebay (and AliExpress). I have to admit they weren't the cheapest connectors but they are a pleasure to work with and well over spec'ed.

As I (currently) have the VFD in the case I've gone for a 16A connection. I'm sure the total draw could be met by a 13A connector but the 16A stuff is just better suited to this sort of situation.

With all my focusing on the main connectors I completely forgot to leave space for the parallel port. Fortunately there was just enough space up the right hand side next to the fan. I actually think I would probably have put the connector there if I had planned it so all in all that's worked out well. The only slight bodge is that it's actually glued into place. The PMDX manual states that the metal outer case for the socket shouldn't be connected to the case (earth loop issues). I didn't have space for a plastic mounting plate so it was hot melt glue to the rescue.

118521185311854

GEOFFREY
15-03-2014, 11:45 PM
That looks really neat and professional. Well done. G.

Wobblycogs
16-03-2014, 06:40 PM
Thanks Geoffrey.

In answer to your question on Neils thread yes it was Lincoln with an l (I found the missing l on the floor). I was surprised how much I liked the place actually and I agree everyone was very friendly. We were up at the top of the hill just by the castle, the walk up and down the hill each day to get into town certainly helped burn off any excess calories!

GEOFFREY
16-03-2014, 09:14 PM
Glad you liked Lincoln, there are some great places in the UK, but I am off to sunny Tenerife for a week tomorrow (won't be burning off any calories though. I bet that there will be a week of wall to wall sunshine here and that it will probably end on the 24th when we return!!! G.

PAULRO
10-02-2015, 12:00 PM
4 seems a bit excessive to me. You can maybe justify two for the slaved axis, but I doubt you'll see much benefit using them compared to DQ860MA on the other axes.
BOB/Spindle control/ESS/Mach3... well if you must.

hi Jonathan, when you say 4 seems excessive to you what exactly are you referring to? if it's the drivers then i'm at a loss because i was under the impression that each motor needed a dedicated driver, or have i the wrong end of the stick? (again !!!)

Jonathan
10-02-2015, 12:21 PM
hi Jonathan, when you say 4 seems excessive to you what exactly are you referring to? if it's the drivers then i'm at a loss because i was under the impression that each motor needed a dedicated driver, or have i the wrong end of the stick? (again !!!)

I was referring to stepper drivers. You do need one per motor, but my point was you don't have to use the same driver for every motor. There are arguments for getting different drivers for slaved motors which don't apply to the other motors.

Clive S
10-02-2015, 12:21 PM
Paulro Have you posted in the correct place as this post is 9 months old!!! >Clive

edit Obviously not !!

Jonathan
10-02-2015, 12:23 PM
Paulro Have you posted in the correct place as this post is 9 months old!!! >Clive

edit Obviously not !!

He was referring to post #64 ... had to Google search the forum to find out!

PAULRO
10-02-2015, 01:37 PM
i'm still getting to grips with the forum!!!:stupid: i'll have to get one of my young fellas to bring me up to date with all the new forms of communicating:shame: just when i think , yeah, that's makes sense ! that's the way to go ... i read on and find one of you bright sparks has another alternative and that really gets the hamsters running in different directions

Wobblycogs
21-11-2023, 05:14 PM
Hello again :calm:. Seems a bit quite around these parts, where's everyone gone?

Anyway, it's been a while (cough, 10 years) but I've finally got around to starting to build this CNC. I started by building a sturdy bench for it to sit on. I was on safe ground here. The cross members are 2x9's and it's covered with a sheet of 18mm ply.
32021
32022

I made my first mistake while making one of the support brackets for the x-axis box section. I was rushing and I managed to break a drill bit. In the replacement hole I then managed to break a tap. :applouse:
32023

Moving swiftly on... Here's a shot of all the parts I've put together so far carefully arranged to make it look like I've done more than I have. I was going to mount the linear rail to the box section when it occurred to me that might not be a good idea. Obviously, I've got to get the x-axis exactly perpendicular to the y-axis. I feels like I should build the y-axis and mount it before drilling the mounting holes for the x-axis. Does that sound about right or should I mount the x-axis rails now and then try to adjust them square to the y-axis later? To adjustment them I'll have a tiny bit of play in the rail mounting screws or I can get more by opening up the holes that mount the x-axis rail assembly to the frame. It feels like I'll only get one shot at this so I don't want to screw it up.
32024

Neale
21-11-2023, 10:36 PM
Not quite sure what the problem is about squaring - it might not be as bad as you think!

On my own machine, I bolted down one rail to the frame as straight as I could. I then used the gantry to align the second rail to it, to guarantee parallelism - just enough slack in the mounting holes using M5 (from memory) bolts to allow the second rail to move slightly, then slide the gantry back and forth to get it parallel and tighten bolts. I built the gantry as accurately square as I could, and once the machine was almost finished I could take test cuts and then adjust the gantry to be as square as I could measure from the test pieces. How to square it? Again, in my case, I was driving the gantry at both ends so I could adjust the home position at each end of the gantry so that each time it homed, it automatically re-squared itself. I had arranged the gantry itself to be bolted to "feet" at each end which carried the ball slides and during the initial squaring, I could slacken the bolts slightly to allow the gantry to swivel slightly, then tightening once I was happy. The process does rely, obviously, on having a master and slave drive to the gantry, one at each side, and the motion controller needs to support separate homing of master and slave axes.

There is an underlying design principle in all this. For a home-built machine, where you do not have the machining and assembly facilities to build accuracy immediately into the machine, you need to build in wiggle room so that once built, you can adjust everything. "Adjust on assembly" is an expensive nightmare in a commercial setting so you pay a fortune for manufacturing capability to build in accuracy. As amateurs we don't have that luxury but our time costs a lot less so we do it the hard way! But it works - witness the number of successful machines described on this forum. Good luck!

Wobblycogs
22-11-2023, 12:32 AM
Thanks, Neale. I'll have another think about it. What you describe is pretty much what I'd planned on doing and then I was standing there yesterday looking at the parts and I started wondering if it was going to work. My concern in terrible ascii art is something like this:



y axis (in home position)
V
________
/ /
/ / <-- The x rails are parallel but not perpendicular to the y axis
/ /


Ah, now I've drawn it out I can see the method you describe will work. You just use the second x-axis stepper to pull the gantry into alignment.

Thanks again. I'm sure I'll have plenty of other stupid questions :-)

Neale
22-11-2023, 09:23 AM
I had a quick look back through the thread after I wrote my last post (thread started a long time ago!) and it seems that you are driving both gantry ballscrews by belt from a single stepper. In this case you could adjust gantry squaring by tweaking the belt position on one or other of the ballscrew pulleys a tooth or two at a time - I guess that that is what you were planning? I was able to just move a home switch trigger point slightly but at the cost of doubling up on steppers, drivers, switches and wiring, and general configuration complication!

Wobblycogs
22-11-2023, 09:57 AM
Cheers, I've just looked through the thread and noticed there's no good images of the back of the machine. There are two steppers on the x-axis, they are set down a little and then connected by short belts to the screws. It allowed the machine to be a little more compact.

This build has, indeed, taken a really long time for a variety of reasons, just over ten years at this point I think. I'll get there eventually.

I don't suppose you know a good source for ball screws do you? The ones I bought originally have seized. Somehow condensation must have formwed in the ball nut and they have rusted solid to the screw. I thought my workshop was dry and warm enough for that to not be a problem.

Neale
23-11-2023, 08:39 AM
No personal experience of buying ballscrews from anyone still in business, I'm afraid. BST Automation via Aliexpress is often recommended in this forum but I have never dealt with them myself.

phill05
24-11-2023, 12:08 PM
I use this company www.wmh-trans.co.uk I have had good ball screws and rails from them.

Phill