Advice on Mammoth CNC Router Build - Steel Framed - 2440mm x 1220mm cutting area!

[Wejjmeister]

So I've been planning to build this thing for quite some time now, finally I have taken the plunge and started to finalise the specs. I have read, re-read and read again many of the posts on here looking for advice and ideas, and I am now ready to lay out my designs in front of you guys for some no doubt frank but highly constructive criticism hehe!

My machine will be used for every type of material between balsa (model making) and Aluminium, and must be able to cut 1220mm x 2440mm sheets of the above materials at times. I will type out my specs a little later on as I'm limited on time at the moment, but I have put together the design in Sketchup for the general idea, and wondered before I go any further if you can give me ideas as to whether it will be strong enough or not?

110891109011091

[EddyCurrent]

It looks quite strong but you don't have any diagonal bracing on the ends. Also with a machine this size I would estimate the amount of material and calculate the weight, you will probably find it need to be made in bolt together sections, the bed being one section on it's own. The method of bolting may therefore have a small effect on your current design. It also looks like you have very little depth of cut in the Z direction.

These threads are worth reading.
http://www.mycncuk.com/forums/router-build-logs/6674-stell-frame-cnc-1500x1000-adjustable-table-your-opinion-please.html
http://www.mycncuk.com/forums/router-build-logs/6457-sturdy-fast-all-steel-cnc-my-first-build-3.html

[Wejjmeister]

Thanks pal, yeah I read both of those threads as part of the planning stage for this build. The Bed will be bolted together in the three sections shown. I will try to incorporate some diagonals too, thanks for that advice. As for the Z axis, what do you recommend? I did think about putting another length off 100x100 between the carriage and the Y axis, to lift it a little further but wasnt sure??

[EddyCurrent]

This is the side arrangement I went for, the linear rails sit on top of the upper blue beam, top of the bed will be flush with the top of the lower blue beam.
11102

The top rail bolts to the bottom section via those red flanges, this allows me to level up the top faces using shims or compressing epoxy putty in the joint.
I'm planning on 150 to 200mm for Z travel, you have to account for your longest and shortest cutters plus your workpiece thickness plus any sacrificial beds you might add in.

[Wejjmeister]

Thanks for the comments. I have made a few adjustments to the design based on comments made and also my own thoughts (dangerous!)

I have added a second 3m length of 100mm x 100mm box along the top, this gives me around 200mm on the Z axis. The plate of Aluminium on the bed will be 20mm thick and surfaced flat. I will set the rails along the 3m steels on Epoxy as in Jonathan's build so they are as true as possible, and the Bed will also be supported with uprights rather than diagonals, purely because the frame must bolt together and it is easier with uprights than diagonals to do so, without hopefully compromising the overall strength, I can add feet below the uprights if you deem it sensible too?

Let me know your thoughts guys as I would like to start on the frame soon. Ignore the frame for the Z at the moment, its in its very very early sketchup stages and will change A LOT from now...

Thanks in advance!


11191

[JAZZCNC]

I wouldn't use that method to build it.? I would drop the extra top rail and just extend the uprights to get your extra height.
It will save on material and resonance because those will be 2 large drums you'll be creating. To be honest I wouldn't use 100mm Box I'd go with 80mm and thick walled.
I'd use the design Eddy showed with top rails sat on plates which can be adjusted into plane much easier. The work involved with Epoxy is quite involved and to be honest a complete waste of effort if your going to use those Cheap round rails has there will be more error in there base than the surface your fastening to.! . . . . I also doubt you'll get it any better with epoxy than using the above method and it will take much longer and not be has flexible, it will also cost lot more money.

Show us the same picture but with out the bed base so can see the bracing etc.

[Boyan Silyavski]

On your last picture it seems you were thinking mainly of cutting aluminum :loyal: . If you continue with this design, take care that the rails on the gantry should be maximum low, near towards the table and as round rails don't like that position nor milling aluminum something has to change there, for starters -square rails.

[Wejjmeister]

Ok I will spend some time converting it all to 80mm box, what wall thickness would you recommend?

Those rails are only for an idea of where everything will go, in the end I will use Hi-Win rails all round as I have read a lot about there flexibility in use and strength, I just couldn't find a model of then in Sketchup so put the SBR's in just for reference. My main priority is getting the frame completed as I would like to do the build in phases.

If I extend the uprights and use the plate method to true it all, will five uprights be enough/too many? Or should I leave that as is?

Also, should I leave the bed in 100mm box as surely that will need as much strength as possible?

Thanks in advance guys appreciate the time taken to help!

[mekanik]

Hi M8
with such a large machine i think resonance could be an issue, i think Jazz commented on another build and suggested loosing the symetry of the supporting members, i am sure he will comment shortly.
regards
Mike

[Wejjmeister]

Hi M8
with such a large machine i think resonance could be an issue, i think Jazz commented on another build and suggested loosing the symetry of the supporting members, i am sure he will comment shortly.
regards
Mike

Hi Mek,

I considered this whilst planning it all, all boxed sections in the side of the frame would be filled with sand, and then the holes plugged afterwards, including the diagonal braces, would this be enough to damp the resonance?

[mekanik]

Hi
I am a new to this game and was just passing on information i have picked up from the forum, Jazz uses sand filled sections, for a definitive answer you might be better waiting for advice from our senior members that have actually built and used a machine(hint)
Regards
Mike

[Wejjmeister]

On your last picture it seems you were thinking mainly of cutting aluminum :loyal: . If you continue with this design, take care that the rails on the gantry should be maximum low, near towards the table and as round rails don't like that position nor milling aluminum something has to change there, for starters -square rails.

Would it be better to place the rails (HiWins) on the underside of the gantry then? I can incorporate this into my changes too if so? The issue then will be travel...hmm!!

[Lee Roberts]

Would it be better to place the rails (HiWins) on the underside of the gantry then? I can incorporate this into my changes too if so? The issue then will be travel...hmm!!

Hi Wejjmeister,

The ideal rail arrangement for these types of rail (supported round) is like what you see in the first pic below on the Y Axis (machine bed), so what your looking for is the base/bottom face of each rail to be facing each other. This dose make it harder when you come to true the rails to each other and get them on the same plane's but it is feasible.

This is true for any axis you use this type of rail for, so if you look at pic two below i'm selected in green the faces I think you should use to mount your rails. If you mount them on those faces, the other nock on effect is that you are also bringing the gantry closer to the bed.

.Me

[Jonathan]

The ideal rail arrangement for these types of rail (supported round) is like what you see in the first pic below on the Y Axis (machine bed), so what your looking for is the base/bottom face of each rail to be facing each other.

Except, as specified in post #8, he's not using supported round rails. Most Hiwin (or similar) type rails have equal load ratings in both directions, so just mount them in the way you think is easiest. How they are orientated in the current drawing is not a bad idea as you can use epoxy to level all the surfaces, but there are some disadvantages.

[Lee Roberts]

Except, as specified in post #8, he's not using supported round rails. Most Hiwin (or similar) type rails have equal load ratings in both directions, so just mount them in the way you think is easiest. How they are orientated in the current drawing is not a bad idea as you can use epoxy to level all the surfaces, but there are some disadvantages.

Bugger missed that post, o well its their for everyone else then lol, thanks Jonathan.

So, are you going to go as far as epoxy Wejj or just shim the rails when the time comes?

.Me

[Wejjmeister]

Epoxy I think although I will also be taking a leaf from Jazz and making it so the 3m lengths can be shimmed to give a better starting point, then on with the epoxy!! Funtimes!

So to clarify, I can mount HiWins as in the initial design i.e on top of the rails.

Jonathan, can I trouble you to expand on the disadvantages??

Thanks guys!

[Lee Roberts]

Epoxy I think although I will also be taking a leaf from Jazz and making it so the 3m lengths can be shimmed to give a better starting point, then on with the epoxy!! Funtimes!

So to clarify, I can mount HiWins as in the initial design i.e on top of the rails.

Jonathan, can I trouble you to expand on the disadvantages??

Thanks guys!

I don’t know if I’ve misunderstood Jazz here or you with regards to shimming...

Surly enough accuracy can be achieved when cutting and then wielding the frame that any small differences would be "levelled" out from doing an epoxy bed for the rails to then sit on?

To confirm, what it sounds like your suggesting is, having the ability to “shim” the 3m lengths, along their lengths.

.Me

[EddyCurrent]

Ok I will spend some time converting it all to 80mm box, what wall thickness would you recommend?

I would say 5mm, it's quite heavy but if you are attaching rails directly to it then you need some thickness for the screw threads. I used 3mm box so I'm having to glue a piece of 6mm flat strap inside to give it a 9mm thickness. The best way I've found to decide on material thickness for threads is to see what thickness a nut is for any particular size bolt and use that as the minimum.

[EddyCurrent]

I don’t know if I’ve misunderstood Jazz here or you with regards to shimming...

Surly enough accuracy can be achieved when cutting and then wielding the frame that any small differences would be "levelled" out from doing an epoxy bed for the rails to then sit on?

To confirm, what it sounds like your suggesting is, having the ability to “shim” the 3m lengths, along their lengths.

.Me

That's the method I'm also using and I found that after welding, one corner of the frame was 3mm out of level with the other. This might be regarded as too much depth for epoxy treatment alone. I don't know for sure ? but at least I can shim the rails to as near level as possible then use epoxy from there.

[Clive S]

That's the method I'm also using and I found that after welding, one corner of the frame was 3mm out of level with the other. This might be regarded as too much depth for epoxy treatment alone. I don't know for sure ? but at least I can shim the rails to as near level as possible then use epoxy from there.

Eddy You might have seen this before there is no problem with the thickness of the epoxy Jonathan and I have used it I think Dean uses some sort of epoxy putty which is stiffer, he has explained that on other threads.

I poured this to about 5mm deep it is very slow setting 10 - 20 hours and don't touch it for 48 hours. Wests system
..Clive

[Jonathan]

expand on the disadvantages??

If the rails are on the bottom surface then the bearings, i.e. your supporting points, are closer to the tool/bed. The overhang is therefore reduced, so the stiffness is correspondingly higher. It does however make mounting them on epoxy more challenging - although you could conceivably mount the Y rails and X bearing blocks on the same epoxy leveled surface and get very good accuracy. I'd try drawing both orientations and see which looks the most promising.


Surly enough accuracy can be achieved when cutting and then wielding the frame that any small differences would be "levelled" out from doing an epoxy bed for the rails to then sit on?

I tend to agree - there becomes a point when an extra couple of mm of epoxy doesn't cost much compared to the added material/time required to get the frame more accurate. The epoxy is pretty strong/hard and the damping properties may be useful, so I wouldn't be too concerned about minimizing the thickness, within reason. If you look in my build log the epoxy is quite thick, but don't treat that build log as gospel - there's plenty of room for improvement.

[Lee Roberts]

That's the method I'm also using and I found that after welding, one corner of the frame was 3mm out of level with the other. This might be regarded as too much depth for epoxy treatment alone. I don't know for sure ? but at least I can shim the rails to as near level as possible then use epoxy from there.

Thanks Eddy, any idea why this happened or?


Eddy You might have seen this before there is no problem with the thickness of the epoxy Jonathan and I have used it I think Dean uses some sort of epoxy putty which is stiffer, he has explained that on other threads.

I poured this to about 5mm deep it is very slow setting 10 - 20 hours and don't touch it for 48 hours. Wests system
..Clive

Hi Clive, in your picture to make your epoxy "gutters" you look to have used MDF, i dont know if you've seen Jonathan's recent machine but he made his using aluminum and then left those in place once the machine was ready. My question is, what are the edges like on the epoxy once it "sets", i.e dose it need protecting from damage and so on to keep it together.

Did you do anything to yours after the machine was ready to be used to keep it safe?


I tend to agree - there becomes a point when an extra couple of mm of epoxy doesn't cost much compared to the added material/time required to get the frame more accurate. The epoxy is pretty strong/hard and the damping properties may be useful, so I wouldn't be too concerned about minimizing the thickness, within reason. If you look in my build log the epoxy is quite thick, but don't treat that build log as gospel - there's plenty of room for improvement.

Thanks JB, I didnt see your post untill i posted mine.

.Me

[Clive S]

Hi Clive, in your picture to make your epoxy "gutters" you look to have used MDF, i dont know if you've seen Jonathan's recent machine but he made his using aluminum and then left those in place once the machine was ready. My question is, what are the edges like on the epoxy once it "sets", i.e dose it need protecting from damage and so on to keep it together.

Did you do anything to yours after the machine was ready to be used to keep it safe?
It is 10mm MDF the box section is 60x60 the MDF is placed right on the edge (just on the round part of the box) and then a very thin bead of silicone was run around the inside of the moat to stop any leaks. I also rubbed a little oil on my finger and ran that around to stop the epoxy sticking to the MDF.

The epoxy is like water and runs around ok. When it was cured the MDF was removed (the epoxy creaps up the inside of the MDF a little) Then I used a scraper and a flat file to get rid of the edges that were raised, it is easy to see when you are scraping it because it will start marking the centre.

The rails were then screwed right through the epoxy into the steel. Job done. The epoxy can be chamfered off at the edges as it can be very sharp like glass. ..Clive

[JAZZCNC]

I've used both methods in the past and each work good and each have there own +/-

Full Epoxy level is ok and works good but has you can see with clives pictures there's plenty of prepartion required and on large machine like what the OP wants this means a lot of work. This prep has to be very good and fully sealed other wise sag can happen and being 3mtr length the chances of this happening are plenty.
Yes it's not the end of the world and can be corrected if it happens but it's a ball ache when it happens.!! . . . . I'm sure Both clive and Jonathan will tell you when pouring and it starts leaking your running round like headless chicken trying to suppress it and again over 3mtr and between 80-100mm width that's a fair amount of epoxy your fighting with.! . . . . It's also not cheap.

Now with Epoxy putty feet and shims you don't have this problem has there's no liquid epoxy involved but you do still have to do careful measuring and setting up to get rails on same plane. Unlike The above epoxy method which takes care of this for you, thou if it's not perfect which you won't really findout until machine is built and working then it's a lot more work to correct.
The epoxy shimming is very easy to correct on the fly and while machine complete and working. You don't need to strip it down to correct.

Now to be honest I think the OP is correct to use Both methods has 3mtr is a long length of boxsection to ensure is straight and true, which it won't be, so would need some work even with Epoxy putty method and the Epoxy liquid method will make light work of this. The Epoxy putty and Shims will give him the fine tuning he will need over this length machine.!!

[Lee Roberts]

It is 10mm MDF the box section is 60x60 the MDF is placed right on the edge (just on the round part of the box) and then a very thin bead of silicone was run around the inside of the moat to stop any leaks. I also rubbed a little oil on my finger and ran that around to stop the epoxy sticking to the MDF.

The epoxy is like water and runs around ok. When it was cured the MDF was removed (the epoxy creaps up the inside of the MDF a little) Then I used a scraper and a flat file to get rid of the edges that were raised, it is easy to see when you are scraping it because it will start marking the centre.

The rails were then screwed right through the epoxy into the steel. Job done. The epoxy can be chamfered off at the edges as it can be very sharp like glass. ..Clive

Thanks Clive, I liked the tip on silicone, will add that to the mental archive for when the time comes. Good idea on the chamfering, i was thinking about what i could use to protect it once done but it sounds like its quite durable.

Clive, where is your build log?

.Me

[EddyCurrent]

That's the method I'm also using and I found that after welding, one corner of the frame was 3mm out of level with the other.


Thanks Eddy, any idea why this happened or?

Because I assumed my garage floor was level and I just built it from that datum.

[Wejjmeister]

1121911220

First stage of the MkII done, Im still working on it so bear with me. I decided just to use the epoxy on the tops of the rails as plates and shims plus epoxy will be a mare. It may be slightly better but I dont intend on moving it once set up so I think Epoxy should suffice.

The whole of the frame will be sand filled too, so not worried about resonance.

Next step is the rails and ballscrews and placements of. Will add more a little later hopefully, time permitting!

Thanks ever so much for all of the comments though guys, a real help!

[Boyan Silyavski]

The diagonals at the sides, it seems to me help nothing except add weight. Should work fine without them, but that's me :-)

[Jonathan]

Looks ok... but probably a bit OTT. Don't leave the ends of the long pieces of box section cantilevered - support them with a piece at 45°.


I'm sure Both clive and Jonathan will tell you when pouring and it starts leaking your running round like headless chicken trying to suppress it

I've not had any leaks at all. The guiding aluminium strips were firmly bolted to the frame, and the wooden pieces used covered with grease, which makes a good seal, and clamped with plenty of G-clamps. If it had leaked then I would have addressed the situation in a calm manner, using the glue gun and tape I had on hand just in case. The running round can be left to the three birds in the garden...

[ba99297]

Sorry for interrupting but I have an idea
Could it be possible to combine both methods (epoxy and shims) BUT instead of epoxy use clear coat for car painting. The clear coat will be the mean to show us which is the true level between the rails, so then we put shims and putty and fit the rails on that level. For that size of machine that may be 2-3 kgr of epoxy will be needed, the difference to the final cost will be significant After the putty pures we could even took off the clear coat as I don’t know how strong it is in getting pressure from the rails. In that way we don’t pay too much money for the epoxy as I think clear coat is much cheaper and we won’t waste too much time to true level the rails as we use the clear coat as guide. Regards the pure time of clear coat we could use a catalyst that will make pure time longer.
Also instead of clear coat we could use anything chaper that has self leveling properties
Just an idea I don’t know if it is correct

Vagelis

[EddyCurrent]

The diagonals at the sides, it seems to me help nothing except add weight. Should work fine without them, but that's me :-)

Even if there was 3 each side, one at each end and one in the middle, or even 2 each side, one at each end, that would be enough.

[EddyCurrent]

Also instead of clear coat we could use anything chaper that has self leveling properties

water ?

But the idea with epoxy is that it finds it's own level then hardens to that level and job done, with the method you suggest how would you measure the depth of clear coat (water) to know that the frame underneath was level ?

[ba99297]

water ?

But the idea with epoxy is that it finds it's own level then hardens to that level and job done, with the method you suggest how would you measure the depth of clear coat (water) to know that the frame underneath was level ?

What i suggest isn’t so different with the epoxy method. Clear coat is also a resign but I think much more cheaper that will harden when it cures. Water isnt hard unless it becomes ice. Water could give us a perfect level but the question is how we could “capture” that level. What about self levelling floor resigns ?

[EddyCurrent]

ba99297, Sorry, it sounds like I did not understand what you meant with the clear coat, I did not think there was enough strength in clear coat when it hardened.

[JAZZCNC]

I've not had any leaks at all. The guiding aluminium strips were firmly bolted to the frame, and the wooden pieces used covered with grease, which makes a good seal, and clamped with plenty of G-clamps. If it had leaked then I would have addressed the situation in a calm manner, using the glue gun and tape I had on hand just in case. The running round can be left to the three birds in the garden...

Oh I'm sure I'd read somewhere you had a Leak.??. . but obviously not so I stand corrected you obviously don't know.!! . . .But I assure you that you would run around trying to stop a leak if it happened no matter how COOL . . :hysterical:. . . you think you are.!!

[EddyCurrent]

Anything of use here ?
Epoxy Resins Shopping Pages (http://www.epoxy-info.co.uk/resins%20shop.htm?gclid=CLK81KfY77sCFfMPtAodsWgAoA #StructuralPouring)

[JAZZCNC]

What about self levelling floor resigns ?

The problem comes from the type of resin.? I think Car paints and self leveing Floor paints which are manily polyester resins contain high solvent levels which evaporate and can cause sinking etc.

[Wejjmeister]

The problem comes from the type of resin.? I think Car paints and self leveing Floor paints which are manily polyester resins contain high solvent levels which evaporate and can cause sinking etc.

They wont shrink or sink but they do not harden to the same levels as paints are meant to be somewhat flexible to resist cracking etc, so it wouldn't be suitable for this particular job as it would 'give' unlike the epoxy resins used in these forums before which cure fully.

[Wejjmeister]

Even if there was 3 each side, one at each end and one in the middle, or even 2 each side, one at each end, that would be enough.

Should I remove these then? Im all for saving material, especially considering the prices of steel these days!

[JAZZCNC]

They wont shrink or sink but they do not harden to the same levels as paints are meant to be somewhat flexible to resist cracking etc, so it wouldn't be suitable for this particular job as it would 'give' unlike the epoxy resins used in these forums before which cure fully.

I belive it does shrink and it's why Only Solvent free Epoxy resin is used for Epoxy Granite to avoid voids, sinking and cracks appearing when solvent evaporates.!!

Regards the steel and bracing then yes you could reduce quite a bit of that steel.

[Wejjmeister]

Next round guys Mk3!

11240
11238

Removed steel form sides and changed bed supports. Any better? Saves a lot of excess steel.

11239

Started on ideas for rails etc... let me know your thoughts please.

11241

11242

Think this should provide a strong setup, also thinking about using BK blocks to mount Ball screws? Will these be suitable?

Thanks!!

[Jonathan]

The covers round your ballscrews will fill up with swarf and cause the ballnut to fail prematurely.


Think this should provide a strong setup, also thinking about using BK blocks to mount Ball screws? Will these be suitable?

Best way is to use a rotating nut. If not, then at this length you'll want two angular contact bearings at both ends to get a respectable speed. Search on Google 'ballscrew critical speed' for more information.

[Wejjmeister]

Thanks for that, as for the former comment, I would have covers all along the open section to prevent swarf ingress, would this suffice?

[Jonathan]

Potentially - depends on the covers.

[Wejjmeister]

Potentially - depends on the covers.

Could always put the slot in the outer sides and make it so the gantry overhangs a little more and they attach on the side? Was trying to keep the rails and screws as close to each other as possible thats all.

[Jonathan]

Was trying to keep the rails and screws as close to each other as possible thats all.

That is a sensible priority.
I'm being intentionally vague here since, as you've realised, you've got to weigh up the factors and if I just say how I'd do it, then you might miss out on finding your own better solution. Ideally you need to minimise distance mentioned for rigidity, not allow swarf to build up on the ballscrew (especially if rotating nut) and not cost an unreasonable amount.

[Wejjmeister]

Exactly what I feared!! I will do some more thinking, need to research rotating nuts first as I'm not at all familiar with these, then I can hopefully come up with something. The covers are pretty resilient though and will stop most foreign bodies, with those plus regular cleaning I think it may be the best option as it gives the most strength.

I'll give it some thought! Any other glaringly obvious mistakes in the design??

[JAZZCNC]

Best way is to use a rotating nut. If not, then at this length you'll want two angular contact bearings at both ends to get a respectable speed. Search on Google 'ballscrew critical speed' for more information.

100% Agree I wouldn't use rotating screws at this length.


Thanks for that, as for the former comment, I would have covers all along the open section to prevent swarf ingress, would this suffice?



Could always put the slot in the outer sides and make it so the gantry overhangs a little more and they attach on the side? Was trying to keep the rails and screws as close to each other as possible thats all.

The covers will still fill up the way you have them now. The little extra the Gantry will extend to make it so slots are on outside won't make any differance to how this machine performs so it will be worth doing no matter if rotating screw or Nut.

Don't get caught up in the over analysing the machine because while important your only talking a very small differance which won't a make a JOT of differance to the cut quality or speed needed for cutting woods, plastics and even for the level of Aluminium cutting this machine will be capable of dealing with.

Shit in the Screws or Nut will make you REGRET NOT doing it EVERY DAY were has you'll never even notice the screw being closer to rails.!!

[JAZZCNC]

Any other glaringly obvious mistakes in the design??

Not glaring but you have gone too far with removing the diagnols but not far enough in other places.?

I would have diagnols at each end of the sides. But would remove the central upright and re-space others on all the Bed support frames.
Would also have a Bed support frame on each side upright and not laid flat in like you have in centre.

The rest looks ok.

EDIT: I would also on the outer End bed supports have diagnols on inside lower beam which fastens to the sides to help hold frame square while setting up and bed frame is fastened. Like this but larger and to suit your design.
1124311244

[Wejjmeister]

Will make the recommended changes tomorrow in sketchup, im getting square eyes now and RSI from that software, its good but frustrating!!

Also done some research on the rotating nut idea and like it a lot, need to make a few mods to allow it to work. Can you guys tell me which Blocks you use to hold the ballscrews in situ tho, are they FK blocks?

Also where do you get the actual rotating nuts from as I can only see the manufacturers, not any suppliers etc, Zapp dont seem to list them??

Any help would be very much appreciated!

Thanks again!!

[Jonathan]

Answer to both your questions is make your own (http://www.mycncuk.com/forums/linear-rotary-motion/3340-rotating-ballnut-design-ideas.html), or win the lottery so you can buy some.

[Wejjmeister]

Haha, thanks, just found this article

http://www.mycncuk.com/forums/showthread.php?t=3340

Awesome!!

[Clive S]

Also where do you get the actual rotating nuts from as I can only see the manufacturers, not any suppliers etc You could try Jonathan:wink: ..Clive

[Wejjmeister]

Read the entire thread last night and decided exactly that!

Jonathan, is there any possibility you could produce a pair of the mk3's from said thread? I would gladly pay you for these as I simply don't have access to the required tooling (currently have a large lathe, that's all)??

If so, do you have any sketchup diagrams so I can incorporate them into my design?

Many thanks.

[Boyan Silyavski]

The Sketch file i made is in Jonathans thread post #83 http://www.mycncuk.com/forums/linear-rotary-motion/3340-rotating-ballnut-design-ideas-9.html#post52272. so you can download it and incorporate it in your model. However i advise you to produce it, when you 100% decided on the exact motors/steppers, servo, size, etc. / and the machine drawing is finished, as maybe you will need to redesign the motor holding plate, in context of your design, or just leave that plate at last, however have in mind that if you decide to drive the gantry with 1 motor for example, this will affect the whole design, as the hole where the belt goes in should be at different place. You can start though with ordering the bearings, the rotating part +the lock nuts and pulleys, that's what i did

[Wejjmeister]

Bringing this one back to life guys. Life's been pretty crazy lately but I should have a little time for this now, I've sold a few of my projects and am trying to consolidate everything so that I have a little more focus.

Question for silyavski... Are the rotating nut diagrams proven, ie a replica of Jonathans mk3 or are they a custom design you came up with? Just asking as I want to know if they work wether I can send the file to the machinist to get them made or not.

Cheers!!

[Lee Roberts]

Bringing this one back to life guys. Life's been pretty crazy lately but I should have a little time for this now, I've sold a few of my projects and am trying to consolidate everything so that I have a little more focus.

Question for silyavski... Are the rotating nut diagrams proven, ie a replica of Jonathans mk3 or are they a custom design you came up with? Just asking as I want to know if they work wether I can send the file to the machinist to get them made or not.

Cheers!!

I think its his own design, the files maybe in the open source section, its been awhile now so I could be wrong.

.Me

[Wejjmeister]

I think its his own design, the files maybe in the open source section, its been awhile now so I could be wrong.

.Me

Thanks, shortly after posting that comment I found some file Jonathan had uploaded, so Im currently trying to integrate that into my design.

Here is the Mk4 frame layout, any opinions guys? Ive taken on board other designs in this one, Im making the bed adjustable too on holes inside the legs, so it will have further movement, will that affect anything else do you think?

Thanks in advance!

[Wejjmeister]

Is this any good chaps? Changes a fair amount to incorporate it...

124691246812470

[Wejjmeister]

Can anyone just gimme an idea as to wether Im heading in the right direction please, I dont want to go too far with any major flaws. Thanks.

[Neale]

How are you planning to allow fine adjustment of bearing/ballscrew nut position? This isn't a criticism of your design but I'm working on exactly the same issue on my own design! I'm assuming I can build a structure to within a millimetre or two, but will need to permit tweaking to get exact alignment afterwards. I can't see how you can move the bearings, for example, in both dimensions, unless you plan to use shims?

[Wejjmeister]

For the top rails, I will use the epoxy method, both on the x and y. For the ballscrews and blocks it will be shimmed to suit, hopefully that will be sufficient?! Im a newbie though too so looking for advice throughout the process just to keep me on track!

[gavztheouch]

Looking pretty nice. I would love to build something like this dedicated to profilling plywood. 8*4 sheets

Do you have a rough idea of the material and parts cost?

[Wejjmeister]

Thanks. So far the budget is around 5k but Im expecting it to be a little over as I am unable to machine some of the parts in house. Doh!! You could cut costs in some areas though as this is also designed for Aluminium sheets, if you just wanted to do ply, you could save a little I would think?

[gavztheouch]

Why do you need the bearing blocks to hold the ballscrews if you are using a rotating nut.

[Wejjmeister]

Why do you need the bearing blocks to hold the ballscrews if you are using a rotating nut.

I will fix them, they are just for reference to begin with to get the spacing etc correct. Couldnt find a skp file for anything else to be honest!

[Wejjmeister]

Can anyone just say yay or nay to the frame as I would like to make a start soon?

[JAZZCNC]

Can anyone just say yay or nay to the frame as I would like to make a start soon?

Frame looks fine to me and if anything it's over engineered in places and under in other. By this I mean you don't need that much diagonal bracing near the floor but it would benifit from more diagonal braces on the sides and ends/cross sections..
If you don't plan to put on feet then I would also lift the lower horizontal beams up a few inches to help with levelling. Just the stubs of the vertical members will be better and easier.

Other than that then it looks fine and will be plenty strong.

[Wejjmeister]

Frame looks fine to me and if anything it's over engineered in places and under in other. By this I mean you don't need that much diagonal bracing near the floor but it would benifit from more diagonal braces on the sides and ends/cross sections..
If you don't plan to put on feet then I would also lift the lower horizontal beams up a few inches to help with levelling. Just the stubs of the vertical members will be better and easier.

Other than that then it looks fine and will be plenty strong.

Thanks Jazz, I can add some bolt in diagonals on the top if necessary after the majority is built, there will be feet too on the bottom to level it all up so got that covered. Thanks for the green light though, very much appreciated!

[Wejjmeister]

Right so I have now officially ordered the Steel for the frame, I have gone with a local company, they've agreed a good price and to cut to a decent tolerance level so I'm happy. The bed Will be tweaked slightly as I've never been happy with it, but the frame is the same.

Questions for those in the know:

1. Does this look to be strong enough guys? The ball-screw will be driven by a pulley on the end so that the pitch can be played around with.
2. I'm not sure where to mount the motor though?
3. Will 20mm profiled rails be sufficient?
4. What kind of motor should I use, as I specced all Nema 34 12.1N's throughout so far but do I really need these on this axis?

1258912590

The gantry is a nightmare, I am trying to keep everything close together to keep it strong, whilst using the space well too and trying to save on machining costs etc, if you have any thoughts so far, please let me know as I am dreaming in CAD lately and my thoughts are all over the place!!!!

[JAZZCNC]

Ok well it's plenty strong enough but I would use 2 x ballscrews on the Gantry with this wide Y axis design to stop any racking affect but connect with timing belts using one motor.
I would use 20mm diameter and 20mm pitch screws at this length with a ratio applied. This will give more torque allow lower screw speed and still maintain decent rapid speeds.
Regards the motors then be aware that large nema 34 motors spin slower and require high power drives. So for this reason I'd go with 6Nm or no more than 8Nm motors as they will spinn slightly faster but still will require high power drives to get the best from them.

20mm linear rails will be more than enough.

With this wide gantry design you make it more stable/stiff but your making life hard regards the Z axis and you'll need to work on this area because that simple frame won't be enough. To be honest I'm not a Fan of this style of gantry design because it makes life harder than really needs to be and for not much more gain really. If your just cutting wood then you need to ask your self do you really need this complexity and my answer would be No.
Also to take advantage of the extra stiffness the gantry allows you need to have a powerful spindle so again not much point if your just using a plane old router or 2.2Kw spindle as a conventional gantry design built strong will be more than good enough.!!

[Wejjmeister]

Thanks Jazz. I will change the design to incorporate the twin screws with belts in my next drawing. I could change the motor to whatever you recommend though, it doesn't have to stay at a Nema 34?

The frame for the Z axis is just a starting point, the actual Z will mount to the box section but wont be relying solely on the box. It will be a solid construction all round and will cut every material possible. The main material I will be cutting will be 6082 aluminium so the stiffness will hopefully provide me with a great finish on the cutting surface, rather than the opportunity of mounting a monster spindle. i will stick to the 2.2Kw items most people use.

I know it may be overkill, but I think it will be great to have the best machine you can build and afford, rather than something you know you could have improved and done better on, that would just bug me.

If there are any other changes I need to make, please comment.

Thanks as always for the quick replies, really appreciate it!

[JAZZCNC]

Ok well lets see the rest of the design for Z axis and work from that.

My suggestion for nema 34 on 2020 screws was mostly for the long (X)axis to keep screw speed down to reduce whip and still maintain 10mm pitch speeds but I've just noticed it looks like your intending to use rotating ballnuts which if correct then you could use nema 23 motors on 10mm pitch.
That said 20mm pitch would allow a ratio to increase torque and have a lower nut speed for same speed as 10mm pitch which if using 20mm dia screws (which I would use at this length) may be better with the larger nut on smaller motors.

This doesn't help the Y axis thou which will still need nema34 motor if you use twin screws on single motor. That said Mixing motor sizes isn't a problem if you use High powered drives because they run from mains AC so don't have a mix matched DC power supply which would be needed for the lower voltage DC drives running the nema23 motors.

[Boyan Silyavski]

What you miss with this gantry design is that the rails must be as low as possible, at least one of them. Otherwise is pointless. Now you are just making a big lever.
As it is now is weaker than normal gantry.
The big+++ of this design is eliminating overhang and stiffening at the bearing blocks and lowering the assembly and eliminate gantry twist.
But as i see it you just lengthened the lever and didn't eliminate twist. Ok, square z casing will be stronger but will push more the gantry beams and bearings.
Ideally 3 rails at the gantry will be perfect as i see it with this design.
But adding third couple of bearing blocks on a normal gantry Z will do the same, its easier and cheaper. And 2 beams soldered together will be stronger than 2 separate beams as is your case.
So, try to lower one of the rails.

The point being here is that this design needs to be done correctly cause it will allow high speed machining or deeper dig, but you will need as i see it at least 3kw spindle, or it will become the weak point. As Dean said before.

[Wejjmeister]

Ok well lets see the rest of the design for Z axis and work from that.

My suggestion for nema 34 on 2020 screws was mostly for the long (X)axis to keep screw speed down to reduce whip and still maintain 10mm pitch speeds but I've just noticed it looks like your intending to use rotating ballnuts which if correct then you could use nema 23 motors on 10mm pitch.
That said 20mm pitch would allow a ratio to increase torque and have a lower nut speed for same speed as 10mm pitch which if using 20mm dia screws (which I would use at this length) may be better with the larger nut on smaller motors.

This doesn't help the Y axis thou which will still need nema34 motor if you use twin screws on single motor. That said Mixing motor sizes isn't a problem if you use High powered drives because they run from mains AC so don't have a mix matched DC power supply which would be needed for the lower voltage DC drives running the nema23 motors.

Ok that makes sense. I will swap out the motors for 23's on the X axis and then install a 34 on the Y axis in the fashion you stated, with pulleys and dual screws. I had intended on using 10mm pitch on the X axis, do yoou think that 20mm would prove advantageous, or would it be a case of pro's and con's?


What you miss with this gantry design is that the rails must be as low as possible, at least one of them. Otherwise is pointless. Now you are just making a big lever.
As it is now is weaker than normal gantry.
The big+++ of this design is eliminating overhang and stiffening at the bearing blocks and lowering the assembly and eliminate gantry twist.
But as i see it you just lengthened the lever and didn't eliminate twist. Ok, square z casing will be stronger but will push more the gantry beams and bearings.
Ideally 3 rails at the gantry will be perfect as i see it with this design.
But adding third couple of bearing blocks on a normal gantry Z will do the same, its easier and cheaper. And 2 beams soldered together will be stronger than 2 separate beams as is your case.
So, try to lower one of the rails.

The point being here is that this design needs to be done correctly cause it will allow high speed machining or deeper dig, but you will need as i see it at least 3kw spindle, or it will become the weak point. As Dean said before.

Im not sure how I can get the gantry any lower without it interfering with the material on the bed? Can you possibly sketch something up to demonstrate what you mean, as I am not following. I could weld another box section piece below the existing long section of the Y axis, but these would then protrude down into the cutting area, and reduce Z height.

Also just to clarify, the Z axis will be inside the boxed section, not on the front, overhanging it, so the loads will be evenly spread across the frame as far as I can see?

Running a 3kw spindle would be nice, but to begin with I think I would opt for the 2.2kw as I can find out the limitations of the machine with that and then start to push things later on in the process. The option of quicker cuts would be nice, but quality is my first concern to begin with.

Help?

[Boyan Silyavski]

I mean- to lower one rail at least on the gantry. To be nearer to the bed. Not on top like its now. I know- more difficult.

My point is that if you don't lower the gantry rails like Jonathan's friend machine, you gain nothing with this gantry design.

Yes, i get- the Z will be enclosed and strong, but long...so bearing blocks will be the weak spot.

Jonathan's design is great but changing this and that can make it weaker and not worth it or opposite, better it.

Obviously you haven't seen 2.2kw and 3kw spindles. There is 1hp difference, which could be quite important at low revolutions where there is not enough power. Also the bearings are bigger if i remember correctly.
A 100€ difference including VFD, which i believe is worth investing into.

[Wejjmeister]

Ok question whilst Im back to the drawing board.... Can I run profiled rails upside down, as in, on the base of my gantry box section? If not I will have to build the Y axis completely differently to the way I intended and therefore the Z too.

Also I have seen the difference between the 2.2kw and 3kw spindles and am aware of the differences, also the diameters are different 80mm vs 100mm I recall? I will use the 3kw at some point, but as with everything, budget has to come in somewhere.

[JAZZCNC]

Ok question whilst Im back to the drawing board.... Can I run profiled rails upside down, as in, on the base of my gantry box section? If not I will have to build the Y axis completely differently to the way I intended and therefore the Z too.

Yes no problem.

[Wejjmeister]

Thanks, makes it a lot easier. Will that put the twisting/leverage issue right?

[JAZZCNC]

Thanks, makes it a lot easier. Will that put the twisting/leverage issue right?

Can't answer that without seeing a complete design.?

[Wejjmeister]

It would be too simple wouldn't it!

[Boyan Silyavski]

Ok question whilst Im back to the drawing board.... Can I run profiled rails upside down, as in, on the base of my gantry box section? If not I will have to build the Y axis completely differently to the way I intended and therefore the Z too.
.

A picture from Jonathan's build (http://www.mycncuk.com/threads/6484-A-sufficiently-strong-machine)which shows exactly what i mean. I assume you know it by heart till now :toot:, going with that design. I wouldn't be over inventive, especially knowing that his machine works wonderfully.

http://www.mycncuk.com/attachment.php?attachmentid=9851&d=1377723791



Also I have seen the difference between the 2.2kw and 3kw spindles and am aware of the differences, also the diameters are different 80mm vs 100mm I recall? I will use the 3kw at some point, but as with everything, budget has to come in somewhere

Then just alter the design so you can fit it in the future.

[Wejjmeister]

Before I go any further...How am I doing so far? Is this along the right lines?

Also, mad question, but could I get the top, bottom and sides cut from 15mm Steel plate rather than Aluminium and use the epoxy method again? It would cost a LOT less than 20mm 6082!!?? The Gantry ends would still be 25mm Aluminium...

[routercnc]

Hi Weijjmeister,

The double beam 'MechMate' style gantry is a difficult design especially if you want to exceed the performance of a well made standard gantry. Hats off for trying it.

The design in post 83 is very limited by the rail being mounted to a flat plate which is mostly cut away. Also, the overall gantry is effectively 2 'C' shapes. As each 'C' shape is joined to the other only at the ends then you've lost a lot of stiffness.

One of the other tricky areas is the bit you haven't drawn yet which is the Z axis and the options for joining it to the Y axis including fitting the Z ballscrew etc. The design will quite literally hinge on what you do there which is making it difficult for everyone to critique the whole thing.

If you study Jonathan's design you will see that the gantry creates 2 box sections (to resist the twisting forces that your 'C' shapes will not) which are joined by a single top and bottom panel. You will also see that the top and bottom panels, plus the sides which take the rail are about 20mm (?) thick, whereas the outer panels are more like 10mm. This is an economical use of material where it is needed.

I think that you would be better off starting from your earlier post (#70?), but mounting the rails onto the inner faces of the box section, but with a thick plate between the rail and box section. Or you could put the thick plate inside the box section so long as it spanned most of the height of the section. This is because with this layout the rails are being twisted off the frame (rather than pushed in or out on a conventional gantry), and the relatively thin wall of the box section may not be good enough.

[Wejjmeister]

Thanks for the above, I really appreciate the time taken to reply and include so much detail. I am working on the gantry again so hope to ppst a new revised design taking everything on board.

This is what arrived today!! Exciting!!

[Wejjmeister]

Revision 5 of the Y axis with the beginnings of the Z axis too for your opinions on guys if possible. The Z axis is not finished, the upright plate will be mirrored to the other end and boxed in with the profiled rails mounted to it, the ballscrew will run internally on the same face.

126941269512696

[Lee Roberts]

Looking nice, any chance you could extend the width that little bit more to allow for the plate, shaped like a T, to fit to the full width. This would allow you to box in the bearing units and then come up the sides and across the top as one plate, if you do that on both sides it makes the whole thing a box and so stronger all round.

I like the way your going with it though :thumsup:

.Me

[Jezzag60]

Did this ever get finished? Looks like exactly the kind of machine I want to build.