First build - Mid-scale gantry router design

Hi all,

As a little disclaimer, I have been researching building a machine for around 24 hours. I realise the learning curve is huge and the questions I am likely to ask will be the ones you're all dead bored with answering. With that in mind I will try to keep things to simple yes/no answered questions whilst I design my project.

----

Basic parameters:

-I would be machining almost solely aluminium, often 6082T6 although I use 7075T6 and other aerospace grades regularly also.

-I'd like to shoot for 500x500 working area, a little smaller is fine. Unsure about bed to spindle distance, will know more as designing goes.

-High tolerances. I'd like to be able to machine bearing pockets and such and have some control over the fit. Hoping the relatively small scale using parts typically found on larger router builds, coupled with the right construction, will achieve this.

----

I am aware that Hiwin type rails are ideal. I am pretty sold on using them for the Y and Z axis (HGR15 size?) as they are preloaded, which makes them infinitely more versatile. However, they aren't cheap and I'm afraid I do have a budget.

My first question is, assuming for a moment that my design is rigid and viable: is it realistic to expect high tolerances in aluminium if my X axis runs on supported round rails, such as SBR16s? I would mount them flat, so the weight of the gantry is what provides preload. I think this would be okay, opinions extremely welcome. This is purely a cost cutting measure, no point hiding it.

To add to that, casting an eye over the vague description above and yelling at me if I'm dreaming is also invited.

----

I have yet to deeply research what steppers/drivers I'd need for this scale machine, I will come to that once I have designed a generic frame.

Always surprising how much you can write without actually saying much...

Cheers and thanks in advance,
Ellis

Edit: I suppose this is punishment for posting so early. I have since read that typically folks would recommend the HRG20 size over the 15s for several reasons, and that in fact via Aliexpress, even with import, they're pretty affordable. Although the supported round rails are even more extra affordable!

Quote:

---

Edit: I suppose this is punishment for posting so early. I have since read that typically folks would recommend the HRG20 size over the 15s for several reasons, and that in fact via Aliexpress, even with import, they're pretty affordable. Although the supported round rails are even more extra affordable!

---

Hi and welcome to the forum. You have answered you're own question in that 20mm rails are preferred because of the spacing ( the 15mm are just to tight to get in) If you are want to cut ali with any repeatability you will need square rails all round.

Don't buy any motors, drives of any electronics until you have the frame sorted.

So as usual try and get a design together using sketup or whatever. and the good people on here will guild you through the maze. Good luck with the build. ..Clive

Hi Clive, thanks for the reply.

I am designing as we speak. I use Autodesk Inventor for all projects, so I'll draw something up and post it here, hopefully later today.

Am I right in looking at sellers such as: http://www.aliexpress.com/store/prod...287879562.html ?

Am thinking the above (or the same from elsewhere) with a central stepper and timing belts to dual ballscrews for the X axis. I believe this is fairly standard.

Cheers,
Ellis

Quote:

---

Originally Posted by Ellis

Hi Clive, thanks for the reply.

I am designing as we speak. I use Autodesk Inventor for all projects, so I'll draw something up and post it here, hopefully later today.

Am I right in looking at sellers such as: http://www.aliexpress.com/store/prod...287879562.html ?

Am thinking the above (or the same from elsewhere) with a central stepper and timing belts to dual ballscrews for the X axis. I believe this is fairly standard.

Cheers,
Ellis

---

The folks on here quite often use Chai at liniermotionbearing2008 or BST at http://es.aliexpress.com/store/314742

With a machine 500 x 500 you might get away with one screw on X (others will chime in here) but if use two screws the choice is up to you, one motor driving both screws or two motors slaved.
Have a good read through some of the build logs to get ideas and copy if need be. ..Clive

Hi Ellis, welcome to the forum. For the size that you are building, and the materials you wish to cut, I personally think it would be worth thinking about a fixed gantry rather than a moving gantry. With a machine like that and 20mm rails and a single 20/25mm central ballscrew I think you could build a really rigid little machine that would suit your requirements, but would of course require a longer X axis. Good luck with whatever you decide, keep posting your designs and thoughts and some of the very experienced members here will offer all the help you need. G.

Welcome to the forum,

My machine has hiwin type rails on Z (15mm x 300mm) and Y (20mm x 700mm) but the X axis (900mm) is 16mm supported round rail. I can cut aluminium well and hold a good enough tolerance to press fit bearings into a bore, so it can be done with a mix like that. However, the hiwin (square / profile linear rail) are so much better and stiffer and the prices these days means that if you are starting from scratch you stand a better chance with hiwin all round.

As for 500x500 single ballscrew, my experiences are this is OK for wood, but aluminium is less than ideal as the gantry will rack a bit when you cut at the extremes. I upgraded from single to twin and it is much better. You might be OK cutting in the middle for very small parts but I personally wouldn't go there again.

Spindle to bed clearance is a personal design choice, and 150mm or so is popular, but if you go larger you will loose stiffness when cutting a full depth. Some designs use an adjustable bed, or use a purpose made 'solid packer block' to bring work up to the cutter. For me simple is best and 150mm or so is worth starting at in your design to see if it will cut everything you need.

You haven't mentioned gantries yet but an elevated X axis allows a superior gantry design(i.e. minimal / no gantry sides). Have a look at the recent build logs as most people go this way now - Jonathan, Silyavski, Eddie, Neil etc.

Thanks all for your responses.

@Clive: I had heard about Chai, but found his prices don't seem to compare particularly greatly to other sellers. I realise of course that any of these vendors could be shipping sub-par kit. Is there a reason particularly why Chai is popular?

@Geoffrey: exactly that had crossed my mind as I can quite see a fixed gantry would potentially be much more rigid than a "live" one. I dismissed it because the footprint of the machine to work area ratio is so much greater with a moving gantry. However it's more important to me that I can hold tolerances reliably, so that style of machine is certainly a potential. Thanks for the thought.

@routercnc: great to hear you have experience with almost exactly my proposed setup! After finding the seller on aliexpress I linked above, who appears to offer every mechanical part for reasonable prices, I think I may push for all-Hiwin rails.

I had always pictured using dual ballscrews driven by a central stepper and belts. It feels right from a mechanical standpoint and it's good to know it's popular. I will be making the Z travel minimal for the reasons you stated. 150mm sounds about right to me.

Thank you for bringing up the raised X rails. That makes absolute sense. I won't even bother with the design I have currently which has the rails actually below the bed height. Cheers!

Thanks again all,
Ellis

Quote:

---

@Clive: I had heard about Chai, but found his prices don't seem to compare particularly greatly to other sellers. I realise of course that any of these vendors could be shipping sub-par kit. Is there a reason particularly why Chai is popular?

---

I think BST is now the preferred source and better quality. ..Clive

Without doubt if you want high tolerences then you need to go with a Fixed gantry. Better still use a design that doesn't use your typical Z axis approach.?
To get the best surface finish and hold tight tolerences it's critical that tool length is controlled and kept to minimum. Typical Z axis is always a compromise between material clearence and tool extension.
A far better way is to not have a Z axis as such but rather have the Axis which spindle runs across raise and lower. This ensures the tool extension is always the minimum it can be regardless of material height. The only extension is the tool stickout from the spindle. This results in much less chatter, less vibration and higher tolerences.
Check this out to see what I mean just make a better job of it and use a proper spindle.!!. . .
https://www.youtube.com/watch?v=4Bj6er322Dg

Make it very strong and Dense to minimise vibrations.

Wouldn't even think to use round linear rail if you want high tolerences as it will be the worst thing you could do and you will regret it for the slight difference in price.
Building a machine to hold tight tolerences is very very hard to do and will require substancial effort so don't spoil it for sake of few quid extra.