Eyeballs and comments sought for noob build

[danilius]

Hi,

I have been lurking here for quite a while and learned a lot in the process, so thanks to everyone for that!

I've spent a while designing my first ever CNC router based on what I have read here and elsewhere on the interwebs, albeit with absolutely zero experience in the CNC field, so would appreciate comments, criticisms and anything else you care to throw at it, bearing in mind that I have probably little if any idea of what you are talking about.

This machine will be used for milling soft and hardwoods, plastics and probably some shallow aluminium engraving. The work are is 900mm x 600mm. Much of it is based around 120mm x 80mm t-slot profiles. I have not made any decisions regarding the electronics yet, since I want to get the design to a workable state first. All suggestions regarding the electronics are gratefully received nonetheless.

The rails (20mm) for the gantry are located under the bed in order to save some width and reduce the amount of dust contact that will inevitably occur. No idea if this is a good idea, or a disaster in the making. Gantry walls are 20mm thick.

All the motors will be drive belts, which will allow me to play around with pulley ratios. I have read a considerable amount regarding the belts-versus-direct-drive argument, but for my purposes I'm keen on using belts.

Some piccies below, looking forward to comments, sagacious advice and no doubt some humorous comments with regards to the sheer noobiness of the design. Please be gentle :-).

20551205522055320554205552055620557205582055920560 2056120562

[Tom J]

Hi
In general this is easy to build design but I would change it slightly:

1. Put rails on spindle plate and blocks on Z plate (this will give zero torque at top spindle position, your design create torque on spindle plate regardless of the spindle position)
20675
2. Y blocks I would install on top rather on the bottom (it easy to make guard if you want to protect blocks against dust) - too big arm from bottom blocks to top of the gantry - think about forces
20672
3. Direct drive is transfer vibration to the motor better to use gear and belt.
4. I would use one motor for Y drive driven by pulley
20673
5. Gantry should be connected on the bottom - not open. This make it stiffer and side plate a bit thin
20674

[danilius]

Loads of luvverly advice, thanks!


Hi
In general this is easy to build design but I would change it slightly:

1. Put rails on spindle plate and blocks on Z plate (this will give zero torque at top spindle position, your design create torque on spindle plate regardless of the spindle position)


Great idea. I don't remember seeing this anywhere, or probably I did, but simply didn't notice it!



2. Y blocks I would install on top rather on the bottom (it easy to make guard if you want to protect blocks against dust) - too big arm from bottom blocks to top of the gantry - think about forces


I was wondering about the pros and cons, so that's very helpful. Will have to look around for ideas to protect the rails from dust, or at least mitigate the issue.



3. Direct drive is transfer vibration to the motor better to use gear and belt.


Is the direct drive on the z axis such an issue? Should I make that belt drive as well?



4. I would use one motor for Y drive driven by pulley


Is there a disadvantage of using two Y motors? Sure, they can get out of sync, but surely two homing switches/sensors should take care of that between jobs?



5. Gantry should be connected on the bottom - not open. This make it stiffer and side plate a bit thin


I was not sure about that. My thinking was that by using two Y motors, I could save some weight, and the two motors and lead screws would compensate for the added stiffness the connecting plate would provide. Is this incorrect?

[Tom J]

Loads of luvverly advice, thanks!

Is the direct drive on the z axis such an issue? Should I make that belt drive as well?

Is there a disadvantage of using two Y motors? Sure, they can get out of sync, but surely two homing switches/sensors should take care of that between jobs?

I was not sure about that. My thinking was that by using two Y motors, I could save some weight, and the two motors and lead screws would compensate for the added stiffness the connecting plate would provide. Is this incorrect?

belt transmission gives you extra flexibility, you can adjust gear ratio means=feed, easier do deal with resonance if any occurs, allows to mount motor in two position (up side down and shaft up).
Z axis takes most vibration as is driving spindle up and down, so those shaky shaky gets to the motor and disturb in high speed. My first machine was driven by coupling - cheap Chinese, than with plastic insert, odiham finally end up with 1:1 pulley.

Y drive, if your power supply can handle biggest nema 23 3Nm motor without extra voltage than one motor should be ok and will wait only few hundreds grams more than 1.9Nm. Advantage of having one motor and two ballscrew is lack of need for squaring the gantry (one motor less, 2 sensor less)
Motor have to be powerful enough so they do not loos steps, if you loose steps you loose your job, material, time.

Gantry - more braces make it stiffer - joining them under bed will increase the weight but benefit you gaining by stiffer, more accurate machine are far more important.
I hope somebody else will give you some inside, I only build 3 machines and even on last one I would make something different, now I designed another one will post it soon.

Good luck with the build

[danilius]

Thanks for that. I'm a little concerned that a belt that long will introduce other issues, so will see what to do about that. Will attempt to alter the z axis for a belt drive. I don't suppose it should be too difficult.


belt transmission gives you extra flexibility, you can adjust gear ratio means=feed, easier do deal with resonance if any occurs, allows to mount motor in two position (up side down and shaft up).
Z axis takes most vibration as is driving spindle up and down, so those shaky shaky gets to the motor and disturb in high speed. My first machine was driven by coupling - cheap Chinese, than with plastic insert, odiham finally end up with 1:1 pulley.

Y drive, if your power supply can handle biggest nema 23 3Nm motor without extra voltage than one motor should be ok and will wait only few hundreds grams more than 1.9Nm. Advantage of having one motor and two ballscrew is lack of need for squaring the gantry (one motor less, 2 sensor less)
Motor have to be powerful enough so they do not loos steps, if you loose steps you loose your job, material, time.

Gantry - more braces make it stiffer - joining them under bed will increase the weight but benefit you gaining by stiffer, more accurate machine are far more important.
I hope somebody else will give you some inside, I only build 3 machines and even on last one I would make something different, now I designed another one will post it soon.

Good luck with the build

[Tom J]

Thanks for that. I'm a little concerned that a belt that long will introduce other issues, so will see what to do about that. Will attempt to alter the z axis for a belt drive. I don't suppose it should be too difficult.

use good quality gates HDT 15mm braided one - there will be flex which shouldn't give you error bigger than 0.02mm (I referring here to my steel machine which was 115kg)

[danilius]

Food for thought. Thanks! Time to get back to drawing :-)


use good quality gates HDT 15mm braided one - there will be flex which shouldn't give you error bigger than 0.02mm (I referring here to my steel machine which was 115kg)

[paulus.v]

Hi,

Connecting the gantry at the bottom will give little benefits but, by moving the legs far apart, the base will be a lot weaker.

Why would you put the X rails "hanging"? Only for dust/swarf protection? The side plates extension for guide blocks will deflect a lot and put pressure on the rails/blocks trying to twist them. I don't think it's worth it.

The gantry side plates are beefy but only when looking from one direction. If you cut out one third longitudinally and bolt it perpendicular you will have more rigidity in the Y direction without any loss or weight gain. And you can connect to your t-slot profile overhang.

[danilius]

OK, I think I have understood all this correctly. So, I'll move the rails up top, and stiffen up the gantry walls.

[danilius]

So, I have made some changes. I wasn't convinced that the extrusion on the gantry was going to be rigid, so that has been dropped in favour of some solid lumps of metal. It is also much shorter, and the rails have been moved up to the top.

20926

20930

What I'm wondering is if there is an optimum distance between carriages for the gantry.

20927

I dont have any worksop equipment beyond the basic tools, so don't want to drill into the extrusions or anything, but I'm getting a bit lathered about the hseer size of the machine, given that it's only going to have a work area of 600 x 900. Currently the overall dimensions are roughly 1400mm x 940mm. I can trim a small amount off the length, but is there any way of getting this more compact?

20928

20929

[paulus.v]

Whydon't you put the horizontal plate of the X axis carriages under the gantry side plates. You will have a lot of unnecessary weight removed and by putting the ball nut support under the same plate you can shorten the ballscrew and the main frame (to the length of the rails). The gantry will slide above the ballscrew end blocks. It is also a good point to put the ballscrews closer to the working surface/plane.

I think you need a plate at the back of the gantry to crate a box. Even a thinner plate will increase rigidity.



What I'm wondering is if there is an optimum distance between carriages for the gantry.

I don't think there is an optimum distance... The 20cm that you have in your project looks good to me. But I'm not an expert.

[Boyan Silyavski]

If you are using Hiwin 20 long blocks/ or even short/ , 200mm between both ends from outside measured is good for your machine. 260mm is desirable for serious build and i would say ideal, and 300mm for OTT and heavy gantry and really strong and large machine. So basically with Hiwin long blocks they will be almost touching each other .

The idea is the bit to be in that distance , looked from side. In other words the spindle overhang. But on a machine like yours even if not, its not a big deal, as bearings are much more rated than forces involved.

[danilius]

Took me a minute to figure out what you meant, but that's going to be the next change! Thanks.


Whydon't you put the horizontal plate of the X axis carriages under the gantry side plates. You will have a lot of unnecessary weight removed and by putting the ball nut support under the same plate you can shorten the ballscrew and the main frame (to the length of the rails). The gantry will slide above the ballscrew end blocks. It is also a good point to put the ballscrews closer to the working surface/plane.

I think you need a plate at the back of the gantry to crate a box. Even a thinner plate will increase rigidity.


I don't think there is an optimum distance... The 20cm that you have in your project looks good to me. But I'm not an expert.

[danilius]

If you are using Hiwin 20 long blocks/ or even short/ , 200mm between both ends from outside measured is good for your machine. 260mm is desirable for serious build and i would say ideal, and 300mm for OTT and heavy gantry and really strong and large machine. So basically with Hiwin long blocks they will be almost touching each other .

The idea is the bit to be in that distance , looked from side. In other words the spindle overhang. But on a machine like yours even if not, its not a big deal, as bearings are much more rated than forces involved.

Thanks for that, quite reassuring.

[danilius]

So, having taken your sagacious advice on board, the gantry walls now sit on a block that is shared between the ballnut and the blocks. The ballscrews are essentially level with the bed. So, gentleman, in your considered opinion am I at the point where it's worth putting all the detail in now?

21018

21019

21020

[routercnc]

So, having taken your sagacious advice on board, the gantry walls now sit on a block that is shared between the ballnut and the blocks. The ballscrews are essentially level with the bed. So, gentleman, in your considered opinion am I at the point where it's worth putting all the detail in now?

21018

21019

21020

Yes, looking good now.

[Greeny]

Hi,

Something to consider might be a double plate set up when attaching the y-axis. This allows for easier adjustment.
In essence the y gantry sides have a base plate attached to form a y-axis assembly. This can then be bolted to a matching plate that is attached to the x-axis carriages.
See http://www.mycncuk.com/threads/10403-Project-Beaver/page4/#31 & the following post for a better explanation.

This can be seen in a very recent post of the finished machine here (post #69)
http://www.mycncuk.com/threads/10403-Project-Beaver/page7/#69

Cheers

[paulus.v]

Good point of the double plate for easier adjustment.

If you want a stronger machine...
21064

Regarding the length of the X axes screws and the overall length of the machine I do not see why you need the frame longer than the linear rails. If you lower the ballnut mount with a spacer and move it to the back (closer to the motor) you can shorten the screw at front by at least 10 cm, having the end blocks going under the gantry base plate.

[danilius]

Hi,

Something to consider might be a double plate set up when attaching the y-axis. This allows for easier adjustment.
Cheers

It's definitely something to consider as a future mod, or son-of- machine. I really like the concept. Right now, since this is my first ever build, I want to keep things as crude as possible. From what I have read I have so mucg fun to look forward too, like aligning everything, tramming and smashing through a collection of cutters. Can't tear mt hair out anymore coz I'm bald.

[danilius]

Good point of the double plate for easier adjustment.

If you want a stronger machine...
21064

Regarding the length of the X axes screws and the overall length of the machine I do not see why you need the frame longer than the linear rails. If you lower the ballnut mount with a spacer and move it to the back (closer to the motor) you can shorten the screw at front by at least 10 cm, having the end blocks going under the gantry base plate.

Oooohh, nice one. Definitely need to give that a go. Thanks for that.

In fact, thanks to everyone. Sop far all the advice has been spot-on and seriously helpful.

[danilius]

Being such a noob, it just occurred to me while working on the detail that I have no idea if I should go for tapped holes or use nuts (nylocs?) instead. Is there a rule of thumb for this sort of thing?