[onri999]

Hi,

Just looking for a quick bit of advice. I'm currently in the process of designing a fixed gantry machine. After looking at a few more designs it seems quite common to conceal the screws and bearings round the back to avoid chips etc.

I'm wanting to extend the nut housing on the X axis fabricating a custom one out of a piece of 1 3/4" square solid aluminium bar which is 140mm long. As seen in this video:



Is this extension something to avoid or would it be okay?

Any advice or comments will be gratefully received

[ger21]

I don't think that's a good idea, but the entire Z axis is going to be less than ideal, with the bearings so high up, and the spindle hanging down so far.

[onri999]

I see what you mean with the bearings. Is that why you sometimes see people swap the positions of the rails and bearings. Have the rails mounted to the spindle mount-plate. Otherwise you sacrifice a lot of Z travel...

Any suggestions on what to do about the nut housing?

[onri999]

Had a quick redesign. Would love to know if you think this is any better:

[ger21]

I'd lift the top tube in the gantry, and put the screw between the two tubes. Maybe move the z motor behind the gantry so the Z axis doesn't need to be so tall?

[wizard]

My perspective on the first issue is do not do it as suggested in your images and text. The thing is this, the extension arm becomes effectively a lever arm. The axis slamming back and forth would effectively try to bend the fastening method or even pull the screws out of the aluminum. Small cross section bar stock does not give you a lot of flexibility in screw placement nor size.

If you really want to place the ball screw so far away from the Y axis saddle you need to consider a more robust solution. Generally though people tend to keed the screws as close to the saddle as possible.


As to your other questions i see some of the design choices as preferences. For instance i like to encourage people to think in box sections but wrapping the Y saddle around the gantry beam bothers me a bit. Mainly for a couple of reasons, one is that you loose clearance under the gantry beam. The other is issues of accessibility for maintenance.

The issues of maintenance are worthy of explanation. If you are machining dusty materials you will need to perform regular ball screw and linear bearing maintenance including cleaning and lubrication. Easy access means the work is far more likely to be done and done properly.

Another issue with the wrap around design is your motor position which impacts cooling. If vertical height is not an issue id move the motor top side simply to promote cooling. Top side also makes maintenance a tiny bit easier.

You are right to be concerned about dust! However moving the screw around doesn't do a lot for the dust that impacts screws the most. That is the dust that forms a cloud around your machine. It gets everywhere and highlights the need for dust extraction. The only other solution for dust is a bellows around the screw and they have their own issues. In other words screw position alone doesn't solve the dust issue.

[onri999]

Hi wizard,

Thank you for your detailed reply. It felt like a bad idea from the beginning

Since posting this I have kept redrawing the design and have drawn a box section as you say for the saddle, I have wrapped it around the gantry though. Not sure how to achieve a box section without doing this. Any precedents you can show me?

Here is a sample of my latest design:



I'm reluctant to place the Z axis motor up top since I don't know where it will live and space will almost certainly be an issue. I've seen people heatsink their motors, is this a way of getting round the issue of over heating?

[KH0UJ]

Well on my opinion every belt is a backlash, you dont want that to happen when machining precision parts, I think it`s just better to direct drive everything and find a way to close all the ball screws for dust debris contamination, the most important thing on a CNC is the ball screws, steppers drives are sealed so there`s a less chance to get contaminated even if you soak it with oil or diesel and run all week it will be just fine but the ball screws....just a little metal chips being eaten on the bearings and your material is done (garbage)

[A_Camera]

I am just curious... why do you have belt between steppers and screw? I mean, I would understand if you would have pulleys which are not 1:1, but in your model it looks like you have 1:1 pulleys, so what is the benefit compared to direct drive?

I would also change the design so that the centre of the spindle ends up at the centre of the frame. That gives you the maximum table movement and the maximum work area. Another thing you should consider is to add some more support for the gantry beam, pointing to the front of the machine. You design looks like a moving gantry design, and one of the reasons to build a fixed gantry machine is to be able to build it as rigid as possible. The weight of the gantry is not an issue in a fixed gantry design, so there is no reason for not increasing the rigidity.

[ger21]

THere's nothing wrong with using belts like that. I think that's a lot better.

[KH0UJ]

THere's nothing wrong with using belts like that. I think that's a lot better.

Yep I agree with you boss probably on wood, plastic and some other soft materials it`s OK, but for gun parts and other hard steel with pilot holes that had to be very precise (0.2mm offset is considered material reject) I dont think the belt will still cling on that accuracy when cutting / engraving hard metals.

[ger21]

I disagree. If you choose the right belts, they'll work fine.

[rowbare]

Read up a bit about Abbe errors...The subject usually comes up in discussions about measuring but they are entirely relevant here. In a nutshell, the further from the tool the nut is, the more systemic error you are introducing. If you can, keep the nut between the guides and the spindle.

bob

[KH0UJ]

Ok boss let`s assume it`s fine on a normal scenario, now im gonna compare it on our working environment, im always washing our units with diesel + kerosene every week to maintain it`s accuracy and to evade overhauling ball screws everytime (it`s a lazy style hehe) not to mention dust debris getting caught on one of the pulleys and throwing the accuracy off little by little, how will a belt withstand harsh cleaning solvents without getting soft and inducing backlash? I know this because our 4 axis chuck that has the belt got backlashes due to kerosene + cooking oil, that`s why im changing it to a planetary gear system or a screw gear system to withstand our kind of working environment

[ger21]

A belt shouldn't be subjected to any chemicals, and has nothing to do with whether a belt is suitable or not.
The router I use every day uses a belt to drive a 70" ballscrew, with a 2Kw servo motor, moving 1000 lbs, and there's no backlash.

[wizard]

Yep I agree with you boss probably on wood, plastic and some other soft materials it`s OK, but for gun parts and other hard steel with pilot holes that had to be very precise (0.2mm offset is considered material reject) I dont think the belt will still cling on that accuracy when cutting / engraving hard metals.

I have seen belt drives on all sorts of machines and properly implemented they have never been a problem. Even on machinery cutting optical elements other parts of the machine become an issue before the belts do. I even know of a case where the torsion in the lead screw was impacting the resulting optic and which was addressed by an engineer with some really fancy math to match the tool path to the expected unwinding of the lead screw twist. The simple timing belt drive was never a problem.

Timing belts come in all shapes an sizes, on injection mold machines I've worked on they have driven ball screws that are about 5 inches in diameter. The only time timing belts are a problem is if the engineer didn't take the time to make sure they where properly sized but other wise they are a very good way to drive lead screws in my opinion. Not the answer in every situation but there seems to be a lot of negativity on the forum with respect to belt drive lead screws which doesn't match my experience.

[wizard]

Hi wizard,

Thank you for your detailed reply. It felt like a bad idea from the beginning

Don't let it bother you. The general goal is to keep the lead screw as close to the axis saddle as possible. That sometimes can't be achieve with odd machines but on a router the general arrangement is well understood and doesn't require odd extensions like this.

Since posting this I have kept redrawing the design and have drawn a box section as you say for the saddle, I have wrapped it around the gantry though. Not sure how to achieve a box section without doing this. Any precedents you can show me?

I haven't reread what I posted but One of us got something wrong there. The saddle on the gantry axis doesn't need to wrap around the beam in my opinion. My fear is that you will loose more than you gain, especially with the loss of clearance under the beam.

Now a saddle plate and even the Z axis can benefit from something more than thin plate. Even C channel can offer an improvement, but don't forget that steel tubing can be hand in a number rectangular forms. How much you will benefit from these forms depends upon how much Z you want. The more the Z extends the more you have to pay attention to deflection to get the results you want. In a nut shell You will not need box sections in the Z for most designs.

Here is a sample of my latest design:



I'm reluctant to place the Z axis motor up top since I don't know where it will live and space will almost certainly be an issue. I've seen people heatsink their motors, is this a way of getting round the issue of over heating?

The first thing to realize is that Stepper motors are designed to run hot, very hot to humans actually. In industry they are often labeled with warning signs or guarded. Lets put it this way getting burned is a big disincentive to maintaining you ball screws and linear bearings. Heat sinking is pretty popular and can be seen on many older stepper based mills. It isn't an absolute requirement but can be beneficial. Today though better drives do help with heat management but so does good air circulation. Technically the steppers don't over heat with a properly running/sized drive as the current is limited by the drive. Rather they (the steppers) just operate outside a humans comfort zone.

You might want to take a trip to a web site of a stepper motor manufacture but one thing you will find that the steppers have rather high temperature rise figures. Often 70 degrees C over ambient and sometimes as much as 100 degrees C. That is extremely hot to the touch. That heat has to go somewhere and what you don't want happening is that heat expanding your aluminum frame distorting the machine in the process.

[A_Camera]

I disagree. If you choose the right belts, they'll work fine.

...but my question was what's the benefit? I understand that there are really high quality belts and pulleys, but I believe those can be very expensive, and even if not, what is the benefit other than it gives a better look and makes a smaller machine? Personally I would not like to use it because I NEED maximum accuracy for one specific job which actually pays my hobby plus some more, so I am with KHOUJ.

[A_Camera]

I have seen belt drives on all sorts of machines and properly implemented they have never been a problem. Even on machinery cutting optical elements other parts of the machine become an issue before the belts do. I even know of a case where the torsion in the lead screw was impacting the resulting optic and which was addressed by an engineer with some really fancy math to match the tool path to the expected unwinding of the lead screw twist. The simple timing belt drive was never a problem.

Timing belts come in all shapes an sizes, on injection mold machines I've worked on they have driven ball screws that are about 5 inches in diameter. The only time timing belts are a problem is if the engineer didn't take the time to make sure they where properly sized but other wise they are a very good way to drive lead screws in my opinion. Not the answer in every situation but there seems to be a lot of negativity on the forum with respect to belt drive lead screws which doesn't match my experience.

Again, my question was not a generic about timing belts, I know they come in all shapes and sizes and qualities, but in this specific case, what is the benefit/reason for THIS design? In fact, for me it is a good enough reason if the answer is "to avoid the use of couplers and allow easier stepper alignment, offering better looks and smaller foot print". Is there any other benefit here? Note that I am not against timing belts driving screws, I chose that to my first 3D printer and today if I'd build a moving gantry CNC I'd probably use a single motor with timing belt driving two screws, instead of dual motors, one for each screw, so I am not generally an enemy of timing belts. My reasons would be to avoid the need of squaring since once squared always squared if one motor is driving two screws using belt, but here I see no real benefit, only extra source of error. We can't really compare this design with industrial machines where costs is not an issue and EVERY component is of the highest quality.

[A_Camera]

Often 70 degrees C over ambient and sometimes as much as 100 degrees C.

Working temperatures are NEVER given as a value above ambient temperature, but as an absolute value. Stepper motors have 60+ Celsius working temperature, but that does NOT mean if your ambient temperature is 30C that your stepper working temperature becomes 90C. If that happens then you are in trouble, unless your stepper is made for that temperature. Basically, the lower the ambient temperature the higher the safety margin. Stepper motors lose torque at high temperature, so even if they may survive the 90C, they will not be very efficient.

That is extremely hot to the touch. That heat has to go somewhere and what you don't want happening is that heat expanding your aluminum frame distorting the machine in the process.

Yes, that's also an argument against the way this machine is designed. Especially the Z stepper is far too boxed for my taste, and the rising heat is trapped in the partial box, heating and expanding not only the aluminium but also the belt. Steppers need food ventilation, so trapping them is not a good idea.