[dlaery]

I have a CNC Taig.
I can't get my backlash in x and y smaller than .007"
This may not be the right way to check this, but I have a maching camera, I jog to the left on the x axis, until I find my edge. Then I have my jog set at
.3% (.003) in Mach and jog right and the display will move .007" before the table moves. When I try to adjust the screws on the nut, if I tighten 1/8 or less of a turn, then its too tight to move. On the gibb, I can move the table freely by hand, when the motor is removed. This machine is a year old, I don't know what a lot of use is, but 10-20 hours a week is about the average.
I have read here of some gettiing .003" I think that would be good. I am making 2 part alluminum molds and would like to get a better finish. The cavities are rounded so the the tool is moving in all 3 axis.
Anyone have any suggestions on how to correct this?
It seems to me like the screw must not be straight, I don't know I will check that.
Thanks,

[awerby]

But you may have worn it down in the spot you're measuring. If you customarily do a part on a certain area of your machine, things can wear unevenly, so that while it's loose there, it may be too tight in the other, unworn part of your screw for the tightening of the nut to be effective. You might try replacing the screw and see if that helps.

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[awetmore]

Check to make sure that your gibs are tight. A loose table can also appear as backlash. The last time that I had noticeable backlash it was due to my X table gib loosening up.

[dlaery]

Thanks, I will try a different spots for measurement.
On the gibbs I have been tighting those back up to where I have no up and down play but still move freely.
When I first got this machine it kept throwing the reset, I have posted here and in Mach forum about this problem, and was loosing everything up to stop the resetting, but that wasn't the problem. I'm not having the reset problem now. One thing that was strange was on the x table a while back when I was having the reset problem and this didn't fix, I took off the motor and could not turn the screw with my fingers. Then after talking with Taig tech, he had me loosen the plate that holds the screw and I pushed up on the plate and tightened the screws that hold this plate and that helped a lot, then I could turn it easy with my fingers, the bearing in the plate seems free. It would bind up when I would tighten the plate and not push up on the plate. I have thought about enlarging those holes a little to raise that screw mounting plate a little higher to see if that helps.
I have had a lot of trouble with this mill. I would not buy another like it.
Thanks for your suggestions.

[awetmore]

If you grab the ends of the table and try to move it do you feel any play? If so your gibs are too loose.

I'm sorry to hear that you aren't having good luck with the Taig. It has been good for me. It isn't as rigid as my full sized mill, but for work that fits it works nicely. It is much better than the Sieg X2 that I once owned.

[lsces]

I have a number of people using their Taig's for very fine work, but the one thing that was common to all of them was getting them to carry out regular maintenance. The starting point on my own machine is always to run the axis's on a full stroke cycle at the start of a session to 'wake things up' this also helps even out wear on the leadscrews. We tend to work in metric, and 0.05mm backlash is the point that we would be looking to tighten things up a little. That equates to about 1 thou. At that even the open loop controller does not have any trouble driving the machine.
You mention '7 thou before it starts moving' ... This could indicate that the plastic pegs in the coupling itself are perhaps not sitting right, and that your problem is at this point rather than in the table. My own method of checking the backlash setting is to set up a dial gauge and jog back and forward 2 or 3 mm and confirm that the difference is in the 0.005mm area. Normally I just do this check with backlash compensation off and this gives me the initial setting for that axis. Once a week check on this then shows how the machine is holding up, and any sudden changes can easily be picked up.
Since you obviously have a closed loop machine, what sort of speed have you got it set to? One of the changes I normally do is to reduce the 'default' to around half of the Taig supplied setting - speed and acceleration. Does not affect backlash, but does help with 'reset' problems.

[dlaery]

On the pegs, I set the gap at what was recomended, I think it calls for 1/32 to 1/16, I used a feeler gauge to set that and had it closer 1/16, then it was suggested that was too loose and causing the backlash and so I have it set to, trying to remember, .040". Now when you say not setting right, do you mean gap?

Since you obviously have a closed loop machine, what sort of speed have you got it set to? One of the changes I normally do is to reduce the 'default' to around half of the Taig supplied setting - speed and acceleration. Does not affect backlash, but does help with 'reset' problems.

I guess you mean motor setting in Mach. Nobody has even mentioned this. The only thiing I have changed there is the Step pluse, it was on 1 and I changed to 9 because someone suggested this.
On the X and Y my Velocity is 60 and Acceleration is at 5, these are the default setting. So 30 and 3? What would I expect to see different? Would it run slower?

[lsces]

On the pegs, I set the gap at what was recomended, I think it calls for 1/32 to 1/16, I used a feeler gauge to set that and had it closer 1/16, then it was suggested that was too loose and causing the backlash and so I have it set to, trying to remember, .040". Now when you say not setting right, do you mean gap?

Actually it's more a case of does it feel right, but with the encoder on the stepper motor it's very difficult to actually check. The spacing is not so critical ... it's if the plastic tubes are 'snug', and at one point these were supplied a little too long so were actually binding.
The fact that x and y are affected is strange, normally only one axis is 'difficult'. How different are the two axies?

On the X and Y my Velocity is 60 and Acceleration is at 5, these are the default setting. So 30 and 3? What would I expect to see different? Would it run slower?

Actually the 60 is OK, the original files had 100 which IS too fast.

[dlaery]

I think is is more the x as far as problems, but is sounds smooth, the pitch of the motor sounds good. The y sounds different and changes pitch from - to +. When I first got this machine, the x motor went bad. I sent it and the controller back, they sent back a new motor, controller was ok. They said the motor seemed to have moisture. I am using flood coolant. Since then I have the motors covered with plastic where no moisture gets in. But that never solved any of the problems. The thing with the reset problem, it always reset when it would raise up and start to jog.
Now I am not having reset problems since I changed the Step Pluse, just would like to reduce the backlash. I am getting ready to cut a new mold and want to have it working the best I can. Thanks for your suggestions.

[strohkirchw]

So just to clarify, you're loosening the center nut and hand tightening the outer two nuts. Then you should be backing off roughly quarter turn and then re-tightening the center nut.

What is the backlash on z?

[dlaery]

Yes, I back off the 2 screws the same amount, then tighten a 1/8 to 1/4 at a time. Then put back togeather and if it tight and throws reset, I take apart and back off 1/8 and put back togeather to check until it works agaiin.
I am getting ready to repeat this process to see if I can get it better.
I am going to take the screw out and check it, clean everything and put back togeather.
I haven't checked the Z, I have had some problems with the Z, and have on adjusted the gibb and the clamps that hold it on the upright.
I will make sure and check that this time.

Thanks,

[Anokiernan]

Many of the parts I use my Taig for require 3d surfing and the ability to return to a constant Z height. If you find you need better repeatability on your taig's Z axis give this mod a shot it's worked wonders for me:
http://www.cnczone.com/forums/taig_m..._backlash.html

In the post I was only able to verify repeatability with a DTI, but here's a part that I ran using a constant climb spiral toolpath showing the results while under load. The part is 1.25" DIA and .050 deep finish machined with a .032" DIA ball endmill. Aside from running this part without coolant I was very pleased with the results: