EZTRAK II DX Z axis drift

[dmac88]

We are currently experiencing problems with our z axis drive. For example:
If the machine is programmed to cut .250 deep it will exceed this by a few thousandths. The deeper the z axis goes it becomes progressively worse. We believe this deviation could be caused by a problem in the resolver. We are unable to address this problem because we are missing pages 6-20 through 6-23 in our manual which explains how to adjust the pots on the amplifier board. If you have these pages please let us know.

[machintek]

!st question. First or second generation Z drive?
1ST gen has ball screw on side of head in a tube.
2nd gen has ball screw in front of quill in a box.
If 2nd gen, does it have 1 or 2 cables coming out of the Z axis motor?
1 cable means no encoder on the motor.

George

[dmac88]

The machine has a 2nd generation ball screw and also has 2 cables leading from the z axis motor.

Thanks for the response.

[machintek]

Let's look at something. Is it possible that your pitch error comp is corrupted on the Z?
Once you home the machine, press the * key.
A box should pop up on the left side of the screen.
the code is 11235.
try adjusting the Z pitch error.

George

[dmac88]

Thanks for the advice. We followed your instructions and found that the machine had been set to a z pitch error of .001". We changed that to .0004" and are thinking of cutting another test block. Do you know what a reasonable pitch error setting for z on this machine would be? We are seeing as much as .034" deviation over a 2" depth when cutting.

[machintek]

If you use gauge blocks, or a step gauge, how much does the inaccuracy change per each additional inch?
This would confirm a issue with pitch error comp.
If you can get into DOS, go to PFM.exe cursor down to tuner, and in there jog the z axis back and forth and see if the numbers are balanced (position error and command).
Your drive may need tuning.

George

[dmac88]

Thanks again for the great advice. The inaccuracy in the z axis we are seeing at 1" is .019" and at 2" is .034". I was able to access the PFM.exe file, per your instructions, and noticed that when I jogged the z I was left with a postion error of -.0002. The position error on the x and y axis was
.0000 and .0001 respectively when these axis were jogged. I am not sure what the degree of position error equates to. When you say "command" do you mean DAC-CMD. I see this term displayed but I do not know what is means. For the z axis under DAC-CMD I get a value of .0003.

[machintek]

You may have multiple issues. First of all your error is cumulative. Either your pitch error is off quite a bit (I have never seen this greater than +/- .0035 inches) or you have some other issues. Has the motor been changed recently? Or ball screw?
If board alone was issue, error would be the same where ever the position was.

George

[dmac88]

The motor and ball screw have not be replaced. We have been the only owners of this machine and it has given us good service for 10 years. I am begining to think that it is the encoder. Which is easy to replace. However, I do not know how to adjust it and we are missing pages 6-21 through 6-23 in the manual.

[machintek]

The concept of the encoder misreading is valid. Lets assume it is missing counts. Then the axis would move further with the same count and would return correctly if missing count was bi-directional.
I do not believe the Z motor can be changed with the X or Y because when installing these Z axis, the motor pulley was keyed. The X or Y are not. Otherwise the typical was of verifying the problem would be to swap motors and see if the problem follows the motor.

George

[NC Cams]

Ifyou use the machine a lot, it may be time for a motor refresh. Clarkson id the factory servcie rep fo rthe SEM motors. His service is impecable. I recently sent hiim a motor to refurb - he tested it and said "no refurb needed". Sent it back at no charge.

BTW, he also tests encoders and tach F/B as well.