Losing Steps in X and Y

[LeonS]

Good day all,

I tried cutting a really intricate vector representation of the Aztec calendar last night. It has thousands of vector lines, circles and polygons. It is a real acid test for controller accuracy.

I found that I was losing registration as the job progressed and the latter cuts tended to be incorrectly placed. The latter cuts were offset in the positive direction for both the X and Y axis.

I noticed that periodically during the cutting process iI would here a "thump" that was more distinct than the typical micro-steps of the head movements. I suspect that those were the incorrect steps that accumulated over time.

I have grounded the cases of the CO2 PS and DSP directly to chassis ground. My stepper drivers are wired on a DIY perfboard circuit. It is shielded from the CO2 PS by the 24v PS.

I will check the tension on the belts and drive elements but I suspect this is an electrical problem.

Is this the kind of errors that were noticed on the systems that were experiencing CO2 PS interference with the stepper boards?

Thanks in advance for any suggestions on how to proceed,
Leon

[twehr]

I have seen the same thing, on occasion. It is not so much a test of the controller accuracy as it is a test of your entire mechanical system. I believe the controller is doing exactly as it is instructed. The mechanics, however, may not be able to make very short fast changes of directions without some slippage, skippage, or stretching.

Try it at a different speed. Chances are that it will not be off in exactly the same places. That likely indicates mechanical. If it is always off exactly the same place by exactly the same amount, regardless of the speed you are running, then it could be the controller's ability, but 1000dpi capability should be accurate enough for any job your are doing.

[r691175002]

Twehr is right in suggesting that it is a mechanical problem.

There is a little more to the story though, because if your controller does not correctly honour acceleration limits it might be commanding the steppers to perform motion that is not physically possible. I highly doubt this is the case though.

My suggestion would be to lower the cut speed and/or increase stepper voltage and see what happens. Measure the voltage the steppers are seeing under load, my Chinese power supply is rated 36V 10A but drops to ~26 as soon as two steppers are drawing a couple amps.

[LeonS]

Thanks for the advice Gentlemen.

I'll explore the problem a bit more. I wouldn't be suprprised if it is mechanical. I think the DSP is fine; I am more concerned about the electrical interference on my home grown stepper driver board. I'll go back to some possible fixes posted by Bart back in the rev 1 days of his stepper driver design.

Cheers,
Leon