Really stuck here and looking for some help...

Trying to get past some other issues, we changed from ACPI to standard PC per the XP optimizations. We used to get a steady pulse frequency on the diagnostics page, but now we get something that bounces around in the sub 100 range (2, 38, 15, 80, etc). It will drive the machine, but we are having issues that we don't know if they are related to this, noise, or something else. The first thing we'd like to do is get the pulse frequency issue resolved. When we run the drivertest, it reports that it is 'excellent', but the diagnostics still look bad.

Desperate for some guidance here...

Can you attach your xml and I will see if I can replicate it here.
Hood

here it is. much obliged!

Seems fine here, get a good steady pulse frequency of 24999 so thats fine.
Do you see a time in Int on the diagnostics page and if so what is it?

Maybe try the SpecialDriver.bat and see if that helps but I am thinking it wont.

Hood

Did an XP system restore because I couldn't get it to leave "Standard PC" mode (it would not see the ACPI stuff). Once we did that, and the machine is back to ACPI mode, the pulse frequency reads steady at 25010.

Now...back to the *other* problem....where it goes off the reservation a random time into cutting a piece. *sigh*

Thanks so much for taking the time Hood!!!

Ok the problem, seems you are saying the axis or more than one axis will just go wrong?
When that happens what do the DROs and Toolpath View say? Are they also showing it going off?

In the xml you attached you had the pulse width at 0, you may need to increase this depending on the drives you are using.
Hood

Sorry I didn't get back sooner...

The toolpath thinks it's where it should be, but I'm not quite sure I know what DRO is.

The pulse width was done basically by following the config videos (and/or books) which said to run through the motor tuning and take the defaults for pulse width.

All that having been said, we've been having a monster of a time with this and have completely rewired everything with shielded wire. Once we got it back into ACPI mode (with appropriate pulse frequency reading), we got back to a limit switch problem. We had removed the caps we originally put on the limits with the hope that all the shielded wire would obviate them, but found it didn't, so we put them back on.

After all that...it *seems* to be cutting right finally. We've had some other "false positives", so a few more runs will be necessary before we call it good. This whole thing has been exacerbating because it all of the sudden went to hell in a hand basket after working okay previously.

Back to the pulse width...these are KL-4030 drivers running with a C10. Do you think we should crank up the pulse width? We toyed with the idea of sherline, but didn't get there (yet).

DRO stands for Digital Read Out and is the name given to the displays of position for axis, for example the X Y and Z positions. It is also any similar readout in Mach for example the Tool number DRO.

If the Toolpath is showing things correct then Mach has output the correct steps so its hardware issue, either the hardware itself or noise.

I would try maybe 4 for Step Width and Dir on motor tuning page and see how that goes.

Hood

After WEEKS of fighting with this thing, we think it's FINALLY there!!

Checked every mechanical connection and rewired the entire thing with shielded wire. You name it, we tried it! Last step was increasing the pulse width, which we were considering but you nudged us over the edge Hood and THANK YOU!! It's now working better than it ever has before!

Sometimes it the simple things that work Some hardware needs a wider pulse and also it actually depends on your computer but good you got there and shielding is always good so its not been wasted .

Hood