On 06/27/2015 03:57 PM, Jon Elson wrote: The error occurred several more times, it seems it was worse on a cold start, and then eventually began to behave. So, I left it running with Halscope monitoring an axis, both axis.n.f-error and ppmc.0.encoder.nn.velocity, and did other stuff in the shop. It began to fault more often, and then I caught it with gigantic jumps in the DRO position indicated. A quick calculation showed these jumps came to 16.7 million encoder counts. The only thing that could cause that would be a misread of one or more encoder bytes, causing the software to perform a carry from the 24-bit hardware to the 32-bit raw position count. I jiggled some connections, and the problem appears to have stopped. I need to do the same machining job where I last saw the problem, so I ought to find out whether there really was a bad connection between the PPMC boards and the PC parallel port.
But, it sure looks like it was, indeed, a hardware problem. I still don't understand why the Halscope trace did not show this massive jump in position as a velocity spike and a spike in pid.n.error. On the discrepancy between pid.n.error and axis.n.f-error, there definitely is a difference between tuning using pid.n.error (which I had been doing since EMC (1) days) and axis.n.f-error, which apparently is really the operative thing that causes the following error to trip. Also, every couple years, I need to tweak the velocity amp offset a TINY amount to zero out the following error at standstill. This needed to be done again, and reduced the following error quite a bit. So, I THINK the problem is fixed until whatever got out of whack develops again. Jon ------------------------------------------------------------------------------ Don't Limit Your Business. Reach for the Cloud. GigeNET's Cloud Solutions provide you with the tools and support that you need to offload your IT needs and focus on growing your business. Configured For All Businesses. Start Your Cloud Today. https://www.gigenetcloud.com/ _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
