Bruce Klawiter wrote: > Problem solved. > > Well I hope it is solved and I don't even know how I did it. OHHHhhh NOOooooo! That's the worst kind, as it can come BACK! > Here is what I did today, > Jon said "You may be able to use the AC range on a DVM, and check while the > machine is moving, that might have an effect. The AC range of the meter > should > show just about zero, anything over 0.1 V AC is highly suspicious that there > is > ripple on the supply due to failed capacitors." > > I did this while the X axis was moving and all the axes were twitching, I got > a steady reading on the meter. > These are the readings DC volts -15.02 on AC range .101, DC volts 15.19 on AC > range .001 > Well, this is NOT OK! .1 V AC on the minus power supply to the op amps in the servo amp is a real problem. And, it isn't your meter, as the + side is reading .001, which is fine. > I think I remember Jon saying the PPMC motherboard is not grounded but could > be, by grounding the D-sub body, I checked and it was grounded to the chassis > so I thought I would find out how. I unplugged everything from the PPMC > boards and found if the amps where plugged into the DAC or the parallel cable > was plugged in, the D-sub connector was a dead short to the chassis. > Don't know what that means but I made a not of it and put everything back > together. > > Yes, the DAC outputs have a signal pin and a ground pin. The encoder board also has a ground pin for each encoder. > With all this talk of grounding I thought I would unscrew the the PPMC > motherboard from the chassis and just let it float in mid air. I did this and > ran the mill again and the twitching went away, I thought cool I found it, > but to make sure I knew I had pinpointed the problem, I remounted the PPMC > motherboard back were it was and the twitching has not returned so I don't > know what I did. > > > Any idea what I did that corrected the twitching You had a bad electrical connection. It may have been that the PPMC card cage was NOT grounded to the chassis, and noise was coming from the computer through the parallel port to the PPMC. Anyway, somewhere in this mess, there must have been a bad connection, and in the process of taking it apart and reassembling, a loose screw got put in tighter, paint that was insulating something got scraped away, or something that wasn't plugged in securely got reconnected better. Unfortunately, with older CNC equipment, some of the connectors on them start out being of dubious quality, like tin-plated Molex and AMP connectors, and as they age the connections get more and more flaky. If it comes back, start wiggling and reseating connectors one at a time until you find the exact culprit, and then you'll know what to repair. Make sure the PC power supply is properly grounded to the panel. Another thing might be to run a short ground strap directly from the DB-25 on the PPMC motherboard to one of the mounting screws of the PC motherboard. They look to be only a couple inches apart in the photo.
The bad news is this sort of stuff is REALLY common, depending a bit on the environment the equipment has lived in. But, I'm sure glad you got it solved, at least temporarily. (I'm not trying to jinx you!) This proves it is not a fundamental problem, but is almost certainly in the bad connection category. Those can STILL be maddening. Jon ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2d-oct _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
