PCI 7340 communication with Applied Motion SV7-S-AF servo drive

[Earthtech]

The motor does move during the tuning, just not in any rational way.  90% of the time, using anything for P (I and D at zero), the motor barely moves.  Something like this, sometimes in the opposite direction too:

But if I keep clicking Step Response using the same exact PID numbers, in other words, without changing ANYTHING - occasionally, I will get things like this:

Essentially, I have erratic behavior.  Maybe I'm missing something, but its so hard to troubleshoot when you can't get it to repeat the same behavior.

I don't think the controller is using PMW - it's just in this "step response" mode that it produces a pulse of 10V.  Other times when the controller does manage to move the motor for a significant period of time, it's not a pulse, but a steady DC voltage.

[Brian_Beal]

Marissa-

I think you are right about the PWM.  The specs say it is a 16-bit DAC, so that is a true DAC.  I would still try the open loop move.  Either disable that axis in configuration or configure it as a stepper.  You will then be able to set a voltage on that DAC through your application program, which I am assuming is LabVIEW, using the Load DAC function.

[Earthtech]

I'm using Measurement and Automation Explorer.  It should be the most straight forward way to get this thing to work, so I can be sure I just haven't made a goof in programming LabVIEW.

I don't really see how to run a servo in open loop as a stepper.  The controller has separate outputs for steppers (different pins).  The output from the controller for a stepper is a bunch of pulses on one output and a + or - voltage on another to tell it direction.  My driver is specifically for servos and has no way to deal with step and direction signals.  I did switch the controller to this setting and watch the pulses on the scope (the outputs were not hooked up to the driver).  It does what I would expect it to.

Thanks for the suggestions.  Keep them coming!  My main troubleshooting right now is staring at the thing and occasionally telling it to run knowing full well it's not going to behave the way it should.  I'm waiting on my new version of LabVIEW to try serial communication.  NI is literally within 2 miles of my lab but I have to wait for them to ship it!

[Highland_Controls]

Marissa-

I only suggested to configure it as a stepper so that the DAC would not be mapped, and hence be available to control directly from LabVIEW.  What version of LabVIEW do you have? Any version should be able to communcate serially.

Regards,

Brian

I just noticed that I login automatically from different computers as different users.  Brian Beal and Highland Controls are both me.

[Earthtech · Accepted solution]

I've been using LabVIEW 7 on a Windows 7 machine for probably 6 months with no problems...until I tried to communicate with a serial port.  It seems to be the only problem.

Soo....um.  I've fixed the problem.  I could kick myself.  It's the sort of thing that I spent hours on the phone with tech support about.  I just KNEW it was a settings problem in MAX, but there seem to be so few so after going over them several times on the phone, it seemed a dead end.  None of the application engineers could come up with any of my settings being wrong.  But by randomly changing things I've solved the entire issue.  The encoder pulses weren't being read right.  The default in MAX is to have A, B and the index be Active High.  I tried switching them all to Active Low and that didn't work so I hadn't messed with it again until now.  Apparently, it needs to be A is Active Low and B is Active High.  I've looked through the Applied Motion literature - it says nothing about this!  Grrrr...

Ok folks, issue resolved.  What a pain.  So much time wasted for just one radio button on the wrong setting!!!!

Thanks for all your help Brian.  I probably would have given up on the controller had not you kept suggesting things and keeping me on task.  Ok, on with the rest of the project!

[checho_CR]

Hi Marissa, im using a similar model Driver Sv7, and a PCI7344.

According to the datasheet of the PCI7344, the controller needs complementary signals. A+, A-,B+,B- .. etc.

But the Driver Sv7 does not, ABZ is enough for the Driver to work.

My encoder only generates A+ B+ Z+.

How do you managed to connect only this signals to the motion Controller without A-,B-Z- ??