7i77 ground vs. servo amp ground

Hello -

I'm working on a second revision of my milling machine controls and I'm attempting to get my 5i25/7i77 pnp kit working with 3 Aerotech BA20-320 servo drives.

I've done some work and gotten to the point of trying to command the first drive to move the X axis. I have the 7i77 set up with X/Y/Z axes, field power from a 24v power supply, 5v from the 5i25, and I've verified that the drive is enabled (after debugging the active high/active low settings on that input) and my limit switches are connected and are supplying power to the limit inputs. The enable light on the drive is on and follows the "power on" soft button.

FYI, the encoder on axis X (axis 0) works and reads as it should.

I've verified that the AOUT0 and GND0 for axis 0 on the 7i77 output a positive voltage with respect to each other when I command a move (jog on X).

However, the drive doesn't output any voltage to the motor. I've verified I'm using the correct motor connections and that my motor works, and that the voltage is being applied to the correct pins on the drive (measured at the drive) but nothing happens.

I did read in the 7i77 manual that I should connect the AOUTN and GNDN as I have done, but also:

"the drives common or GND signal should be connected to the 7I77 power GND with a separate wire"

My questions are:

1) Could a lack of this GND wire be causing this to not work? I'd assumed that they should already have the same ground potential, but now I'm not sure.

2) I have located the drive's common ground for the control pins, but what do I connect it to on the 7i77.. field power ground?

Thanks,
Erik

Normally the drive analog input ground is at frame ground, You might verify that this is true
If not and the drive has a separate input ground, you can ground this at the 7I77's GNDN pins

Does the drive have differential inputs? (sometimes labeled IN+ and IN- or REF+ and REF-)

In this case AOUTN should go to IN+ (or REF+) and GNDN should go to IN- (or REF-)

are you able to move the motors by supplying a small voltage to the input terminal of the servo amplifier ?
i tend to use a 3v dc supply made from 2 batterys for testing , if it moves then you should be ok from the 7i77 , if not then it could be the enable is activated and needs to be reversed i.e it's on when if needs to be off

The drive does have differential inputs, and AOUTN goes to + (non inverted) with GNDN going to - (inverted input).

For reference, here's the drive manual:

www.servo-repair.com/documents/aerotech/BA10_20_30.pdf

I did verify that the drive would control motor speed before I installed it by testing on the bench - I used a 9v battery with a potentiometer. However, in that case I had the drive jumpers set to invert the limit switches and enable line (which this drive calls shutdown) so I didn't need to connect them.

I will re-verify by using a battery to feed the drive a voltage. If the motor moves, then that'll verify both that the drive is good and that I have the enable and limit switches wired correctly.

From my bench testing, I believe the enable LED for these drives won't light unless the drive power stage is enabled, which requires both limit switches and the shutdown line to be "on". For the limit switches this is +5v (logic low, I think?) and for the drive enable pin it's logic high (0v, I think). I had to invert the drive shutdown input already because I had it wired to provide +5v when enabled, which didn't work.

I'll check the drive DB25 "common" pin against the common ground for the frame. I'm assuming that the 7i77 ground will be the same? Right now the 7i77 is grounded through field power and its connection to the control PC... if it's supposed to have an additional ground to the frame I missed that.

Erik

Okay, I think I have most of this worked out.

I tried using a battery to move the servo by connecting it to the command inputs. It moved just a tiny bit and stopped. I ran some extra pulses of my way lube pump and things worked a bit better. I had to use a 9v battery to get the servos really moving though. It's been a while since I moved the axes on my mill, so they were a bit dry. I ran them back and forth a bit to distribute the lube.

It turned out that I didn't have a good ground on my 24v field power supply, so once I fixed that I experimented with settings. The example file I've been using for my 7i77 config had a P (gain) value of 1 in the ini, so I upped that to 60 experimentally. Then the axis would move back and forth similar to what I'd expect, although it still barely moves and does so jerkily when I have the jog speed set low (to eg. 4 ipm jog speed). I think this is partly due to PID settings in LinuxCNC and partly due to the input gain settings on the servo amps. I also might need to adjust the gib strips, although from looking at them they're not very tight.

So, tomorrow I'll be looking for information on basic settings for the PID loop values.

I also found out that my limit switches are being ignored completely, which made me want to not fool around with higher speeds yet. I scratched my head for a long time, then re-read the servo drive manual. The critical sentence there was "when the ilimit input is pulled to its active state, movement is inhibited". You can move jumpers to set the limit switches to be active low or active high. For some reason I read this as being able to set the limit switches to be NO or NC. That's not the case, I need to set the switches up with a relay to invert them or else set them all up as NO, so when they get hit they complete a circuit between either +5V (active high) or GND (active low) and halt the drive. As wired they either do nothing or always halt the drive when connected.

So, assuming I can figure that out and get some reasonable starting values for tuning the PID loop, I should be ok to get the rest of the drive wiring done and attach the home switches, plus maybe wire the spindle VFD for RS485 and wire the solenoids for the power draw bar air valves. Lots to do.

Erik

I've been referencing your site.. thanks for putting the info out there.

My amps actually have a fancy feature.. two limit inputs. Each one only restricts movement in one direction, so even with hardware limits I can jog back into normal territory

I'm thinking I'm probably going to have to install 6 DC SSRs to handle inverting these limit inputs, but I think it'll be worth it to use them. I'll run e-stop through the 7i77 since the Mesa stuff has a hardware watchdog and fails safe. My enable inputs actually de-power the drives, so it's a real e-stop.

Thanks!

One more question, in case anyone has run into it with a 7i77.

Using the suggested inputs for a manual pulse generator, how long can the cable to the mpg be?

I have a nice industrial pendant I got surplus, it has a 5 or so foot coiled cord on it that would stretch out to maybe 10-12 feet linear, and I want to mount it on a monitor arm hanging from my mill's main body, which would put a total of about 20-21 feet of wire in between the mpg and the 7i77. It'd all be shielded, with one amphenol connector set (male/female) between the MPG and the 7i77.

Is that too much, or should it work? The cable will consist of four wires from the MPG, plus two for the e-stop, plus about 14 for all the switches on the thing. I'd expect to run all that into the inputs on the 7i77.

Should I do this, or should I instead use something like a 7i73?

Thanks,
Erik

if the MPG is TTL then i'd say it should work fine , so long as the cable is sheilded and of resonable cross section
i have some running longer distances ( 20ft ) without any problem , although i did change the cable for some with larger cross section

if i was going any further distances i'd use a 7I73 , or convert to RS422