Trouble testing PMDX 126 with Gecko 320X

[Jackal66]

I have my Gecko 320x drives connected to my PMDX 126 board.

The 126 board has a test feature where you can check the Geckos before connecting them to Mach 3.

I have all of the dip switch test settings correct on the 126 ( I think).

The encoders are AMT 103, with the dip switch setting to 500 counts.

As soon as I turn on the power to everything, the Geckos go into fault.

I have one axis ( the X) motor armature connected.

If I push the button a couple of times the X-axis motor with spin at 90 degree steps @ 1 revolution per second. After about 2 seconds it stops.

The LED's on the 126 board flash just as the manual said they would.

About 4 steps per second, and then constant steps.

My thinking is it's the way my Geckos are adjusted (or lack thereof).

The Geckos are set at the factory ( default) settings.

The Gecko manual says to have a jumper from the #5 ERR/RES terminal to the #7 ENC + terminal for testing.

Mine isn't this way. I went ahead and hooked up the #5 ERR/RES terminal like it showed in the PMDX manual for running purposes.

Could this be the problem?

I have read about installing a pull-up resistor & capacitor on the Gecko's, but not sure this is the case.

The #10 common terminal on the Gecko, is connected to the PC-GND tab, but there is an option to connect it to PC +5v.

On my own, I don't want to just start experimenting and changing wires around, without theory and reasoning behind it.

Thanks for any & all advice.

JAckal

[Al_The_Man]

I haven't used those products so it is a shot in the dark, but try swapping the motor armature leads over ( I assume they are DC servo's?).
It sounds a bit like encoder motor direction/phasing is reversed.
Very quick to try anyway.
Al.

[morestuff]

Hi Jackal66,

The instructions in the Gecko manual regarding initial
encoder and motor phasing should be followed before
trying to use the self test on the PMDX-126.

The purpose of the connection from terminal 5 to
terminal 7 on the G320X is to reset and enable the
G320X. This can be a momentary connection, in
which case it will need to be repeated after each
fault; or it can be a continious connection in which
case a fault condition (such as following error) will
cause the drive to continiously reset and try again
at a rate of once every couple of seconds.

The RST and ERR connections on the PMDX-126
serve the same purpose, but with the ability to
use Mach or other software to issue a reset when
the operator desires. See our application note here:

http://www.pmdx.com/Doc/AN125-2_GeckoRestart_01.pdf

It is possible to have a momentary push button between
terminal 5 and terminal 7 of the G320X drivers in addition
to the connections to the PMDX-126, they both do
the same thing and no conflict will occur.

Servo faults on initial setup are most commonly due to
phasing of the encoder A and B signals versus the
motor connections, and because of unsuitable tuning.

Tuning a servo motor with no mechanical load on it is
sometimes touchy and will definately be much different
from the tuning required with the load connected.

Also, any G320X that receives step pulses while the
motor is not connected should fault because the motor
does not follow the commands.

If you need additional help, feel free to contact us
by email or phone.

Regards,
Steve Stallings
www.PMDX.com

[morestuff]

OK, the Yahoo URL magic fails yet again. To see our
application note, please cut and paste the URL into
your browser.

http://www.pmdx.com/Doc/AN125-2_GeckoRestart_01.pdf

The note will be titled PMDX-125, but it applies in the
exact same manner for a PMDX-126.

Regards,
Steve Stallings
PMDX

[Jackal66]

Thanks guys.

I wondered about going through the Gecko initial set-up before connecting the motor armature and using the PMDX test.

It looks like I need to.

Steve,

I have my ERR/RST wired like the link you provided.

That made me think that the PMDX test could be just for testing existing systems that had already been tuned, or had been through the Gecko start-up procedure.

I tried to cheat, thinking that if it was all wired up, the "Beginning Gecko" steps could be skipped.

Guys,
Thanks again for the quick responses.

JAckal

[Jackal66]

Well,

I got a little farther along.

After temporarily disconnecting the ERR/RES wiring to the Geckos & 126 Board, I went back the the Gecko manual.

With a jumper from #5 to #7 terminal, the lights on the Geckos do what they should when you move the motor shaft.

With Mach3 sending steps, the B_O_b showed activity on its lights.

The Geckos never faulted like the were supposed, too.

They have continuity to the B_O_b and I even switched the step & direction wires, to see if that was it.

After hooking up the armature of the motor to see, it only moved slowly in 90degree increments, in one direction. Never faulted either.

Do I need to adjust the following error or something?

There is something I am missing.

It will be stupid simple when someone points it out, too.

Thanks,
JAckal

[Jackal66]

I figured it out.

When starting up Mach3, I was loading an .xml called "Hurco".

This was supposed to be the .xml file from PMDX.

It wasn't. This was one created back in March when the retrofit started.

If found the PMDX sample .xml file and after it was loaded, everything works.

The step & direction wiring terminals were different.

When I was trying to get X to move, my signals were going to Y or Z.

Also, the armature wires + & - had to be switched as Al suggested.

Maybe this will help someone with the same problem.

It was stupidly simple.

Thanks for the help,

JAckal