Anyone got any tips for tuning servo motors?

[Darc]

I haven't actually used the tuning part yet but in the next couple of weeks I'm going to put a couple of Allen Bradley servo motors and drives on to our 6 axis machine.
I would just like to understand the procedure before I tackle it.
Because it's a tuning function, does this mean it actually changes settings inside the drive to run the motor more efiicently or does it just change the output signals?
Because the machine runs steppers (very slow I might add) I'm a bit worried I'll need to put a gearbox on the end of the motors to compensate for the difference needed (but I'll work that out later).
One of these motors (MPL-A330P-HK22AA) I've got says a maximum RPM of 5000, can these run quite stable at a low speeds e.g 250 - 500rpm (I'm only guessing until I remove the old stepper motor) or should I really alter the ratio to allow the servo to run near it's maximum rpm when running flat out?

[Al_The_Man]

The amplifier drives should be set up in the torque mode and the gain as high as it will go without oscillation, something most don't seem to look at is calculating the motor to load inertia, at least if you use one of the software sizing packages you can get a feel as to what the effects are on motor to load inertia, play around with free programs offered by Kollmorgen or Allen-Bradley, I have posted a link to the emerson one in the utilities section, I assume you have a Galil card seeing as you posted in the Camsoft forum.
If you view the on line video by Jacob Tal on the Galil site, this will give you an idea of tuning the system control loop with the Galil auto-tuning software.
Al.

[keithorr]

AFAIK:
The tuning function doesn't change the amp drive settings. The tuning on the camsoft side only changes the way the +/-10v signal is determined. You should ask camsoft direct about their features. There is an autotune function. I couldn't use it as my encoders are attached to the axis, not the motors. You can also get the Galil tuning software for about $150US.

As to servo rpm speed, there is the common accepted idea that they run better at a higher rpm. However, just for kicks, I took the end caps off of my 1.3kw Sanyo Denki servos just to see what they were doing. On many of the moves, the motor shaft is running much slower than 1rpm and the machine shows no following error and is very smooth, no cogging. The servos, amps and machine are all stock, I only added CamSoft and an ethernet card.

[Karl_T]

Just a heads up if you use Galil's WSDK or their other software on your Camsoft computer. Galil will install drivers that cause Camsoft to not run. AND the drivers from both companies will not uninstall. I ended up reformating and starting over.

If you are using XP you could set up a system restore point and roll your computer back to a prior state after using Galil to avoid this mess.

Karl

[Darc]

Thanks for the heads up

[intrusion]

It's not that hard really to get both programs to work on the same computer. I noticed that if you are using the latest version of CamSoft then install Galil software the drivers installed by Galil are no longer any trouble. If you have an older version of CamSoft then the procedure if pretty simple, you basically call or e-mail CamSoft and they will give you a list of driver files to delete. You install the Galil program then delete the driver files then re-install CamSoft to put the proper drivers back in place then both programs will use the same set of drivers.

[Darc]

Well, I finally got a chance to install the servo drive, (I don't think I've got the settings correct yet, e.g in the start up file).
When I physically grab the motor shaft and turn it the DRO changes, I just can't get the motor to move itself.
When I Auto tune it, the motor doesn't turn but it gets to 48% and sits there for about 40 minutes, I aborted after that, should the motor be turning during the tuning process. Also how long should it take to tune the motor?
I've attached ther terminal PDF (with a few changes).
I think I'll have a few questions over the next couple of days.

[Al_The_Man]

Take a look at the manual and see what is required for pin 49 and 50 it is labeled brake or drive enabled, you may have to jumper this out depending on what the book says to enable the drive, you won't need to hook it up to the ICM2900 to try it.
It should take a couple of minutes to auto tune the drives.
Al.

[Darc]

All the below are terminals for the drive
Pin 49 & 50 are for the brake +/-, but my motor doesn't have a brake.
I did notice that pin 20 is Drive enable, when you say jumper this out how do you mean? (I labelled it on page 4)
After reading the manual (see attached) I think your right, I'm just not sure where to connect it.
Also Pin 24 & 25 are for drive ready +/-, not sure if this is of any relevance.
I attached the PDF.
Thanks (Yet again)

[Al_The_Man]

Ok I see it now, Pins 5 & 26 and 6 & 13 are the power source for operating remote switches or opto etc as shown in fig 6.1, 6.2, 6.3 & 6.4 for now all you need to do is to take pin 20 to pin 5 or 26 and that should enable the drive, later you can operate this with a relay or opto etc if you wish from the ICM2900, also the brake enable output can indicate if the drive itself is ready or in fault when fed back to the system, again not needed for testing.
Al.

[Darc]

Thanks Al, I just tried it and it's still not working, I'm not sure if it's hardware (second hand drive) or software, is it possible to connect a power source to the drive to make the motor move (just to ease my concerns, incase the drive is knackered).
Also what is the DCBus + & - terminals on the drive (yet again, see attached).
Your patience is incredible......

[Al_The_Man]

The DC bus comes from the rectified AC input,I would assume this feature is if you want to use an external DC source instead of the AC mains.
If the drive is powered from an ac source you should measure the resultant dc bus voltage, you should be able to test the drive in a stand-alone condition by enabling it and place a variable DC source into the command +- , change polarity and the direction should change, I use a battery box with a pot to give me +9 -0- -9v but any DC cell from 1.5 to 9vdc into the command input should produce an output to the motor.
When you test it hooked up to the ICM2900 and command a move, there should be a + or - voltage on the command input.
Most drives have a LED to indicate when the drive is in the enabled state.
Does your drives have the 7 segment status LED that will tell if the drive is enabled?
Al.