Pico Systems boards

Bob La Londe wrote:

I have been having pretty good luck with full plugin PCI LPT ports, but have had mixed results with onboard LPT ports. CNA still has plugin LPT ports for sale. I would check for Linux drivers though before buying. I'm still a Windows XP Pro guy so I have not checked.

Windows XP drivers have nothing to do with EMC2 motion control. The problem with the PCI parallel ports is they are faster, and the original USC board from 2002 was not tested with them. Older motherboards usually used ISA-compatible parallel port chips that would ignore small glitches on the IEEE-1284 (EPP mode) WAIT/ line. PCI parallel ports would take a small glitch as a signal to continue, and so the PPMC driver and the USC board would get out of sync. I made changes around 2006-8 to improve communication reliability between the PC and USC.

The current version of the USC board supports most EPP ports, but there are still a few that don't work. And, of course, even the manufacturer of the ubiquitous NetMos parallel port chip admits the EPP mode does not work.

"Linux driver" doesn't apply unless you want to print through it, too. EMC2 provides its own driver for direct step output or the several controller boards.

Jon

Jon, thanks for the information. I am begining to look for some drivers for some servos I just picked up and wanted to see if the Pico BLDC drivers will support these.

They are Servo Dynamics IG23CS-64-IE2000-S found here:

www.servodynamics.com/Drawing/23S-IE.pdf

It is an eight pole rotor with Hall Commutation, three commutation channels (4 Cycles/360 electrical for eight poles). I think it is a 120 degree hall spacing. What do you think?

I tried with a different model of Servo Dynamics motor a while ago, and had a lot of vibration. Apparently, Servo Dynamics' motors are made for real sine-wave drive, and not trapezoidal drive waveforms. So, I can't say for sure, but I'm afraid that the results probably won't be satisfactory.

I couldn't find any mention in their docs about the preferred drive waveforms.

Jon

jmelson wrote:

I tried with a different model of Servo Dynamics motor a while ago, and had a lot of vibration. Apparently, Servo Dynamics' motors are made for real sine-wave drive, and not trapezoidal drive waveforms. So, I can't say for sure, but I'm afraid that the results probably won't be satisfactory.

I couldn't find any mention in their docs about the preferred drive waveforms.

Jon

I wonder what motor you tried? The motor I listed is a current motor out of their product line and their current brushless amplifiers are trapezoidal. Here is one of their current brochures:

www.servodynamics.com/Specs/1525bls_spec.pdf

It stands to reason that their drives and motors should match.

andypugh wrote:

Is there any way to do sinusoidal commutation with the Pico drivers? Looking at the docs it looks like there is only one PWM signal per axis, but if that was split out into into individual phases (which I assume would be firmware and driver changes) then the bldc hal module can calculate the sinusoidal amplitudes required based on the encoder counts (after starting up in trapezoidal mode)

Well, yes, there are a couple ways, depending on the servo amp. My current brushless amp has only one PWM input, so it is not suitable. It also has no current feedback, so it can't work in torque mode.

Using an amp like the AMC/Copley drives that take two analog values, then you could use two PPMC DAC channels to drive their amp. The ABZ from the encoder would go to the encoder counter board as usual. You would have to also connect the 3 Hall signals to the digital inputs for initial commutation control. And, then use the bldc component to compute the U and V drive signals.

Some other amps that take 3 PWM drive signals could be driven with a bunch of Universal PWM Controller boards.

I'm thinking about making a new brushless servo amp, and either following the AMC/Copley scheme (2 analog inputs for torque mode) or a 3-PWM input bare PWm drive. But, this wouldn't be anything that would be available real soon.

The problem with a bare PWM drive is that is is NOT a torque-mode drive, so you have to come up with some scheme to measure motor current and feed this back to the bldc hal component. I haven't figured a simple way to do that, yet.

One scheme that does make some sense is to send the drive a PWM signal that is converted to a number in the drive's CPLD. Current is measured by a sense resistor and digitized and fed to the CPLD. The CPLD also receives encoder signals as well as Hall signals. So, the bldc component is moved into the drive, and the PWM signal from the UPC board becomes a torque command. Of course, this means that there is a servo loop in the drive, so then you get into tuning issues.

Jon

RobP wrote:

jmelson wrote:

I tried with a different model of Servo Dynamics motor a while ago, and had a lot of vibration. Apparently, Servo Dynamics' motors are made for real sine-wave drive, and not trapezoidal drive waveforms. So, I can't say for sure, but I'm afraid that the results probably won't be satisfactory.

I couldn't find any mention in their docs about the preferred drive waveforms.

Jon

I wonder what motor you tried? The motor I listed is a current motor out of their product line and their current brushless amplifiers are trapezoidal. Here is one of their current brochures:

www.servodynamics.com/Specs/1525bls_spec.pdf

It stands to reason that their drives and motors should match.

I went back through my emails from 2008, and maybe this result was not as bad as I'd remembered. I did get some significant humming from the motor at maybe 600 RPM, and I was worried about that. The motor was a
IG34CK-32-IEB8192

I am sure the commutation signals are totally compatible.

Jon