_
I'm in the process of wiring P5 to my Fluxeons. Here's the pinout, in case anyone needs it. PDF shows 14 encoder wires- Surplus Center servos were apparently custom made for Rockwell and differ from standard P5.
Standard 8 wire P5 is pictured below.

(Copley version is in the second PDF, so pick the one you like best..)

I think standard P5 has also 14 wires because the same cable carries also 6 hall sensor wires. There seems to be differences in wire colours though.

Wiring to fluxeon goes 1:1 with P50B05010PXKS7.pdf. Similar signal names have been used in VSD-A data sheet. Leave hall sensor unconnected and watch out not to short circuit them against each other or anything.

The PDF shows U Red, V White, W Black. My motor seem to struggle against itself when connected to pins 7, 6, 5 of the power connector. Swapping U&V returned Follow Error, output disabled. Can you confirm the wire colors on your 300W motor?

Or could it be the encoder?

What do you mean with struggle? If it works corretly, drive starts holding position after power up. Current consumption should be small and drive or motor should not produce much heat.

I tested connecting U&V wires wrong. I got strong shaking and finally follow error.

Drives are shipped with "default" configuration for P5. So if it's wired properly, then you could even jog it from step/dir without running tuning software.

Wire colors seem to be equal for all Surplus Center P5's. At least the ones in data sheet for 100W model applies also for 300W and 1kW models that I have tested.

Ok, that makes sense. It holds position and the rest is due to the PID settings.
I'm starting to really like this drive.

I have 4 300w p5 motors here, got them off ebay rather then the surpluscenter deal, and the encoder wires are 8 pin not 15. I assumed that this would be exactly the same as the surplus center ones, just with single ended outputs instead of differential.

From testing it looks like these motors don't have commutation signals, only differential encoder outputs.

Does anyone know how you get the back off these motors to get at the encoder? Its sealed so the back looks to be glued in place. Im hoping that its still got an encoder with commutation, and the wiring is all that has been changed. If not im totally screwed, as i cant afford to buy another set of drives to get around this.

motors are labelled p50b07030dcs50

also just to make sure, when im looking at commutation signals from drive they should be logic high for 1/3 of the rotation, and logic low for 2/3 right?

If it has quadrature encoder outputs, you're lucky (assuming that you have Fluxeons :). Commutation (hall sensor) signals are not needed or used by VSD-A.

Commutation signal has 50% up and 50% down during rotation just like quadrature signals have. Commutation signal has only few edges per revolution while quadrature has hundreds or thousands.

actually im quite unlucky, as i dont have any of your nice drives, as i was hoping to make do with some amc ones until i scraped together enough cash to upgrade.

i dont have a scope here, so my testing method is a bit creative. That said i have identified C and !C lines, so it must be wired as a differential encoder without commutation.

My choices now seem to be either find a way to get the back off and hope the encoder actually has commutation, and its just a custom wire (only one encoder listed with this range on sanyodenki, and it has commutation tracks).

That or buy 4 VSD-As now, which would be a nice solution if i had that kind of cash lying around. Also i had planned to be doing just torque control on the drives, and closing the control loop in EMC with a mesa interface board. The changeover kind of ruins that idea.

ok i give up, the only way im getting into these motors is with a hacksaw! managed to get a 100w apart and the encoder is not exactly modular, as it looks like you have to unsolder the lightsource to get the codewheel out. Even if i could swap encoders i have no idea how to align the commutation tracks.

Some Sanyo Denki motors have a "combined" encoder/commutation output, where the commutation signals are output for about a second after power is applied to the encoder, then the outputs are switched to the encoder signals. I don't have a data sheet for those motors handy, but you should be able to google for it. It will make things clearer.

Hey Walter,

How are you powering those motors?
I'm assuming you are talking here for the 1000w sanyo motors. I have 4 of those and also planning to buy some fluxeons but I can't find any reasonable priced PSU.

P.S. We are on 220v/50 Hz here.

Darik, you probably need much less power than the sum of motor power. In average application I would reserve 300-500W continous power per motor. Use transformer based PSU because they can deliver high peak power just like motors can.

Darik,

This power supply came from Ebay guy johnango. You can use it on 220v; toroid is 30lbs, I measured 156vdc on the drive. Check him out, you can buy parts and make your own (vs $300 assembled).

btw, these drives are sweet!
_

Walter, that must be the cleanest design DIY control box that I've seen so far :) Really nice work.

I noticed that there are some bent power stage pins in the middle drive, check that pins are not touching each other before powering up.
Clean compressed air might also be good idea for removing any metal chips that might have come from drilling.

Good catch, thanks!

Walter,

What is the VA rating of the PSU you are using?
I've checked johnango's site (http://www.toroid-transformer.com/) and his power supply's go up to 1500 VA. I'm planning to add a 4th axis later so if I reserve 500W for a motor I'll need 2000W. From VSD-A data sheet I've estimated that I'll need at least 3kVA to run safe 4 motors. Is that right or I can go with 2kVA one?
Also did you measured any voltage drops under load?

@ Xerxes,

I'm a bit confused with the power supply voltage calculations after I saw Walter's PSU voltage rating. How is the calculation for an AC motor done? Those Sanyo's are rated at 100VAC.

1. Should I calculate DC or AC voltage at the motor?

100VDC needed/88% effective voltage swing ≈ 114VDC at the drive?
or
100VAC needed/88%*1.41=160VDC at the drive?
Can you give an example?

2. Is there any breakout board you can recommend? Could also be DIY one.

Darik, the VSD-A datasheet is very rough about power supplies and talks about worst case load. In practice you probably need a fraction of the worst case power. I hope Walter would publish some real world power consumption data after he has his machine running :) I would bet it to consume less than 800W power in average.

Sizing PSU for servomotors is not exact math. Even voltage requirement can be difficult to estimate. For instance I have 300W and 1000W sanyo P5 servos which both are 100VAC and 3000 rpm but the 1kW model seems to need nearly 200VDC to run at full speed while 300W runs happily at 150VDC. There shouldn't have been any difference but there is because motor voltage ratings seem to be very loosely specified.

The latter of your equations gives a quite good voltage estimate.

Darik, it's 1.5KVA, 170Vdc 9A cont.

Xerxes, I'll post something as soon as I can get it going. Appreciate all the help and info here on these threads- good job guys!