[jsheerin]

I'm trying to find a way to run a 3 phase AC servo motor cheaply at a constant speed without feedback (specifically a Yaskawa 300W motor). I happened to buy a motor cheaply and would like to use it as a spindle for a hand held tool I'm making. I initially thought that because it's a 3ph motor, I could run it using a VFD. I have a bunch of VFD's sitting around, so I thought that would fulfill my low cost requirement. Unfortunately it doesn't work - the VFD tries to turn it and then shows an error - over current, I think. I searched here and found others having the same issue. All I need it to do is spin one direction. Speed control would be nice, but not necessary. Does anyone know of any other way to do this?

I've been poking around on ebay for cheap servo drives and reading manuals, but I'm not finding any that I'm confident I could make work. I've also looked at DC motors but didn't really find anything that looked promising. I'm looking for something that turns in the 1500-3000rpm range with around 300W of power and could reasonably be hand held (or maybe supported by a spring loaded tool balancer), so this servo motor seemed to be a decent choice.

Any tips, experience or thoughts appreciated.

John

[Al_The_Man]

You can use a drive in the velocity mode with 0-10vdc analogue control with a pot, but you need commutation feedback in the form of a resolver or hall effect-equivalent for BLDC mode.
There are plenty of the larger DC servo's you can use with a cheaper SCR drive such as KB or Baldor, these also have the option of tach feedback for precise speed control, if you get a servo with one fitted.
Al.

[jsheerin]

Thanks Al,
Are you saying use a VFD drive in velocity mode or a servo drive? If a servo drive, any particular recommendations?

I think I do have some sort of hall effect outputs on the encoder - they're labeled in the data sheet as u, u bar, w, w bar, v, and v bar. I'll give them a test with my scope and check to make sure.

[Al_The_Man]

Not a VFD but a DCBL drive such as A-M-C etc.
Al.

[jsheerin]

Thanks Al. I found an AMC drive on ebay that should work. I'll give it a try once it shows up and post the results.

[jsheerin]

Okay, I have an AMC drive, a B20A25. I've got my Yaskawa servo connected to it - the hall signals with power and ground, and the motor connections. I've got it set to run in open loop mode and in the test mode. It locks up the shaft, but I can't get it to turn more than about a third of a turn using the test potentiometer. It sort of flips the shaft and then stops and the current draw goes way up and then tapers off. I've tried every motor connection sequence and they're all about the same in terms of the behavior they give. I do have a green light on the drive, and it doesn't change when disconnect the motor wires and rotate the shaft by hand which I think means the hall signals are okay? I have the hall sensor phasing set to 60 degrees. If I set it to 120, I get a red LED.

Anyone have any hints on getting this working? Should the test potentiometer make the motor spin continuously?

[Al_The_Man]

If using a AMC with a BLDC mode motor then it should be set to 120deg.
If you get a red led then the hall sequence could be wrong, you need to disconnect the stator leads and power up and rotate the rotor by hand and the LEDs should stay lit, if they alternate red/grn, then the sequence is wrong.
Once they stay green then you can connect the stator, but you may have to try the 6 combinations.
60deg is usually used for running a brush DC motor on the drive using the U V connections only.
Al.

[jsheerin]

Thanks Al. That cleared up how it's supposed to work for me.

I was playing around with it some more last night, and I found out the problem was that the drive can't supply enough current to run the encoder. The drive can supply 30mA and I measured that the encoder draws about 160mA. So adding an external power supply for the encoder solved the problem - I had it wired correctly the first time. So I've got it going! Thanks for the help,
John

[Al_The_Man]

I usually use the PC 5v supply, just tie the commons only, together.
Al.

[jsheerin]

I used an old wall wart from a discman - 4.5V at something like 0.6A. It seemed unregulated - unloaded it was around 6.3V. Yep, tied the grounds together only.

[Al_The_Man]

It always puzzles me when wall warts are used, the PC has a perfectly good 5v supply, the majority of WW's are grossly unregulated, they depend on the appropriate load to supply the rated voltage.
Al.

[jsheerin]

My setup doesn't use a PC.

[Al_The_Man]

That makes sense
Al.