Printable View
Hello, I recently purchased a parker Compumotor Z610 for use as a spindle motor on my CNC build. I had my eyes out for a while looking for a lenze servo, or an equivalent, and this motor looked to suit the bill and then some. I may have made a big mistake, as I only recently find out the driver is more than twice the cost I payed for the servo!, Is there any possible way I can run this thing with a different driver from parker? Please advise..
Hi,
it would appear that servo is fitted with a resolver in which case there are now few choices for a compatible drive, most
drives are for encoders.
You might investigate Granite drives, if memory serves thay have a resolver to digital converter.
Craig
This looks like a very versatile drive, but unfortunately it only has a max 1500W rating. This motor is 3.34 KW. Would it be possible to run this servo with a VFD? I have heard of some people doing that, sans the benefit of positional accuracy - but should work as a spindle motor.
Hi,
Yes you can but the speed range is not great. I have a 2.7kW servo I bought as re-manufactured that has a resolver, and I bought it for the same purpose as you.Quote:
This motor is 3.34 KW. Would it be possible to run this servo with a VFD? I have heard of some people doing that, sans the benefit of positional accuracy - but should work as a spindle motor.
When I realized that the resolver made finding a drive for it difficult and/or expensive I tried using a VFD.
It works but you have a limited speed range where it works well. In my case the servo has a rated speed of 3500rpm and a maximum of 4000rpm.
A reasonable speed range for a VFD is about 10:1, so I could have my servo do 4000rpm down to about 400rpm. With tuning I got it a bit better than that,
down to about 300rpm.
In the event I decided to make my own Field Oriented Drive based on a Texas Instruments 32 bit floating point capable micro controller.
All in all it has been a long steep learning curve but it works OK. Unless electronics is your thing I hesitate to recommend that course,
in my case it became a hobby project in itself.
In the mean time I found a second hand 1.8kW Allen Bradley AC servo and drive VERY reasonably so I bought it and am using it as
a spindle motor, and very successfully to.
It might be a bitter pill to swallow but you might have to consider your servo 'effectively obsolete' by virtue of the cost/availability of a drive,
and consider buying anew but this time with a clear understanding that you need a servo AND a drive. Servos seem to outlive drives
so there are plenty of second hand servos but the required drives are as 'rare as hens teeth'.
Craig
Finding compatible drives can be hard. You need to check part number and the datasheet for exact version. I screwed up and bought a Baldor Flex+Drive 2100 watt servo drive that I thought was encoder version but it is resolver. Lucky I still can use it since I have a big Bosch servo with resolver sitting in my junk pile of motors.
The drive only cost me $100 so not a big loss.
OP - yes, this is easy.
Looks like the 610 pulls 15A stall, 40A peak.
1. Remove the resolver from the parker servo.
2. Install a CUI AMT commutation encoder on the servo. Get one with differential signals and Hall outputs. Using the AMT software, program the encoder with a low (1048) encoder count as you're going to be running it at high speed - don't want to overwhelm the drive inputs.
3a. Buy a used Allen Bradley Ultra3000 2098-030-X drive, OR
3b. Buy a Copley Controls Xenus XTL or XSL-230-40
4. Get some servo cables off Ebay with enough conductors to accomodate the AMT encoder. You'll need 11 minimum (A, A-, B, B-, Z, Z-, U, V, W, GND, and +5vdc). Neither of those drives needs complimentary signals for the Halls.
5. Get some MS connectors and get rid of those odd-ball ones on the Z610. Parker standardized on the 18-pin MS style later on, so you can find male sockets (and female plug end cables)
6. Solder the AMT encoder outputs to the connector on the servo, so you now have a 'standard' encoder output. Do the same for the power wires (and brake if you have one).
7. Buy a matching set of connectors for the drive end and finish making your cables. Suggest sticking with flying leads and a DB15 (high density) breakout board from Amazon while you're testing, but get good connectors when you're ready.
8. Hook it all up and - using the appropriate drive software - let the drive auto-phase the Halls and auto-tune the servos.
Bamma Jamma.
This works because both the Copley and the Allen Bradley were designed to take 'universal' signals. The Copley in particular was designed to use just about any servo manufacturer's motors & encoders. Neither of them are 'black box' drives.
Can't be true? Feast your eyes. That's a hall-only BLDC motor rated for 13A @ 300VDC, converted to a servo with an AMT encoder, and driven by a Copley Xenus. The AMT encoder didn't miss a beat up to 8kRPM.
https://www.cnczone.com/forums/attac...d=443632&stc=1
Once I figured this out I can now drive just about any servo out there. If it has a compatible encoder, great. If not, get a new one from CUI. I've driven Parker BExx, SMxx, N-series, Baldor BSM resolver servos, and a couple other brands no problem. If you find a 180-200V servo you like, fine. Power the drive with 120VAC instead of 240VAC.
The only thing the Copley and A-B won't do (I think) is drive serial-interface encoders - the really crazy proprietary things. Found a $50 Fanuc? Throw a $50 encoder on it and go to town.
-Ralph
Ralph - I love you. This is the exact information that I have been looking for. In my case, I cancelled the order of my particular 610 - mainly because I did not know I could do this - and it was NOT cheap. I will try again with another servo motor (hopefully for a better price), and will compile every bit of kit needed to absolutely make sure I can get it to work.
Sweet, hope it helps.
The cables and cross-connectivity are the biggest pain. The CUI encoder connector is a hyper-fiddly little thing, and they want over $100 for anything longer than a 3' cable and it isn't a continuous-flex cable. Unless you're a ninja with crimping microscopic connectors, then buy the shortest CUI/AMT cable you can and chop it. Then solder it to the backside of the standard Parker MS connector after identifying the Parker "standard" pinout.
Soldering 15 or so 26awg wires all to an MS connector sucks, but is better than trying to crimp the micro pins and getting them all in the right holes on that stupid JIS (or whatever) thing CUI has.
Or whatever servo motor brand you want to use. I've mostly gone with Parkers as their BE series have stupid high torque for a given size (NEMA34's). And the connectors are pretty easy to come by.
Whatever you do, standardize the encoder & cable connections you're going to use. Pick a servo MFGR and use their pinout so you don't have to figure out which cable is compatible with what servo/encoder combo.
And don't buy a servo where you can't easily find the manufacturer's data sheets. There are too many to pick from that if you have to spend more than 5 minutes on the internet finding all the motor data - pick a different one. If you can't find the encoder pinout then you better plan on spending $100 on top of whatever the servo cost: new CUI encoder, new CUI encoder cable, and probably a new 'standard' motor-end encoder connector.
Once you've got the cable/encoder/power pinouts standardized, it's easy to label a short flying-lead test cable and connect it to a drive with a breakout board for testing and initial tuning. The reason you want flying leads is that the Copleys and Allen Bradley Ultra3000 have different connectors for brakes.
The Copley has the brake output on the 24VDC connector, and the U3k has it on the main I/O connector. Flying leads from the motor means you can connect up to either drive and don't need a specific cable/drive combo for bench testing.
If you're shooting for a new spindle motor with 1:1 speed (no drive ratio) the Baldor BSM90-x75 series looks about the most powerful, fastest motor I've been able to find with decent data sheets available to drive a BT30 spindle. The high-speed Fanuc A06B-085x's and similar fan-cooled, dedicated spindle motors are probalby the cat's pajamas, but finding all the motor data is next to impossible. And the Parkers have lower (published) max speeds than the Baldor/ABB servos.
Kollmorgen have a few that look nice, but they're generally very expensive even used. And if you can figure all this stuff out you should be able to figure out a 24VDC or 220VAC cooling fan & shroud for whatever motor you choose.
If you are willing to do an over-drive pulley system your motor options increase dramatically. Pick your spindle top RPM, and then find a motor with a wiring configuration that gives you a 2:1 ratio at the motor's torque peak. So if you want an 8K spindle, get a motor with a max speed of 4k-5k and rated speed of 2500-3500. The lower speed winding motors can have gobs of torque down low (where you need it because of the overdrive) and you're still able to speed it up for those little 1/8" end mills that need no torque.
If you want the best of both worlds check this out (German site, use Translate): https://forum.zerspanungsbude.net/viewtopic.php?t=26643
That guy is my hero.
CAUTION: I am seriously NOT an expert on this stuff. More like a monkey banging away at a keyboard who eventually created a few legible sentences. But the combos have worked more than once for me so it's not entirely dumb monkey luck.
-R
EDIT:
The reason I suggested a 2:1 ratio is that it makes it super-easy to deal with the spindle index signal. If you run 2:1 belt drive and have a spindle with drive dogs (CAT40/BT30) for an ATC then the servo's index signal works - and both the Copley & U3000 have encoder-out signals for your controller to use. For anything other than 1:1 or 2:1 (etc) you'll need an additional spindle-mounted index sensor to home the spindle. For anything other than 1:1 (only) you'll need a spindle mounted trigger for rigid peck tapping, otherwise the threads won't start in the same orientation each time.
I was just looking at that German's work on the Fadal stylle belt setup yesterday! I would absolutely love to run such a setup, but would want to shoot for an even higher RPM, say 10,000 peak. Your information has by far been the most helpful of all my research on the matter. I very much appreciate it.
If you look at his last posts and go snooping around his instagram posts, it looks like he's planning (or has finished) a 10kRPM BT30 spindle upgrade. I can't see his instagram posts any more as I refuse to sign up, but he may have made some progress since I last checked.
That guy is brilliant and persistant.
When I saw that post I ran over and looked at my buddy's Fadal 6040. They didn't even use a servo for the spindle - just a huge 10hp induction motor. The only encoder is on the spindle itself, and once you know what the trick is you can hear it change gears around 2500RPM. I think the big Baldor induction motor is rated for about 5kRPM, and Fadal used 2:1 and 1:2 belt ratios. They also use a hydraulic system for belt tension/changes rather than pneumatic.
I serously considered doing that setup, but I punted and am just going to live with a slightly lower top end and less torque - and compensate with an over-sized motor. I just bought a CNC lathe and have to learn how to run it - no spindle belt drive project clogging up my brain right now.