[mecheng10]

Hi everyone. First I'm new here, and I think this site is awesome. Almost every CNC search I did ended up coming to this site, so I thought maybe I should join. I'm working on doing a cnc conversion for my full size mill, and I'm just about ready to make the big plunge and order all the parts for the controller. I wanted to get some feedback from folks here to see what people think about my design so far.

I already have (3) Reliance Electric E-675 servo motors with 500 line TTL encoders mounted on them. (Not my first choice of encoders as I wanted differential encoders, but since I got all three motors with the encoders mounted for about $500 I'm not going to complain too much.) These motors are 120VDC nominal, 2.6A cont., 15A peak. They have a torque constant of 5.75 lb-in/A, and back EMF constant of 68 V/krpm.

My plan at this point is to go with the Rutex R2020 drives. I am planning on buying a Plitron 625 or 750 VA toroidal trandformer with dual 40V secondaries connected in series to give me 7.8A, 80VAC which would be rectified by a 600V, 35A cont. bridge rectifier from Digi-Key. I am planning on a 10,000uF, 250 V capacitor from Digi-Key to filter the AC output from the rectifier. I figure this should give me approvimately a continuous 110-115VDC under no load, and 5.5A cont output of the power supply. Question: how much would the voltage be expected to drop/ripple under load?

I would like to go with the R2020 so I can fully utilize the full power of my motors, but recently I have been thinking about backing off and running 85-90V so I can buy the smaller R2010 drives and save some money on the drives and a cheaper 100V cap. My question is, does anybody have a feel for how much performance I will lose. I know the lower voltage will give me a lower top speed, but am I really gonna miss it? My gut feeling is I will kick myself for going the cheap route at a later point.

At this point I am undecided as to which software to use. Eventually I would like to outfit my spindle motor with a VFD tied to the CNC software, and I like the deskcnc controller because is has a PWM spindle control output that I could interface with the VFD when I get it. However, I really like the potential of the SPI interface on the RUTEX drives because the PC I have set aside is an old P2-300, and It looks like Mach1/2 requires more than this. I don't really like the idea of having to rely on Window to send precisely timed signals to my controller. I think the SPI would be more reliable. One question is: is there any nice software that can interface with the SPI currently? I have not found any. I would like to be able to use the SPI and the Rutex tuning software in conjunction with the Rutex MB for simplicity, but would I be albe to have spindle speed control or use the deskcnc controller interfaced to the Rutex mb somehow?

How does that all sound?

Thanks for the input!

-Brian

[Kookaburra]

Another Avenue is to e-mail [email protected] (which is desk cnc) and ask about their new drive / software combination, I am sure they can help you with some drives that will suit your requirements in driving the bigger motors.

[Ken_Shea]

My gut feeling is I will kick myself for going the cheap route at a later point.

Yep, you will, kick your self that is

[rutexus]

Ken,

I don't know if you'll kick yourself or not. 87 volts is 72% of the 120 volts you mentioned. How fast will the motors run at 120vdc? The speed reduction should be fairly proportional. Keep in mind that you will not get full rated speed of the motor with a pwm servo drive. Probably around 85%. For an explanation of this, I recommend reading the motor and drive matching document found under the Rutex.com web site under 2000 products.

We have a new setup and tune document now for the R2000 series. It has not had its final editing yet, but it is definately usable. I can email it to users until I get the editing done.

There are no software controls that use the spi yet, but Vladimir is presently re-writing the cam program (pcut) to run under Windows. This program runs on the Rutex R510 plasma cutter system. It uses the spi rather than step and direction. I think the software re-write will be done in the early part of November. I think it is kind of Vladimir's premier test of the R2000 spi in a whole control environment (He didn't say that, but that's the way I look at it.) A plasma control is a good test.

That's the latest info I have.

Tom

[mecheng10]

I would be interested in the new setup and tune document when it is ready. Does the pcut software run three axis such as would be required for a mill?

-Brian

[rutexus]

Brian,

If you want a copy of the document, email me at [email protected]. I'm almost done editing it.

Tom Eldredge
Rutex LLC

[rutexus]

Brian,

Please go to the Rutex.com website and read up on the R510b. It is a 2.5 axis controller for plasma cutters.

Tom

[rutexus]

Gentlemen,

I read through the recent posts. I do not have much experience in brushless motor control, however, I have used both the Copley and the Rutex brushless drives, with other brushless motors (with hall sensors) and neither of them give these symptoms. Even though the commutation is square wave (each hall is either off or on) I have never noticed any of this jittering at the transitition point on other motors. A user in Germany just got his hall effect brushless motor (configured with a motor from Romania and an after market encoder/commutator from Italy, and it works fine with a R992, and with none of this dithering. I have used hall commutating brushless motors which had to have absolutely no following error, and obtained it with this type of motor before. I suspect that there is something unique about the commutator on these motors.

I also get the 12 position chatter, exactly like the user did with the AMC drives, so let's not get too hard on the drives.

I told Vladimir about this situation to see what he has to say about it.

I think before we throw in the towel on these motors we need to try someone else's encoder/commutator. That will involve taking at least the back end apart.

One Sanyo Denke engineer told me they are 8 pole motors, and users have said 4. I know the dithering is at 12 positions and the motors are three phase.

That's all I know at this point. I try not to say anything I'm not sure of, and as I said, my experience in this area is limited. Vladimir knows the inside of motors much better.

Tom Eldredge
Rutex LLC

[rpm mill]

I will research this but I think the Sanyo motor will want to be driven by a Sinusoidal drive, Trapezoidal drives may cause the motor to cog at high loads. I am not familiar with the Rutex R2030 drive, I think it is a Trapezoidal drive. If anyone knows for sure please let me know.

[vmax549]

Some things I have learned so far
P5 sanyo is sinusoidal
Trapezoidal drives tend to not work well with a sine motor, cogging,heat, missed steps.

Sine drives are expensive and there are not as many that take step and direction. $700-1500 ea

There is what is called a Suedo-Sinusoidal drive that works well with a sine motor but is less costly. I am looking for a reference on the web.

I found a site last night that explained it all but I can't remember where it was. As soon as I locate it again I will let everyone know. TP

[ger21]

Guys, the previous 3 posts belong to the "other" thread. I think Tom posted in this one by mistake. I can't move the posts, but try to keep future posts in the right thread. Thanks.