Some questions on Rutex R2010 drive & tuning

I got a few R2010's used off ebay after initially expecting to use gecko drives on my mill, which I've set up a power supply for 75 VDC with a 1500VA Antek toroid, 4x 60A bridge rectifiers (2x per side of the transformer secondary in parallel), 2x 56000 uF capacitor bank for the motor drive side (hey they were cheap).

I then bought some R2210 interfaces, hooked to HEDS 5500 encoder on the Electro-Craft motor I'm trying to figure out tuning on first, well second I guess. I've built a tuning cable as described in the R2000setup document, had some problems with the parallel port on the PC I was using it with (BTW you should revise that document for that - turns out PS/2 or AT modes in the bios on my Dell don't like to work with SPI & your drives, but ECP mode worked fine.

First question is to answer some ambiguity - the R2000setup document says the R2010 needs to see .03 ohms/volt or a braking resistor needs to be installed. Since 2 of my motors are old and I can't get data sheets on them, I measured one and came up with 4.8 Ohms, subtracting 1 ohm for my DMM I get 3.8 ohms, if I calculated correctly the R2010 needs to see 75Vx .03 ohms = 2.2 ohms, if my servo's number (3.8) is more than that do I need to do anything or am I ok to run it as is? Also do you have any reference as to how to build a good shunt regulator as referenced in the R2000faq document?

First problem is that I have a drive/encoder/encoder interface that I attempted to hook up to another motor that I purchased to replace a missing one on the mill, this motor had a CUI 103 encoder fitted, regardless of swapping wires or checking swap boxes on the tuning software the motor slowly runs away, at low speeds the encoder seems accurately counting but if I spin the shaft faster (with motor disconnected) the tuning software is clearly missing a lot of encoder counts. I haven't put my scope on it but suspect a noise problem, does rutex have any recommendations or can/should I try what Gecko suggested for HEDS encoders here: Using HEDS Encoders - Servo Drives - Application Notes - Support ?

Next is tuning. The second motor I tried, an old electro craft with a HEDS 5500 seems to be working as it should but I'm not clear on what I'm aiming for as far as a tuning end result.

I read the R2000setup document beginning to end, and the tuning part several times. I wish your website would've stated that the R2xTuneVB6 program allows you to adjust settings at the same time you watch the tuning graph, I figured that out the hard way.

I started by increasing servo loop time to 2 ms, increase Kp til I get oscillation, play with Kd and Kd index till I get better damping so I'm not oscillating when steps are commanded. Now the R2000setup doc tells me to increase Ki till I get overshoot? How much overshoot should I be aiming for? I notice that I don't get the same amount of overshoot depending on direction the motor is rotating, is that a problem? How fast is the software generating pulses when it commands the servo to move however many steps are in the box on the tuning graph window?

Also I notice the "ideal" tuned graph has a very steep ramp up to the commanded number of steps, even when I increase Kp quite a lot the incline of mine is still pretty tame, is that a function of the size of the motor and speed at which it is capable of accelerating or is there more to it than that?

Thanks for answering all my n00b questions I really hope to be making chips soon!

Did you ever get a response on this or try contacting them directly? I have some drives that I bought new a few years ago that I have installed but not wired yet on a new build. I've tested them with no problems.

Quote:

---

Originally Posted by kevincnc

Did you ever get a response on this or try contacting them directly? I have some drives that I bought new a few years ago that I have installed but not wired yet on a new build. I've tested them with no problems.

---

I just called them and talked my problems over - I'm a lot more comfortable now than I was when I made the OP.

For anyone else who is wondering - if your motor measures out above the .03*(input voltage DC) ohms, you don't need any sort of braking resistor

They warned me that they've had problems with the CUI capacitive encoders like my 103 so I think I'll just be replacing it with another HEDS 5500 unit as is on my other motor that I have been able to tune.

You do need to have some overshoot, because of the (un-attainably accelerated) speed at which the tuning program commands the motor to accell and decel. With reasonable accelerate/decelerate values in Mach or EMC and coupled to a load, the motor shouldn't be overshooting significantly in real world applications.

I wish more of this info was volunteered across the multitude of documents from Rutex that I've read, but at least I got to the bottom of it.

Hello Rutex users:

I have not been over here to check this forum for a while.

The best way to get help for using Rutex products is to call me on the phone. (573) 341 1528. I have been answering questions over the phone for over 10 years now, supporting Rutex products. I believe I can give much better answers over the phone than on line. That's why I prefer the phone. It allows me to ask questions and give answers much more quickly than on line.

I'm in Missouri, Central time zone, USA.

Otherwise, you can contact me at [email protected]. I check there normally twice a day on week days. However, I like phone the best.

I did answer the questions the fellow who posted today when he called me today, and I gave him several tips for tuning too.

Tom Eldredge
Rutex LLC