[johndmcmaster]

I have an XY table with DC servo motors consisting of RSK MSA6703 encoders. The motor is a I'm trying to figure out my options for a COTS control system as opposed to a DIY option.

One option might be to use Gecko G320X drive using comparators to convert the sine wave outputs to quadrature encoding. Does anyone have experience with this? Is this easy/feasible? I've seen a few hints to this regard on some threasd here but was hoping someone with more experience could confirm this before I blow my beer money.

But I didn't post this in Gecko section because I'm also interested in alternatives. Anything else I should consider? It will be either drive from linuxcnc or something like that.

Some additional details can be found here for the curious: mcmaster:tech_inst:k2_ind_zygo [Silicon Pr0n]

[mike_Kilroy]

what is it you want? Your wording is mixed up and makes no sense. you want a new motor? mew drive? new cnc control? new encoder?

Do you need specs on your RSF encoder?

What is your question?

[johndmcmaster]

I have DC motors and RSK MSA6703 sine-cosine linear encoders. I do not have a control system for it and need to build one. I am looking for suggestions on how one might build a unit that can be controlled from something like LinuxCNC or a microcontroller. For example, a unit that accepted step/direction pulses would work well but something that supported a serial protocol would also be acceptable

I did some more research and am now fairly confident that the gecko option will work with some reasonable electronics to convert the encoder signals. It looks like I have two options:
-Use schmitt triggers to convert the sine waves to quadrature square waves
-Use an interpolator for better precision. These typically output quadrature square waves

I have some specs on the encoders. The cables are cut so I will need to figure out the pinout on the bare cable. I think I should be able to figure out sufficient specs to integrate them.

[Al_The_Man]

If these are quadrature sine wave output, you could look at using a differential driver if to square them up to 5v, if the amplitude is enough to input the driver.
Do you know the resolution of the basic sine wave count/rev?
Al.

[mike_Kilroy]

We were RSF reps for about 20 years until Heidenhain bought them. Attached is spec sheet on the MSA6703.

NOTE: it is NOT typical 1Vpp sine output of today; it is 11ua output to emulate the old Heidenhain units that used EXE boxes. You will have hard time reliably converting these outputs to anything usable by your drive.

Check Ebay for "exe BOX" or similar words and buy a converter. They came in 1x, 5x, 10x, 25x maybe higher, multiplier of input 11ua sine and output TTL or 1Vpp sine - get a TTL 5v output model. Prob can be had cheap.

Attached is pdf data sheet on that unit with wire colors listed.

[johndmcmaster]

Thanks for the info, that helps a lot!

I picked up this guy: http://i.ebayimg.com/00/s/NzY4WDEwMjQ=/$(KGrHqV,!hcE6boO-qU4BOw74Y3EZw~~60_57.JPG

And will give it a try. The 610 is 5x or 10x multiplier. In my similar microscope (I haven't characterized this one yet) the smallest field of view is 108 x 80 um with probably about 5% error acceptable (say 4 um). So at least 5x (2 um) should do it, 10x (1 um) will be great.

Ordered gecko drives and the interpolator, waiting for stuff in the mail now. I'll update once I get the stuff and try to hook it up.

[mike_Kilroy]

exe-610 is good unit

[srgvorob]

I have two encoders from the company RSF-Electronik
Linear Encoders MSA 654RI and MSA 322RI , can anyone tell me how to replace them with HEIDENHAIN encoders?