What's new
What's new

Yuasa Rotary/Controller Help

seahuston

Plastic
Joined
Nov 12, 2015
Hello,

Wondering if anyone has experience or knowledge with Yuasa rotaries and controllers. I got a used DMNC 170 rotary with a Yuasa UDNC-M1 controller recently and I'm struggling to get it to position properly. Mechanically it seems okay, I've replaced the oil and the leaky brake solenoid. Without the motor attached, the gear turns smoothly.

My issue is that the table appears to be always undershooting it's programmed position. Example: G0A90 results in a position of ~89.578deg. This prevents the controller from registering the move as complete and moving on. If I grab the table and try to rotate it a bit the controller will read the final position a move on.
I assume this is a parameter issue but unfortunately the Yuasa documentation is pretty poor and does not cover the motor parameters. I've not had much luck getting help on this issue from Yuasa USA.
If anyone has dealt with something similar or has a setup that could share parameters I'd really appreciate it.

Thanks!
 
Not sure if it is really a parameter issue, or what.

I have a spreadsheet attached. there are 3 sheets, so if you toggle between sheets, the first one shows a couple 170's parameters that you can compare to.

 
The Yuasa manual for the controller does have all the parameters in it for all sizes of rotaries.

That is a strange problem. What happens when you command different angles? When you send it home?
 
The Yuasa manual for the controller does have all the parameters in it for all sizes of rotaries.
While mostly true, there are way more parameters in the m1/m2 boxes than the manuals show.

Most should not be changed and would never need to be changed. But that doesn't mean that the ones in his box haven't.


I'll also mention, even though not really relevant, that yuasa no longer supports the udnc m1/m2 boxes, in case you ever need the drive/boards replaced. They have a new "nmc" box that is touch screen.

By the way, after reading some notes, the only thing I would perhaps try first parameter-wise, the in-position band, i128
 
While mostly true, there are way more parameters in the m1/m2 boxes than the manuals show.

Most should not be changed and would never need to be changed. But that doesn't mean that the ones in his box haven't.


I'll also mention, even though not really relevant, that yuasa no longer supports the udnc m1/m2 boxes, in case you ever need the drive/boards replaced. They have a new "nmc" box that is touch screen.

By the way, after reading some notes, the only thing I would perhaps try first parameter-wise, the in-position band, i128

I've had the drive mosfets go out. It was a straightforward job to find equivalents and replace them.
 
We've got a few M1 boxes that have been fairly reliable, way better than the UDNC-100 boxes. Transistors failing on udnc-100's is a common thing. I've got 2 indexers udnc-100 powered, but I believe I have 6 "working" boxes and several parts boxes along with several huge cardboard boxes with spare boards (years ago we used to service yuasa indexers and boxes)

I have an M2 box that had several components and runs burn out on the board, and yuasa didn't have any spare parts or a schematic of what was missing/charred, so they sold us an NMC box.
 
Wow that spreadsheet was hugely helpful, I never found Yuasa documentation with this much documentation. In case the problem wasn't clear, the rotary table undershoots angle for all programmed positions. My assumption is that the PID loop is overdamped and never actually getting to position.

I played with the parameters a bit. It does seem like i128 is a helpful factor. Increasing this to your values (from 107 to 500) made a big difference. I also increased the proportional gain and decreased the derivative gain (assuming the system is overdamped). This has it running pretty reliably on bench tests. Small angle moves still have a slight undershoot. I'll try to tune that a bit further.

I'm now seeing that the i128 variable change does not carry over in a power on/off. The PID factors do but i128 keeps resetting to 107 if I power cycle. Any experience or thoughts on that?

Thank you much!
 
Not sure if it is really a parameter issue, or what.

I have a spreadsheet attached. there are 3 sheets, so if you toggle between sheets, the first one shows a couple 170's parameters that you can compare to.


Could you break down with the "Beere Test 1" indicates? Is that the values that you adjusted to in order to get proper performance out of your unit?
 
I'm not really positive. I wasn't the one who did that. Our "beere" indexer was something like this: https://pawsworkholding.com/beere-collet-systems/beere-4-station-5c-multidex-360-tilt-rotary/

It was running off a modified M2 (2 axis) box (modified to power the hydraulic brake clamp for the 5c units). I don't know if "test" means we were tweaking it to get it to work right, or what. I would say that one of the axis (not sure if 1st or 2nd) was the "beere axis" with 5c collets, while the other was a large yuasa indexer, I think a 220?

We still have that beere unit but haven't used it in a bit.
 








 
Back
Top