Yaskawa encoder hack
- JR1050
- Topic Author
- Offline
- Platinum Member
Less
More
- Posts: 450
- Thank you received: 33
21 Dec 2024 15:35 #317089
by JR1050
Yaskawa encoder hack was created by JR1050
Hello there. I have a couple of mills with Yaskawa servo systems, mid 1990 BE series analog drives. They are junk, constantly blow up, very little info available, only support from Yaskawa, send us your life savings, in my case about $5 , so not enuff anyway.Using another brand of drive should be doable, the wrinkle here is Yaskawa’s unknown initial magnet sensing. The motors are sinisudialy communtated, sorry for the misspelling. The oldest ac yaz motors used what I suspect is a coded type hall sensing for initial trapezoidal communtation. The output signals were v,-v,u,-u,w,-w. , output from the encoder to the drive.The new motors have the markings on the encoder for the added signals, they are unused. I suspect the circuitry is there. Both encoders have on the glass disk , an inner ring of what I guess could be grey code. See pictures.Does anyone recognize this and have a hypothesis about how these signals would work? My guess is they are similar to the red cap Fanuc bldc signals. I have not put a scope on them
Attachments:
Please Log in or Create an account to join the conversation.
- jmelson
- Offline
- Moderator
Less
More
- Posts: 817
- Thank you received: 151
21 Dec 2024 16:47 #317094
by jmelson
Replied by jmelson on topic Yaskawa encoder hack
It looks from the PC board that there are 12 outputs from the encoder. A B quadrature, plus index (Z) and U V W commutation, and the differential complement of all of these. That should be enough to run any industry standard servo drive. AMC, Servo Dynamics, etc. have brushless drives that can take encoders like that. Some don't need the UVW complements.
Fanuc had a different scheme for commutation on the older red cap version, the later ones are serial, so there is no ABZ signal at all.
Jon
Fanuc had a different scheme for commutation on the older red cap version, the later ones are serial, so there is no ABZ signal at all.
Jon
The following user(s) said Thank You: JR1050
Please Log in or Create an account to join the conversation.
- JR1050
- Topic Author
- Offline
- Platinum Member
Less
More
- Posts: 450
- Thank you received: 33
21 Dec 2024 18:10 #317097
by JR1050
Replied by JR1050 on topic Yaskawa encoder hack
Thank you. I have AMC amps for this application, these ones require the communtation signals until it sees the index. The thing that threw me off was the signals were differential, I’ve never seen , and I’m not an expert in this field, dual communtation signals. I take it , that I would just pick either the u,v, w or the -u,-v and -w . The AMC amps I’ve used allow for one set of halls signals. Thank you again.
Please Log in or Create an account to join the conversation.
- jmelson
- Offline
- Moderator
Less
More
- Posts: 817
- Thank you received: 151
21 Dec 2024 19:51 #317103
by jmelson
Replied by jmelson on topic Yaskawa encoder hack
Yes, that is common, just use the UV and W, and don't connect the inverted ones.
Please Log in or Create an account to join the conversation.
Moderators: PCW, jmelson
Time to create page: 0.076 seconds