config axis

[Wolfenstien]

I'm making some progress over in the thread called

"Mori Seiki SL1H Oak Board Build"

Just don't want to keep double posting.
Thanks,

[S_Tully]

Marty,
Sorry just been busy, I spoke with them a couple months ago trying to figure out where the new drives were. I got them in about a month or so ago and just got them installed a few weeks ago and found they do work a little better but there are still issues with the encoder signal from the drive to the oak. They say to set the GEAR_NUM to 8192, LINE_NUM to 4095(4095 makes no sense but i went with it) and Centroid counts to 32768. Since the GEAR_NUM is set to 8192 it needs to see 32768 which is fine since Centroid will end up sending 32768, where the problem lies is the LINE_NUM being set to 4095 where according to their manual it will send out 4 times this value which would be 16380 which is not right since Centroid needs to see what it sends out! The weird thing is that with the LINE_NUM set to 4095 Centroid is seeing roughly 32658 which is 7.98 times the value in LINE_NUM and still not what Centroid is needing to see and not what their manual says it will do. The problem before was The dyn4 drives were pulsing the encoder signal which caused quad errors almost instantly, now i am able to jog around a bit longer before this happens but it still happens. I wanted to try some other settings before contacting them back. Since the highest value that can be input into the LINE_NUM is 4095 i wanted to try an lower the Gear_NUM and Centroid counts to see if that will work. They said they did trials but not looking like they did some good trials. They sent a video of their initial test and seemed ok but that was with lower counts and i noticed they still got a quad error right at the end. Not sure when i will get back around to messing with it as i am busy with several custom machines. Hopefully i will get to try some other settings soon. Sounds like you are going to be experimenting with some soon. How new are the drives Have they had the updated firmware installed on them? If you have any questions feel free to message me! I will try to get back to you ASAP!

[cnckeith]

4096 is the number you are looking for

(1024 x 4 = 4096, 4096 x 4 = 16384, 8192 x 4 = 32768 etc..)

[martyscncgarage]

Marty,
Sorry just been busy, I spoke with them a couple months ago trying to figure out where the new drives were. I got them in about a month or so ago and just got them installed a few weeks ago and found they do work a little better but there are still issues with the encoder signal from the drive to the oak. They say to set the GEAR_NUM to 8192, LINE_NUM to 4095(4095 makes no sense but i went with it) and Centroid counts to 32768. Since the GEAR_NUM is set to 8192 it needs to see 32768 which is fine since Centroid will end up sending 32768, where the problem lies is the LINE_NUM being set to 4095 where according to their manual it will send out 4 times this value which would be 16380 which is not right since Centroid needs to see what it sends out! The weird thing is that with the LINE_NUM set to 4095 Centroid is seeing roughly 32658 which is 7.98 times the value in LINE_NUM and still not what Centroid is needing to see and not what their manual says it will do. The problem before was The dyn4 drives were pulsing the encoder signal which caused quad errors almost instantly, now i am able to jog around a bit longer before this happens but it still happens. I wanted to try some other settings before contacting them back. Since the highest value that can be input into the LINE_NUM is 4095 i wanted to try an lower the Gear_NUM and Centroid counts to see if that will work. They said they did trials but not looking like they did some good trials. They sent a video of their initial test and seemed ok but that was with lower counts and i noticed they still got a quad error right at the end. Not sure when i will get back around to messing with it as i am busy with several custom machines. Hopefully i will get to try some other settings soon. Sounds like you are going to be experimenting with some soon. How new are the drives Have they had the updated firmware installed on them? If you have any questions feel free to message me! I will try to get back to you ASAP!

I think line number is a typo. As Keith suggested, try 4096.
I'm getting a back panel wired up with OAK and DYN4's myself. Will do a video on how the benchtesting goes.

Let us know how it works out.
Marty

[S_Tully]

Marty I originally thought it was a typo as the number they gave wouldn't have been incorrect but when trying to put in 4096 ( was what I used the first time) the drive defaults it to zero when writing to the drive, I didn't catch it till I read the drive after having errors in centroid so I double checked the drive settings and it had save the value to zero. So I lowered all encoder counts to what they had in the test video DMM shared with me and I still had the same issues. I knew going into the project that they had encoder issues but wasn't worried about it as I had the servos on hand from an acorn setup and figured I would try and work with them or find a way to get them to work. If someone is looking to get a machine up and running in any kind of hurry I would recommend just breaking down and buying the estuns or yaskawa's as they work extremely well with the oak boards, I always use the yaskawa's and never an issue. I just thought hey what the hell I will slap the dyn4's on just to for fun.

[martyscncgarage]

I am just getting OAK and DYN4 going on the bench and working through some similar issues.
Will report back once they are resolved.

Marty

[martyscncgarage]

I have DMM DYN4's and OAK running in lathe mode on the bench. They have to add a drive reset input on DYN4 and also the ability to invert the drive alarm output. More testing to be done.

Marty

[burke_wl]

I have DMM DYN4's and OAK running in lathe mode on the bench. They have to add a drive reset input on DYN4 and also the ability to invert the drive alarm output. More testing to be done.

Marty

Hi Marty,
Did you happen to find a fix for the Oak/DYN4 encoder quadrature error problem? Im currently working on a lathe build with an Oak and DYN4 drives and have encoder quadrature errors. I can jog the X axis and get a Z axis quadrature error. Also occurs on the axis that is moving.

[martyscncgarage]

I have DMM DYN4's and OAK running in lathe mode on the bench. They have to add a drive reset input on DYN4 and also the ability to invert the drive alarm output. More testing to be done.

Marty

Hi Marty,
Did you happen to find a fix for the Oak/DYN4 encoder quadrature error problem? Im currently working on a lathe build with an Oak and DYN4 drives and have encoder quadrature errors. I can jog the X axis and get a Z axis quadrature error. Also occurs on the axis that is moving.

I have been on vacation for 3 weeks. Recently returned. Before I left, DMM shipped me the new drives/cables with the drive reset input and alarm inverted output. Its been awhile, but I do not recall any encoder issues. Did you set them up for Precision mode?
I am swamped and likely wont get the drives and data cables swapped out until next week. I intend to do a follow up video once done.

Marty