Velocity vs position mode

All things related to Centroid Oak, Allin1DC, MPU11 and Legacy products

Moderator: cnckeith

Post Reply
occustoms
Posts: 129
Joined: Sun Jan 27, 2019 3:51 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: No

Velocity vs position mode

Post by occustoms »

Ok i have talked to a guy who does alot of retrofits on older machines. Reason i started with that is because this isnt something i made up or looked into. But more asking the pros on here that have way more experience than me about this. I can run my drives in either mode. Leaning toward velocity just because of the double loop feedback system of the drive and oak.

But sounds like the precision aka position mode is easier to do.

Is their a benefit from precision other than ease of up and running. Is velocity mode more precise? Following errors or what not.

He told me that it was smoother in velocity mode and that position mode is good just not as good or precise.

Would like to know more about what he is talking about and then i can decide what way i would like to choose. Sorry for the long rant.
aamir
Posts: 96
Joined: Fri Jun 10, 2011 8:31 am
Allin1DC CNC Controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC11: Yes
CPU10 or CPU7: Yes

Re: Velocity vs position mode

Post by aamir »

Velocity mode is more precise and 100% close loop update error on fly.
I did my 600+ installations on velocity or torque mode.
Precision or Position mode did not based on error no PID algorithm.
I prefer and love to do retrofit on either velocity or torque mode only.
Currently centroid may launch Hickory board with EtherCat that would also close loop with high encoder count provision
Thanks
eng199
Posts: 372
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
Location: Howard, PA

Re: Velocity vs position mode

Post by eng199 »

Position mode is the preferred interface.
The control loops are all tightly integrated on the drive in this mode, which gives the most potential for high performance. I didn't see a brand mentioned, but can tell you that Yaskawa will give excellent results in position mode. Other modes require the Centroid hardware to run a control loop. This adds communication delays within the control loop. Analog request and encoder feedback have to communicate between 2 asynchronous systems in velocity mode, which results in the extra delay. Longer time around a control loop limits the amount of gain that can be added to really bring control in tight and lower error.

In a position mode system, the position request to the drive has to be completely reliable or the control and drive can have a mismatch in where they think the motor should be (standoff error). Standoff error can be corrected by p261 and p262, but is not as transparent is it is in velocity mode. In position mode, the Centroid hardware is outside of the position control loop, which does not mean there is no control loop, open loop, not 100% closed, etc. There were some issues with the position request on early systems, which unfortunately made some customers avoid ever using it again.

Do not consider torque mode unless you're an expert like Aamir. This will be the hardest to tune and least likely to produce a good result.
occustoms
Posts: 129
Joined: Sun Jan 27, 2019 3:51 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: No

Re: Velocity vs position mode

Post by occustoms »

eng199 wrote: Mon Nov 21, 2022 1:04 pm Position mode is the preferred interface.
The control loops are all tightly integrated on the drive in this mode, which gives the most potential for high performance. I didn't see a brand mentioned, but can tell you that Yaskawa will give excellent results in position mode. Other modes require the Centroid hardware to run a control loop. This adds communication delays within the control loop. Analog request and encoder feedback have to communicate between 2 asynchronous systems in velocity mode, which results in the extra delay. Longer time around a control loop limits the amount of gain that can be added to really bring control in tight and lower error.

In a position mode system, the position request to the drive has to be completely reliable or the control and drive can have a mismatch in where they think the motor should be (standoff error). Standoff error can be corrected by p261 and p262, but is not as transparent is it is in velocity mode. In position mode, the Centroid hardware is outside of the position control loop, which does not mean there is no control loop, open loop, not 100% closed, etc. There were some issues with the position request on early systems, which unfortunately made some customers avoid ever using it again.

Do not consider torque mode unless you're an expert like Aamir. This will be the hardest to tune and least likely to produce a good result.
yes i ben trying to tune it in velocity mode right now. just on bench not on machine yet. I can get it working great with very little errors during moves. only being out like .007 on begining of move at 100ipm for tiny bit. like its slightly laggy. but what ive ben trying to do is get it better at idle. NO movement right after a jogg move. it wants to bounce back and forth like 7 counts pos to neg. and takes a sec to stabilize on 0.

Current one im using right now is 1.5kw motor with a sdf20kn9. They are cheap chinese motor and drive. their software leaves alot of figuring out. Havent figured 100 on their waveform software so i ben using centroids. I tried to get it close using parameters on servo drive then using centroids after that. i might post a video up later when i have time showing whats going on. Reason im wanting to use velocity mode is because on my fadal im reusing the glentek drives and motors. Also have a 4th axis motor and drive working from dmm currently in velocity mode. So i wanted to get better familiar with velocity mode and heard its a better in velocity mode. By a reputable retrofitter. i know when i have loads on it that tuning will be different. Just trying to get it close on table before on machine and understand it completely.
eng199
Posts: 372
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
Location: Howard, PA

Re: Velocity vs position mode

Post by eng199 »

occustoms wrote: Wed Nov 23, 2022 12:15 am it wants to bounce back and forth like 7 counts pos to neg. and takes a sec to stabilize on 0.
This sounds like oscillation due to kP and / or kI out of adjustment. Generally, low frequency oscillation means the gain is too low and high frequency oscillation means gain is too high. See TB234 (https://www.centroidcnc.com/dealersuppo ... ds/234.pdf) if you haven't already. The key to good results in velocity mode is setting Kv1 as close as possible. The corrections that can be made with kP and kI are rather limited. Of course, this assumes tuning in the drive is already dialed in.
Post Reply