Sign in to follow this  
Followers 0
Gary Burton

Effect of "master position filtering" on an axis (CLX)

34 posts in this topic

I think you are mistaken here Gerry but I'm not doing anything but beating my head against the wall. I follow Peter's lead and give up too. He's yours now Gerry Gary if you post the full explaination of what is going on and what you are really trying to do it would help Gerry out

Share this post


Link to post
Share on other sites
Guys - I know what the problem is , Peter confirmed what I suspected. TWC, you don't understand how interpolated position works with this module...Gerry explained it, you could learn something here. Edited by Gary Burton

Share this post


Link to post
Share on other sites
How can you be so sure? Gary has not provided us with any pictures of the data from the encoder. How can you gear to data that is old? Yes, it can be done but there is a phase delay between the time the master encoder event occurs and is sampled, is sent to the sercos controller, is read by the sercos controller, is transfered to the PLC, is read by the PLC scan, is used to calculate the slave position is send to the sercos card, is read by the sercos card, is relayed at to slave at the next 2 millisecond time interval and used by the slave controller. All of these times happen at intervals and the jitter is 1 interval at each event. In any case I doubt the master and slave positions that are used to compute the error are sampled at the same time. The must be a phase delay and a lot of jitter caused by all the data transfers. I think they are admitting to the phase delay and jitter when the GSV gets a position and a time. Having the time allows one to compensate for the jitter. It may also allow one to predict where the master really is given you know where it has been in the past. Actually, the have to do this because the Sercos controller must download new positions for the slave to go to every 2 milliseconds. An experiment. Run perfect encoder data into the master input and see how much jitter there is in the slaves target position. Try different frequencies to see if the results are different when the encoder counts are not synchronous to the 2 milliseconds and the counts are numbers like 10 one update and 9 next. We have equipment that can do that. We are fanatical about this. The master and slave positions must come into the same FPGA where all feedbacks are sampled in parallel at the same time down to the nanosecond. That way we know the true error between the master and slave. This yields excellent results.

Share this post


Link to post
Share on other sites
The ControlLogix method for obtaining axis positions concurrent with the position of the axis that had the registration event is the same for Sercos axes and analogue axes. I have no information on its accuracy. The time resolution is the same as the Coordinated System Time, but I have no information on the sample frequency of the position data that is used for interpolation. The registration position and time are stored in the MO2 module (analogue) or in the drive (Sercos) - perhaps in the Sercos case, captured by the drive and transferred to the MO8/16 module. Program scan time does not affect this data. The position interpolation is performed within the motion module, as far as I can ascertain. This implies that all axes in a system are constantly recording position history. As to how camming and gearing is accomplished over Sercos - your guess is far better than mine.

Share this post


Link to post
Share on other sites
So, I hear through the back door that you want 0.002" resolution. In that case, quantization error is likely your entire problem.

Share this post


Link to post
Share on other sites
If that is the case then you are out of my league. Ignore all my thoughts about timing I was just asking Peter about how to really determine the resolution that is truely necessary for the application

Share this post


Link to post
Share on other sites
Yes, you are correct. The problem is quantization.........thanks for your help. I posted a picture on another board and Peter identified it.

Share this post


Link to post
Share on other sites
I am pissed. You are on my s-list and s doesn't stand for short. I don't mind helping but I don't like people wasting my time.

Share this post


Link to post
Share on other sites
I'm sorry you feel that way. Thanks again for the help

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0