Sign in to follow this  
Followers 0
Gary Burton

Motion group coarse update time with SERCOS and kinetix drives

22 posts in this topic

I have a generic question regarding the coarse update time. If I have an event task that is triggered by an axis registration input I have a possible delay of (coarse update time+ SERCOS ring update time)....in my case this could be anywhere from a best case scanario of about .1 milliseconds to about 2.5 milliseconds. Is there a way to make the motion group start its update off a trigger....?

Share this post


Link to post
Share on other sites
\ Hmmmm.....I guess not Edited by Gary Burton

Share this post


Link to post
Share on other sites
Another CL Motion with Kinetix question: When a registration occurs is the position registered in the drive or in the CL data table? ie does the drive store the position or is it merely reported back to the PLC?

Share this post


Link to post
Share on other sites
I have read this several times and I'm not sure I still understand the question. Can you go into a little more detail? From what I understand you are talking about the delay of the execution of your event that is triggered off of your registration input but that isn't really making much sense. Is this a delay of some type of motion execution in your event driven task?

Share this post


Link to post
Share on other sites
Well the event is triggered by the registration input....which is located on the drive. The sercos ring update time is .5 m/s.....and the coarse update time for the motion group is 2 m/s. So if the drive sees an input, it could concievably take .5 m/s to reach the plc and then if it catches the motion group on the front end of the coarse update time it will have to wait till the motion group completes its scan before executing the event task......This is assuming the motion group executes with priority 1 every coarse update time. The motion group and the plc scan are not asynchronous right? Does this make my question any clearer?

Share this post


Link to post
Share on other sites
Yes and no. There is no way to anticipate the registration and adjust the coarse update accordingly. But that does not mean that you are going to have this variation of whatever you are doing with this registration input. It all depends on what you are doing. If you are capturing position this variation can be reduced tremendously. What exactly is the critical piece of information you are trying to capture with this registration or are you trying to execute some motion off of it. Please give details. Also what is the tolerances we a looking for in whatever you are trying to capture?

Share this post


Link to post
Share on other sites
Well it was mostly a question out of curiosity....I was wondering if there was a way to trigger the motion group scan...or switch it off.. I know how to capture the position at registration using GSV and SSV..and then use that for interpolation (used a cam profile) I just like playing with different scanerios One question I do have is when the registration occurs, the position value of the axis used is stored (so it can be retrieved with a GSV).......how long does this information stay in the drive? Until you re-arm the registration input and the input goes high again? I'm still unsure exactly how the registration data is stored and reported.... Thanks for replying to my question. Edited by Gary Burton

Share this post


Link to post
Share on other sites
Caveat: This is how I think it works.... The registration position is stored in the Kinetix system module. It is retrieved with GSV instruction. The motion group coarse update time specifies how often to run the motion task - not how long it takes to execute. The system will fault if the execution time exceeds or equals the update period. Your event task could be triggered but interrupted by the motion task. The motion task executes at the coarse update rate and most likely will execute several times during one scan of the continuous task. Depending on the relative priority of your event task, I think the maximum delay from registration event to completion of the event task would be 3 sercos ring updates plus motion task execution time plus event task execution time. 3 sercos updates - 1 for reg event, 1 for GSV request, 1 for GSV response. Other delays that you may need to consider more carefully are the input filter on the registration input (700 ns) and the response time of your registration sensor. These will affect the stored registration position depending on axis speed. Edited by Gerry

Share this post


Link to post
Share on other sites
Now I am rusty and need to read a little on this but for the analog version, not only is the registation for triggering an event, but it also can log the position of the axis upon the triggering of the registration. This was done within a window of around 1 microsecond. It is not that the position you recieve will vary, there just may be a delay in the time of delivery. I need to look into the sercos but I can't imagine them not including this feature.

Share this post


Link to post
Share on other sites
Yes TWC....that is what the registration does as well, that is store the position on registration

Share this post


Link to post
Share on other sites
Time can also be logged. Then you could take the position and the time difference between the registration and the the execution of your event ladder and... I still am not sure I am on the right track of what you are asking

Share this post


Link to post
Share on other sites
Yes time can be used to set interpolated time on another axis and then get position of that axis......for camming for example

Share this post


Link to post
Share on other sites
So you understand how to get rid of your .1 to 2.5 ms deviation now?

Share this post


Link to post
Share on other sites
Well I can't get rid of it.... I was just hoping there was some other way to trigger the motion group execution instead of by the coarse update period....... I was aware of how to get the registration position.. Thanks for your help

Share this post


Link to post
Share on other sites
AB's all the way!

Share this post


Link to post
Share on other sites
Can you still explain how this deviation will affect your application? I'm just curious

Share this post


Link to post
Share on other sites
Time = distance Edited by Gary Burton

Share this post


Link to post
Share on other sites
I know that Gary. I was asking about your application that you were thinking of putting this in. Instead of giving smart answers why don't you just say it's none of your business or you don't have a clue I was trying to see how much the deviation was actually going to affect your application and see if we could find ways to reduce for your particular application I'll just remember to never put forth any effort into responding to your threads from now own

Share this post


Link to post
Share on other sites
Unreal, the Boks won Edited by Gary Burton

Share this post


Link to post
Share on other sites
Richie McCaw reflects

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