+- RoboDK Forum (//www.sinclairbody.com/forum)
+-- Forum: RoboDK (EN) (//www.sinclairbody.com/forum/Forum-RoboDK-EN)
+--- Forum: RoboDK API (//www.sinclairbody.com/forum/Forum-RoboDK-API)
+--- Thread: Remote TCP speed in curve follow project (/Thread-Remote-TCP-speed-in-curve-follow-project)
Remote TCP speed in curve follow project-Camille-11-30-2020
Hi,
I am working with curve follow projects and I have a speed constraint on my parts.
I use a remote TCP and I noticed the speed I choose in the curve follow project is the one the flange and not the speed of my curve along the TCP.
Since my curve is not a straight line this is problematic for me, specially in the corners where the speed is slower.
Is there a way I could have the desired speed constant along the curve ?
Thanks !
Camille
RE: Remote TCP speed in curve follow project-Jeremy-12-01-2020
Hi Camille,
This is a pretty good observation.
I was discussing this specific issue with a colleague the other day.
Remote TCP are very tricky, so there is never one good answer.
Quote: the speed I choose in the curve follow project is the one the flange and not the speed of my curve
Yes and no. In fact, this is only true if the robot TCP is positioned at the flange of the robot [0,0,0].
The speed you are giving is the speed of the TCP, no matter where the TCP is.
I thought about this problem and I might have some ideas on how to improve that.
最好的人真正取决于你有多少部分need to program.
If it's a small amount, you can use a simpler but more manual method.
If it's a larger amount, you might be better "automating" the process with the API.
First of all, how consistent do you need to be in terms of velocity? (what is your application?)
For the simple method, you can divide your path into smaller sections/parts.
You then create one TCP for each section (locate them as close as possible from the path) and run your programs using these TCPs.
Maybe if you can attach your station with your part or a similar part in case it's confidential, I can give a better idea of what I mean by that.
Jeremy
RE: Remote TCP speed in curve follow project-Camille-12-02-2020
Hi Jeremy,
(12-01-2020, 03:29 PM)Jeremy Wrote:Yes and no. In fact, this is only true if the robot TCP is positioned at the flange of the robot [0,0,0].Yes the robot TCP in my case is at the flange, I forgot to mention that.
The speed you are giving is the speed of the TCP, no matter where the TCP is.
I thought about separating in smaller sections but I don't really want to do that because I'm afraid it would make the robot stop or slow during the process and I can't afford that with the process I am using.
Also, I aim to do a large amount of parts and will probably need to use the API to make this automatically.
I attached a simplified version of the cell so you can have a better idea of what I am doing :
Decoupe GF Remote TCP V6.rdk(Size: 4.8 MB / Downloads: 303)Thanks,
Camille
RE: Remote TCP speed in curve follow project-Jeremy-12-02-2020
Hi Camille,
I think we should have a live discussion on this one.
Envois moi un courriel et on trouvera un moment. J'ai déjà réfléchis à ce problème et j'ai certaines pistes de solutions, mais je n'ai jamais trouvé le temps de les implémenter correctement.
(Si tu n'as pas mon addresse courriel pro, envois moi un PM, je te l'enverai. J'évite de l'écrir directement sur le forum pour des raisons évidentes.)
Bonne fin de journée.
Jérémy