Hello and it's great to be here and to what looks like a fantastic robot forum. My name is Andy and up until a couple of weeks I worked for an integrator in the UK ( I live in Ireland though ) but like many of us in these worrying times was made redundant 2wks ago. I joined the UK company after working for Hanley automation here in Ireland and helped win a project worth just north of 500K but using dare I say here, a cobot!! I did a quick simple POC and won my new boss the work. So having been with my new boss for just over a year and having just been laid off I decided to go it alone and form my own robotics solutions business and have just been called in to a company to help with a couple of issues that they currently have with a welding application. The robot in question is a Fanuc M-10ia with a R30iB controller and a Fronius mig weld system and of course they have Arc Tool installed. There are a number of issues that they seem to be having and one of which is the TCP setup for the mig torch, so today I have also tried to set the correct TCP ( 6point method XZ ) the burn back leaves 17.5mm of weld wire extruding from the nozzle, we have made a solid pointer to replicate this distance attached to a new wire tip and have then followed the procedure in the arc tool setup manual ( version 8.30 ) but no matter what I do I can not get the TCP to move around the stationary point that I have used for the set up, it's as if the TCP is slightly off elsewhere. The 3point setup produces the same result but it's the 6 point method that best suits what we are trying to achieve here. I have checked the zero positions and carried out a Mastering procedure but this did not help and the same result remained.
The next problem is that we are using a search procedure for the part, which may I add is on a single axis horizontal positioner. We move the positioner through 90 degrees for each of the weld planes and we are using only PR32 to collect the data for each search plane and then PR32 dumps the offset values into the relevant PR's for that plane i.e PR10 / PR11 up to PR13 but when we cycle through the program on a Dry run the first plane is ok but the next 3 planes seem to have values in the PR coordinates used for each of the weld planes/positions ( hope I've explained this correctly ) resulting in the offset positions being wrong. Is there something that I am missing here or doing wrong i.e. using a master PR to then send the multiple offsets to all of the others, the main search PR32 returns to zero after each search. I have attached an image from a demo that I did using RoboDK just showing the workcell layout.
Thanks in advance for any help
Andy
