Posts by Chasberrypi

If I change the way an offset is found do i have to go through my program and touch up everything?

PR [11] is the offset being used for all my points after the length has been found.

I taught myself about searches. Sorry if this is a simple question.

Original Method:

Gets a rough measurement of

: bar length

8: J P[128] 50% FINE ;

9: Search Start [1] PR[41] ;

10: L P[145] 600mm/sec FINE Search[Y] ;

11: Search End ;

12: ;

13: Touch Offset End ;

14: Fine tuning bar length;

15: J P[160] 50% FINE Offset,PR[41:2nd bar search] ;

16: Search Start [2] PR[11] ;

17: L P[160] 600mm/sec FINE Search[-Y] ;

18: Search End ;

Newer Method

4: --:Gets a rough measurement of

: bar length ;

5:J P[1] 50% FINE PTH ;

6: SKIP CONDITION RI[1:OFF:LASER GO RANGE]=ON ;

7:L P[2] 1000mm/sec FINE Skip,LBL[99],PR[43]=LPOS ;

8: LBL[99] ;

9: ;

10: --:Adds 10mm to GP1 Pr43 Y

: value ;

11: PR[GP1:43,2]=PR[GP1:43,2]+10 ;

12: ;

13: ;

14: ;

15: Touch Offset End ;

16: ;

17: --:Fine tuning bar length ;

18:J PR[43] 50% FINE ;

19: Search Start [2] PR[11] ;

20:L PR[43] 600mm/sec FINE Search[-Y] ;

21: Search End ;

We had this alarm previously and had a Fanuc service tech come out and changed out the entire wire harness.

Quote from EPeters1

This is the process I use: https://weldbot.com/wp-content/upl…nt-eb033721.pdf

Has given me the most precise zero out of everything I have tried.

This is similar to what we do but this seems more precise than our method.

I'll be trying this out hopefully within a couple of weeks.

Thanks!

Quote from R47

hi Chasberrypi,

why do you need to remaster the robot? is it because 1 or 2 seems are out of position?

If so.... that is not the right way to fix the bended torch.

or do you really get an BZAL or something like that

Hard collisions are the reasons why remastering is done. Sometimes it was bad enough that we have to replace the torch and a whole arm.

Never had to remaster due to the batteries dying, my company is really good about changing them.

Quote from HawkME

Quickmaster

My understanding for quick master is for integrator use only. Also, on our older robots the witness marks are not visible anymore.

Quote from C_hiney

Digital angle finders are great for quick master J2-6 and get you almost perfect. J1 use a dial indicator. After moving them you can use single axis master and with the indicator change master position to the last known position. Either write down the joint position or take a back up before disconnecting, summary DG will have the joint position in it. Always check zero after if not mastering at zero.

Interesting, do you use the baseplate of the robot as your 0?

J:1, to my knowledge, has never been an issue when we have to remaster. Been learning robots/automation for 3 years and have my current job for 3 years.

Question: How does everyone find their Zero Position on their robots?

(95% of our cells are welding cells)

Current Method: We use a point to point method where there is a fixed pointer inside the cell and we add a pointer to the end of the torch. We use three different programs to help figure out the zero for all the axes.

-Bracket program where we make points (indents with a punch) on the robot and used the fixed pointer that "touch" the points on the robot. Each point corresponds to a 2 or 3 axis.

-TCP program that aligns the pointers perfectly to each other (linear vertical) where the tips are barley touching one another. We use this to see what axis needs to be brought in and use the difference in degrees to adjust our zero points

-Zero program is made when the cells come in and we use the witness marks on them to create the original zero point.

Comments: I'm not a fan of this method due to whenever a robot gets remastered we have to go through every program and make adjustments. 8 out 10 times its way off and everything needs touched up which causes quality issues and production interruptions. I'm also not a fan of the bracket program because each point (indent) relates to more than one joint.

First post, if anything needs more explaining I will explain better.