Posts by JonHopper

Honestly I think the only reason it's done that way is so your io setup takes 10 seconds instead of 20 seconds.

A lot of RJ2's do not have the memory card slot. Otherwise yes this is by far the easiest way.

Quote from keyboard7

We use a compact flash pc card adapter found at best buy. Then you need a 2GB compact flash cardalso at best buy Will plug right into controller next to floppy port.
You can call lincoln robotic hot line, they will talk you through the process. Number is 1 -888-532-8001. goes something like this.

menu
files/ enter
select files backup.

Hopes this helps this is how we do it.

Display More

I've used mames[7] successfully on a dual robot linear track with two mastering posts so I could swap pts between them. Worked well. Didn't try changing values for 1 - 6.

OR... use two / three sets of light curtains ... much simpler solution than everything that has been posted so far

Is this going to be coordinated motion, that is, robot + "trunion" moving together at the same time? (for example, mig welding a round tank application)

Or are the movements going to be separate?

If the latter why not have one robot be master of the "trunion" external axis, have a set number of defined positions (as many as you like), and remotely call positions from the other robot with some code running in submit interpreter?

This is very common practice in dual mig robot welding cells where a trunion / external axis is used to reorient the work piece.

GOOD INFO SkyeFire Thanks for posting!

If it's a GM version don't bother, I just went through this

I thank the Kuka Gods every day.

For the same price you can sit and key in your programs one line at a time on a Fanuc, if you really think Kuka 'hates' programmers. Or write a basic search routine that's ~300 lines long instead of 30. Choose your battles. Nobody writes a manual for life..

Quote from HawkME

First, I use a valve that it's either detented or normal locks the tool.
2nd, I set up a BG logic program that requires two separate outputs to be turned on before allowing the tool to unlock. If the unlock command is given without toggling the interlock output first, it will automatically lock the tool.
3rd, you can set a ref position at the tool change station. A ref pos can be set up to turn on a DO. You can then require that DO to be on to allow the tool unlock.

Sent from my VS985 4G using Tapatalk

This is how I've done it in the past, and the only thing that allows the tool to fall to the floor is some dumb@55 hitting the manual valve spool.

We also added dual limit switches to the tool stands so they must both be made before tool is allowed to unlock.

Kind of hard to understand exactly what you mean, but- if the object(s) you're placing the glass onto is static (non moving, non changing) then there should be no need for any type of distance measurement

If your payload data isn't set correctly acc/dec gets funky. Also if TCP is off splines get funky. "Funky" is my new technical term of the day.

I set up something similar, the only issue with this is it only tracks base positions. At full speed the bases will still collide with each other due to stop distance. Also during production robots come within 1 foot of each other.. No TCP / Arm monitoring

If I was in this situation I would try manually bringing the robot to this mirror point, or as close as possible, save the point and look at the data and compare it. S&T values are probably wrong.

Quote from diglo

I'm just using the the variable viewer of the SmartPad since my "C" variable is just a global integer variable.
I know that you can't directly set an $OUT variable without stopping the Advance pointer (unless you use $OUT_C).

Inviato dal mio XT1052 utilizzando Tapatalk

You can use TRIGGER as well.

Forgive me I haven't touched a KRC4 since February but I seem to remember that you cannot view non-global variables while running ?

Kuka was supposed to have some software for this, but didn't. It never ended up being integrated with collision avoidance between the two robots (to my knowledge, I work for a different company now so I'm not sure if it has been implemented at this point or not).

The manual is very vague on this.

For example... can I do this?

TASK PERS wobjdata palletBase{150}:=[false,true,"",[[0,0,0],[1,0,0,0]],[0,0,0],[1,0,0,0]]];
TASK PERS num offsetLayer{150}:=0;


PROC calc()


VAR num n;


FOR n FROM 11 TO 150 DO
   palletLayer[n].oframe.trans.z := palletLayer[n].oframe.trans.z + offsetLayer[n]
ENDFOR


ENDPROC

Basically how do I declare a persistent array for wobjdata? Will every subset of data be set to the value I assign it ?

Well I finally found the right info no thanks to Fanuc (Big surprise)

Input instance: 7
Size 8

Output Instance: 8
Size 8

Config Instance: 1
size 0

use the IO sizes in the ADPRCFG screen and ignore input / output header words

I find it extremely hard to believe NO ONE has any experience setting up a controlnet network on a new(er) fanuc.

Hey everyone.

Trying to setup an R30iA 120ib with controlnet card (SST card). Single CompactLogix L43 with 1768-CNB controlnet scanner. Looking for the Assembly Instances needed for Input / Output / Configuration in the RSL5000 Generic Controlnet Module configuration. I can scan the network with RSNetworx for CNet and everything looks good. Just need the proper assembly instances to get this thing to work.

I've tried calling Fanuc. They either genuinely didn't know or were pulling my leg (said they didn't have the info). I called Rockwell and they gave me a few numbers to try but they also did not work. Any help would be greatly appreciated, nobody sets up these networks anymore and the info is impossible to find online.

I've had this problem when copying new programs / overwriting programs, The only way I've found to get rid of these errors is a reboot