Posts by Devon

This seemed to work, but I didn't want it all the way down to 10, so I tried 30 and it still went mostly black. So I put it back up to 70, and it still didn't work. It just seems to be mostly off when in sleep no matter what I put in there.

Does it bother anyone else that when you leave a teach pendant sitting, and it goes to that kind of "sleep" mode, the screen actually gets brighter? And then when you start using it, it get's slightly dimmer? I think I know the cause, just not the solution.

If you go MENU > SETUP > iPendant Setup the first box at the top is "iPendant Brightness Setup", which has 3 options: Low, Normal, High. But it seems that someone at Fanuc fat-fingered the actual values for those options. Low is the brightest, Normal is slightly dimmer, and High is practically off. I'm pretty sure when the pendant goes to "sleep", it changes brightness to "low", which is the brightest. Then when start using it again, it goes to "normal", which is slightly dimmer, which I guess is probably right. So basically just high and low are backwards.

I'm guessing there is a system variable somewhere for these values, but I've taken a look through and I don't see anything obvious. Does anyone know where they're buried? I'd really like to correct this on the pendants I have to work with. Thanks!

STILL_CNT + MOVE_CNT were super useful to me! BG programs can't use waits, or timers, and the system clock doesn't even have seconds. But this worked great for allowing me to create a delay in a background program

Finally solved this. If anyone else is fighting the same problem, in this case it was an incorrectly setup encoder. Our RIPE network only has 2 robots and encoder 2 was setup to look at RIPE Master 3. A much better fault on Fanuc's part would be to say "Incorrect encoder setup", rather than the useless "No tracking hardware".

Hey everyone,

Working with an LR Mate 200iD on a 30iB Mate Plus.

We have a copy and paste system we're working on. We've taken backups from the original system that is onsite and working, put them on the new system, but we're running into an LNTK-022 error when trying to setup the tracking frame. Does anyone know what "tracking hardware" is that it thinks is missing? Our encoder and camera are both working. And to make it worse, we have 2 identical new cells that we are pasting to. It's working fine on one, and giving the LNTK-022 on the other. As best I can tell, everything right down to all of the vision setup is exactly the same between the 2. Just not sure what to look at next.

These aren't the exact number, but I think the're close. FOV is about 15", our parts are about 4" and we trigger about every 2". Keep from missing things that are only half in the window at certain points. But the N Found that is reported back is per trigger, so you get back a lot of duplicates. Like if 3 parts go by side by side, it will say 3, 3, 3, 3, until they're gone. Which the PLC is going to see as 12 or 15, rather that just the 3.

Yes, its irpicktool. I'm not sure what you mean by how is the tracking attached to the image.

Not all items are picked. The cell is overfed to keep the robots busy. We're trying to determine how overfed it is so that we can keep it to a minimum.

Hi everyone,

I'm trying to determine the rate of good product coming into the cell based on what the vision finds (I don't care about ones that won't get picked because they didn't pass the vision check). I've done this before by pulling $PKSNACT[1].$NUM_FOUND but the problem on this system is that the camera is triggering frequently, and the belt is slow, so the same object is found multiple times. Any other suggestions?

LR Mate 200iD

Hey everyone,

I'm working with an LR Mate 200iD and I'm getting a MOTN-023 (In Singularity) fault during a joint move... Anyone know what's up with that? Does this robot have some weird options turned on or something? I didn't know it was possible to have singularity issues with joint moves, I thought that was have their reason for existing.

Quote from cunaf

I'm don't know about your version of Arctool, but at the 6.4 (ish) versions the description of the variables was wrong. For instance,
$AWESCH[1,1].$CMD_VOLTS would actually control WFS;
$AWESCH[1,1].$CMD_WFS would actually control Trim

At least this was the case for Burn-back ($AWESCH[1,34]) and Wire stick ($AWESCH[1,35])

I recommend setting up a test program, run it and verify volts write to volts and WFS to WFS etc. under your wels schedule!

Can confirm this on 7.7. Thought I was going crazy for a second. I was like "Uhhh, 500 volts at 150A....?"

Haha, I'd love to know that too! That's my other post. Unfortunately no luck yet.

RE: Why is my VR[1].Found_Pos[2] Blank?

Yes, doing exactly that. Vision finds parts, part locations are sent to the PLC, the PLC does some math (see if all are present, or which are missing or out of place) and sends back a correct pick location to the robot.

Ahh, that makes sense, thank you!

Now here's where I think it gets a little more complicated. This is a line tracking setup, so VR[1] gets populated from the queue when called. I don't believe I can pull the offset data I want at that point, correct? So I could need to pull it during the original image capture and get the offsets then? Then also pull 4 VRs when I do my GETQ?

Hi everyone,

I've got an issue where my vision process is picking up the 4 items it should be, but my VR[1].Found_Pos[2] is blank (and so it Found_Pos[3] and Found_Pos[4]). Should it not be showing me the all 4 items found? I need a way to get the position of all 4 so I can send them to the PLC. Vision process attached.

Unfortunately not an option. System has been in place for years, I'm just here to make some vision improvements.

Hi everyone,

I'm running a single-view visual tracking process that finds 4 objects on a conveyor, then all will be picked at once. I'm sending the X and Y of those 4 objects to the PLC from VR[1].Found_Pos[n] (via a PR sent to a GO). The PLC will average my X values to determine where the best fit for the row is, and do a similar function with the Y values to determine the best Y offset. This part is all fine, and I get the correct values back from the PLC. But my VR[1] that is being picked has very different values for the .Offset compared the the .Found_Pos. This is because my vision jobs have a pretty large offset in the Reference Data for X and Y. I could get away with hard coding the values in for that reference, but if anyone else touches the job and changes the reference, the system will no longer pick correctly unless they know to change the hard coded part. What I would like to do is just pull the reference data numbers from a system variables somewhere, but I haven't had any luck finding them. Anyone know where they are, or have a better suggestion? Thanks!

Is there a hidden option somewhere in RoboGuide to allow the use of more than 1 on screen teach pendant? I'm working on a multi-robot cell and it would be really nice to have a couple of teach pendants on my second monitor, rather than having one teach pendant that has to log in every time I click a different robot.