Excellent detailed explanations.
Excellent detailed explanations.
Thanks for your help.
I must admit I have never put my fingers in servo parameters.
Could you point to where these files are ?
The hard drive looks healthy ; I detected no strange behavior in Windows.
I noticed that the LEDs on the drive for motor 6 are indeed black, while it's chistmas for the others : https://photos.app.goo.gl/DcXFEvVR53ZjmEMB6
I made a cold start and I saw that they don't blink at all.
I swapped the intebus cable with another one, and it changes nothing, so the cable is good.
Could be the Kuka Servo Drive cables, but it seems all nice and snug.
Maybe it's the KSD1-08 servo drive module.
Funny, I see on Ebay that the prices for these ddrive modules range from 175 $ to 2500 $, and the KSD1-08 module is generally more expensive than the -16 , -36 and larger ones...
For some unknown reason, I can't move the robot anymore.
I get error 250 : "Incorrect servo parameter A6"
and error 220 : "Power module axis A6 not or wrongly plugged"
I checked the wiring and nothing seems to be wrong.
Any idea how to fix this ?
After further testing with an Ohm-meter, I saw that the current still flows through the cable ; only the plastic is a bit damaged, and I could fix this easily.
But then I realized that one of the pins of the connector on the motherboard was bent so that it didn't connect to the corresponding cable.
Things are back to normal now !
I made a video out of the little tutorial, and added extra warnings in the description : https://www.youtube.com/watch?v=YpKQYHMVEhA&feature=youtu.be
But as Panic mentioned : it's better to think twice when undertaking this kind of task for the first time !
I might be able to get one from Eurobots from whom I pruchased it, I guess they have heaps of hardware for canibalistic purposes.
I also found this interesting video : https://www.youtube.com/watch?v=RyRVA7NDo9o
I'll be more prudent next time, but there is not so much clear and visual tutorials out there for robot hardware repair, unfortunately.
At least, if this whole adventure can help out poor souls like me in the future, it will be a positive thing.
I think that I found what's going on with the pendant and dead man switches.
I noticed that the small dead man switch which is on a sort of cylindrical bump had a strange behavior : when pressed, it did nothing, and then if you press harder it switches on the display of the "X, Y, Z, etc... or A&, A1, A2, etc..." options.
Normally, that second level should also be an open circuit, as with the other dead man switches.
So I checked the ribon cable, and found that it was squished on the edge, certainly it was sandwiched on the nearby screw socket that secures the case closed.
EXACTLY what Spirit 532 warned me about.
Now here come the big question : what is the reference of this ribbon cable so that I can purchase one, (either 0.8$ on Ebay, or 245 $ from KUKA)
The batteries are shipping and should arrive soon.
You should never get an "Active commands inhibited" message without another message that inidicates why.
Err... well, this is the twilight zone then.
And how about the cold starts ? In what circumstance would the system not do a normal restart after a "hibernate" ?
Well, for my defense, I did ask for some advice on June 10th, and only got an answer after bumping the topic on July 31st.
I have looked very hard for ANY picture or description of the inside of a KRC2 KPC and found absolutely nothing.
This is why I took the time to document what I did, hoping that it could help others.
I don't have the money to get an official KUKA training, so I do what I can with friends and forums...
Also, I really don't see what the pendant has to do with the forced cold starts.
I'd love to try another KCP, but it's like 1000 carat diamonds, I don't have many lying around.
Just so that you don't think I'm a raving madman at the helm of a massive industrial robot, Wall-E is a Little Kr6 which runs about an hour a week, therefore I didn't feel the urge to sell a kidney for some KUKA folk to change the oil.
The batteries are shipping as I write this.
The screen is SLIGHTLY brighter indeed.
But the robot is COMPLETELY useless...
I have been getting warnings of low battery voltage for some time.
Could that be the reason why the robot does cold starts instead of just getting out of hibernation ?
Like there is just enough juice to save the mastering data, but not enough to finish the hibernation procedure.
And damn ! How can the drives stop working in manual mode after the first triggering ?
This makes absolutely no sense !!!
Time to re-install ?
Here is what happened this morning as I shut the power on :
-For some reason, the controller did a cold start ; I even had to log into Windows
I used "User=Administrator" and "Password=kukarobxpe2" (thanks forum !).
It managed to load everything fine, just saying something about sceduled maintenance
-I set the operating mode to T1
-I press the dead man switch on the KCP ; "I" becomes green and I hear the relays trip in the controller.
-Keeping the dead man switch pressed, I move the robot in Carthesian mode for as long as I want, in whichever direction
-I release the dead man switch, say, to scratch my butt or dink some Rhum ; "I" becomes red
-I press the dead man switch again ; "I" stays red, no sound in the controller, but I can see the "X, Y, Z, A, B, C" options on the right side of the screen
-If I press any key to move the robot, say, the "X+" key, I get the error "1376 KCP Active commands inhibited" (and nothing else)
I should add that "Automatic" mode works perfectly. I can use the dead man switch as many times as I wish : no error messages.
are there any other messages
do drives get enabled EVERY time you press and hold enabling switch in T1?
No other message.
After exiting "hibernate" mode, I get one shot at moving the robot, and then any subsequent attempt does not enable drives and the same message is displayed.
I know the dead-man button works because it has an effet on the interface as described earlier.
I can still run programs in "Automatic" mode ; the "Motors on" button works just fine.
In fact, it seems to work better now that I blew compressed air up it's ass.
But the T1 and T2 modes are still glitchy, so I can't define new bases and tools
I realized that I had tripped the Emergency stop on the pendant, so I reset it back, shut the KRC down, and the back on again.
I was able to move the robot in T1 mode, but only on a single stroke.
Then the error messages came back.
So I decided to open up the pendant and took out the copper strip from the black terminal, and did the whole Hooba-loo once again : no luck.
For the sake of my mental sanity, I'd like to know where that copper strip is supposed to go anyway.
The dead man switch seems to work normally, since it triggers the display of the "move" options (X, Y, Z, A, B, C in "World" mode for instance), but it only allows one stroke to be done ; if ever I let go of it and press it again, it will not activate the motors (I don't hear the sweet "click" sound in the controller) and display the "1376 KCP Active commands inhibited" message.
I just tried to move my robot with the pendant, and I get "1376 KCP Active commands inhibited" messages, so I guess I did something wrong to the pendant while re-assembling it.
My guess is that it has something to do with the 6D mouse, and the funny copper strip that sticks out of it.
When I popped the case open, the mouse stuck on the wrong half, as I mentionned in my tutorial, and that strip srung out, but I don't know out of what...
So I stuffed it in the black port which is next to the white port where the 6D mouse's data cable goes.
Maybe this is beaking the security circuit...
Before I open up that can of worms again, I'd like to gather some advice...
How hard did you pull in step 7 to rip half the connector off the board?
I used patience and a plastic spatula from my kit.
How do you tell a plastic connector that's supposed to stay with the cable from one which is supposed to stay on the socket ?
If you never experienced it or found clear documentation about it, you can't, thus my photo tutorial.
By the way, the screen is still dim as hell.
I guess I'm good for buying a new one, or perhaps they never get brighter than this.
If that's the case, I'm amazed that everyone finds it normal, because in a well-lit work environment, I really need to squint to read the screen.
It could also be that I'm 46...
It's a good idea to upload the images to imgur and link them here using img tags
I'm done with opening web accounts, so I placed the images in my Google Drive and put the links in "img" tags, but it doesn't seem to work.
Anyways, I hope it can be of some use as it is.
I was too impatient and proceeded to operate.
Now I can see the reason for your advice ! Thanks anyways
I attached a little photo-tutorial that will help with replacing the lamp (or the whole screen ) of the KRC2 pendant.
Having a smartphone repair toolkit can be handy to pry things out without too much damage (there's still the occasionnal little plastic rib that will snap anyways)...
Some pictures have a warning because sometimes what is depicted is not what you should do, but rather what you should avoid doing !
I hope it helps...
AT LAST ! I received my pendant screen light.
I ordered it from Ebay here : http://cgi.ebay.com/ws/eBayISA…e=ADME:X:CPSSPSIB:US:1123
The seller was flawless, but the French postal services in summer.... meh.
Anyways, befor I attempt open hearted surgery on my precious, I'd like to gather some advice from those of you who have already done it.
While I am at it, I will also try to clean up the "Drives ON" key which seems quite capricious...
Those photos bare no resemblance to your previously posted DEVNET.INI data.
For one thing, why are you looking at a single byte when your analog inputs are (according to your previous posts) 16bit? Do you even have your INB0 mapped to the same DeviceNet location as the ANIN? Wading back through the earlier parts of this thread, I no longer have any clue what your current configuration is.
If your analog input has 16 bits of resolution, map an actual 16bits of $INs to it. Get your "decimal" value by using a SIGNAL declaration to map the $IN group to an (unsigned) integer variable like so: SIGNAL AnalogIn $IN TO $IN.
Disconnect the analog input from your plasma cutter, wire a potentiometer into the analog input https://www.allaboutcircuits.c…tiometer-voltage-divider/, dial the pot to various fixed voltage with a meter, and for each voltage, log the value of AnalogIn. Make sure to cover the entire input scale.
Plot in Excel. You should see the linear relationship between your "raw" digilog value and the input voltages.
You are right. I'm a moron. It's written right on the module, and in the online documentation : https://www.beckhoff.com/english.asp?bus_terminal/kl3102.htm
My current configuration is shown in the attached image.
I'm back to pulling my hair out with
Hopefully I can manage to wrap my head around all this.
I wish there was a little program where you would just decribe you input and output blocks (with the fabricator reference perhaps), and it would spit out the proper description for the iosys.ini file...
This is just tedious as hell, and I suspect it is a huge waste of time for many.