Fanuc Robot Problem

I’m having a problem with a Fanuc robot that I hope somebody can help me with. During a problem we found out we could not get the monitor function to work. Normally you can pull up the programs and press the monitor key and you can see where the robot is in the program. I went to our other robots and tried it out and it works fine. When you enter the programs and press the monitor key the screen will flash like it’s going into that function but it just comes back to the programs screen. I thinking that it’s some variable that has to be turned on but the question is which one? Ive looked it up in the book but to no avail. I also tried calling up Fanuc but they are closed for the weekend. Can anybody shed any light on me?

It is a Fanuc R-J
Robot S-420
TPP+ (Karel Run Time)
V3.07PD

Thanks!!!

I understand your urgency...

but, do you have any idea how much we don't know what your situation is?

Are you a programmer?

If so,... what have you found through status checking?

If not,... how can you expect to provide the answers we need to help you?

Thanks for your help Terry,

It's not big of an issue, i just ran into a problem and have run out of resources and i hoped sombody would go...."oh I have have seen that before, just do X" The robot is up and running fine, but if it freezes in a program and i won't be able to figure out which line or what it needs to continue it's cycle. I have pulled up the program status by going to "MENUS->STAUS->PROGRAM But that says that im running a program that isn't there (different problem for a different day)

Im not a programmer, I am the Maintenance Person for my shift and Programming Robots is just one of the many things i do along with PLC's, VFD's, QC, servo systems and such(the list grows every day). It's sunday and there isn't a programmer or Electrical Eng in the plant. Plus I went through this with them friday afternoon and they said call Fanuc because they have never seen this before either.

As far as status checking, I don't even know where to start, thats why im posting this. I looked up the problem in all my manuals for the RJ robot i have and the manuals for a RJ2 but the program(running) monitor screeen just isn't coming up on the pendant.

I thought that the question was pretty simple but I guess im not giving the right information. Im assuming that it is some system variable that needs to changed to make the monitor screen on the teach pendant to come up, but ive gone through most of them and can't figure out which one for this problem.

CD,

I can see that you are in a hateful situation.

While "Automation" has provided a great many benefits, it has also made it easier to make troubleshooting NOT easier.

The worst case being,

• You have a machine that runs under control of a program.
• You have no access to the program.
• There is no display.
• There is no manual.

The machine has been running fine for years. Now it doesn't.

The fact that the machine has been running reliably for so long indicates that the program is probably OK. It doesn't indicate anything about the quality of the program - only that it seems to work. Or at least, it did work.

Now, something has gone amiss...

Without knowing anything at all about how the program is written, you, as the local maint. guy, should have some understanding of how the process works. Or, you might need to rely on operator expertise.

At any rate, if any thing goes amiss on a system that has been working reliably for some length of time, you should feel pretty comfortable knowing that the problem is in the field - not the program. Now it becomes a case of knowing the process and checking the I/O related to the problem area.

None of us have access to the program in our car's computer (if any of you does have access, would you consider sharing?). But we know the general process and we can usually localize the problem.

This is not to say that any program that has been running for x-amount of time has proven itself to be good. All that has been proven is that the program has handled those situations that it has run into.

There is a whole bunch of "what-if's" that might, or might not, have been tested. The more complicated the system, the more difficult it is to test EVERY case. (It is possible to test for EVERY condition... but that can be very expensive in terms of time and labor.) So, only the more obvious cases are tested.

Many cases are dismissed as not being possible in terms of the process - yet, logically, they are possible. How can two sensors, one at each end of travel, indicate that a "thing" is at-this-end AND at-that-end?

• 
  OEM: "This program has no problems."
  USER: "How can you be sure?"
  OEM: "No problems have shown up yet!"

The OEM is essentially challenging the USER to "PROVE the NEGATIVE". That is, Prove that the Program is NOT solid.

In this sense, every USER of an "Automated" system is a "Beta-Tester".

Once a program-fault occurs, the proof has been made! However, if a program fault does not occur, all that is proven is that you haven't run into the necessary conditions to cause the fault to become apparent.

"There ARE aliens!"

This is a case of "PROVING the POSITIVE". To prove this, all we need is to run into ONE alien.

"There are NO aliens!"

This is a case of "PROVING the NEGATIVE". To prove this, we would have to visit EVERY cubic-inch of ALL EXISTENCE (is there anything beyond space as we sorta know it?), AT THE SAME TIME! They could be avoiding us, don't you know?

• 
  JOE: "There are no aliens."
  JACK: "How can you be sure?"
  JOE: "No aliens have shown up yet!"

RESET TANGENT:

One of the more painful aspects of being a USER is that you are subject to what the Program/System will let you see or do.

In some systems, as long as things are fine, then things are fine. But when something goes wrong...

• 
  JOE: "What's wrong?"
  JACK: "It's broke."
  JOE: "How do you know it's broke?"
  JACK: "It doesn't work."
  JOE: "Is it plugged in?"
  JACK: "Yes."
  JOE: "Is it turned ON?"
  JACK: "Yes."
  JOE: "Was the START Button pushed?"
  JACK: "Yes."
  JOE: "What happened?"
  JACK: "Nothing."
  JOE: "Should something happen?"
  JACK: "I don't know."
  Joe: "So... What's wrong?"
  JACK: "It's broke."

This is the "unspoken" intercourse between Man(Joe) & Machine(Jack) in a system that is designed under the assumption that, as long as all is well, Jack (the Machine) will behave as expected.

If the program is not designed, from the beginning, assuming that shizt happens, then that program is saying..., "If I'm broke, I dare you to find out why!"

And so, you sit there, staring at a black-box, wondering what to do. If you don't have the documentation to help you solve this problem, or if the system has not been designed to help you, via self-diagnostics of some kind, then your only option is to call Tech-Support.

Sorry cdlove, No experience with Fanuc.

I have plenty of experience with Panasonic, Daien and Nachi though.
Anyways, check the system settings to make sure this function is not turned off. Can you possibly run a different program on the same robot and monitor the steps?If so,you have a problem with that particular program. In this case try either, renaming the program or copying it to a new program number. Be carefull though, programs are called up different ways check the start allocation.
If your system uses a binary or bcd input for program call you'll need to change some plc code as well.

cd, I've been in your situation before. Thinking that it has to be something stupid that I'm forgetting to do. One time I was trying to get the screen contrast adjusted on a Daihen robot teach pendant. I had done it many times in the past, but could not figure out where the feature was located in the system settings. I looked in the manual knowing that it was in there, but could not find it. I called the tech support and they told me that this particulat series pendant doesn't have this feature. I would need to order a new screen they said. After ordering the new screen,I finally found the setting for the contrast. Needless to say, the tech that told me I needed to order a new screen now takes purchase orders over the phone.

Tim

Tim-

The system settings is what im trying to fix. I just don't know which one. There is a lot in this Bot (nice ryme huh?) Those are some good ideas but i don't know if it will work in my situation. Here is the best example of what the monitor screen is. Imagine if you could watch the scan of a plc. If you could somehow watch a magic cursor go from contact to contact, rung to rung, of the plc ladder logic. Well with the Fanuc Robot programs you can. What you do is Pull up a teach pendant screen that has all the programs in the robot. You can then go into different programs and look at the program, modify it or whatever. But back at the main screen of the programs area there is a key that says monitor. If you press that while the robot is running it will switch screens to whatever program the robot is running, and highlight the line it is on. When the program changes, such as a jump or calling another program, the screen will switch to that screen and a visual indicator will show what line of the program the robot is proccessing. With our robots (and im assuming alot of other peoples robots) is there are small programs each with a specific function. Such as LOAD,UNLOAD,HOME,INITALIZE,MAINT_POS. What happens is that the robot starts up in one program but depending on what it needs to do it will jump in and out of programs through it's cylce usng jump and Call. Now if the robot hangs or freezes up you pull up the monitor screen and see where the robot is at and what it needs to continue. So in my problem it was a sensor, this i knew, but a standard feature of the robot we use for troubleshooting didn't work. It's like using a handheld for a plc but not being able to see the state of the contacts. So if it were somthing more complicated like waiting for data from a plc or some kind of hidden condition to happen, i would have to pull out the written program guess were the robot is at and go through all the I/O's and data registers in the robot one by one. NOT FUN!

Tim for future knowledge what has your experience been with the tech support for panasonic, Daien and Nachi?

Thanks again Terry, usually around here we get
Joe: "it's broke"
Tim: "You fix it"
Joe: "Im going on break"
Tim: "Jiggle some cables"
Joe: "Shoot that didn't work"
Tim: "quick bypass it and leave it for day shift!"

cd,

I can't imagine a robot would stop in a middle of a process and the user not being able to determine the next step in the programing.
If this is your case then yes, you have an annoying problem. If you think that monitoring a program whether it be the steps itself or i/o's is a nice feature, to my knowledge of robotics its a standard feature in almost all robots. How old is this robot? I guess I'm a little confused. What would the screen show in a normal operating state of the robot? When the robot does stop, can't you just block or trace back or forward to see where the robot is at? Does it not show you as your manually stepping through a program at which step the robot is at?
Like I said before, I don't have any experience with the "Fanuc software".I can only give you some things to try, seeing you are stuck at this time. You said the system settings is what your trying to fix. To my knowledge you cannot fix it, but get the software reloaded into the main cpu or an upgrade. I can tell you, I keep over 150 robots going everyday at an automotive plant and have had to reload the system parameters. I just had an incident a while back where the power to the plant had flickered. I had one robot that as it was running the program, the file was no longer any good after the power was restored. Weird!! I had to initialize the memory and reload the system settings and parameters. Files do get corrupted and things do go wrong.

Now when you say "LOAD and UNLOAD" is this a I/O file that contains the necessary I/O's to actuate cylinders etc. or is this just a commented command for an output?
Unless your watching the robot and know its sequence by heart, I can see where this would be annoying. If the robot was to actuate a cylinder at say step 20 and step 21 would be wait for the input from limit switch, at this point you cannot tell where the robot is.
It seems that you have a function that allows the user to do some offline editing where you can go in and change some of the commands throughout the program while the robot is in operation.


Rule out the obvious things.
1.Does it not monitor any programs correctly?
Ans.=yes
Sol.=system setting corrupted or a setting you have yet found to make this work.
Ans.=no
Sol.=Try to monitor another program.

2.Swap hardware with another robot, teach pendant etc.
3.Call FANUC TECH SUPPORT!!! Your company probably paid at one time or another for this support.


Let me know if you have anymore information.
Good luck!
Tim

P.S. You can tell Terry has had some of those critical thinking classes. He's right, trouble shooting is a process and to get from point A to D you must not forget points B and C.

I may be wrong but I think y'all arent looking correctly at what cdlove is saying.

What I see is that when he trys to go ONLINE in MONITOR mode to use the associated software for troubleshooting, it wont connect. I am not familiar with any of the software etc that CDLove mentioned but I do use GE Fanuc Versa Pro which uses Monitor as a means to connect to a plc when running. I take it that he either has a communication or software problem but do not understand why an error message wouldn't display.

Best I could offer is verify the communication setup, I can not offer details for that. Just in case it may be a software/computer issue you may want to scandisk and defrag the system.

Tim said:

cd,

If you think that monitoring a program whether it be the steps itself or i/o's is a nice feature, to my knowledge of robotics its a standard feature in almost all robots.

Click to expand...

That's my problem. This is a standard feature that somehow got turned "off" The million dollar question is what software setting do i turn on to get it to come on. All of my other fanuc work just fine. On the teach pendant you press the following keys "SELECT->F4 (MONITOR)

Thanks Rsdoran, but im not using a laptop, this is the teachpendant/deadman switch. It is always connected.

Thanks to everybody who helped me but my shift is over, and Im going to call Fanuc when they open tomorow. Im sure they will tell me somthing simple. I will be sure to come back and post the answer to this problem.

cd,
Look one more time before you go calling Tech support.
Such a standard feature is got to be in the manual. I know sometimes when I go to look something up in a robot manual, something as simple as this doesn't cover but maybe one small paragragh. When your going through a two - three inch manual its easy to overlook. Good luck and let us know what you find.
Tim

CD,

Seems to me that this problem you are having is one that should have been anticipated by the designers.

Since robots should be able to process Widget-A, -B,... -X, there should be way to easily switch the program AND operating configuration from Program-A/Config-A to Program-B/Config-B to... Program-X/Config-X.

Assuming that there is the possibility for corruption, I should think that they might have included a way to "Recover" a particular operating configuration.

Is the operating configuration the same for your several robots?

Do you have a way to save a particular operating configuration?

Do you have a way to view the "operating configuration parameters"?

Is there a way to print a listing of those parameters?

Can you compare the configuration?

How much grief would you have if you completely "stupidified" the machine followed by a re-load and boot?

Would you have to re-teach it? Or, is there a file, somewhere, that contains the last saved "operating configuration" which can be re-loaded?


Steve Bailey,

Do you know anything about these units?

Is it possible to output a configuration listing to a printer?

Sorry Terry,

I have plenty of experience with GE Fanuc PLCs. I have limited experience with Fanuc CNCs. My closest exposure to Fanuc robots is having seen a few of them. I expect cdlove23 is on the right track in thinking that there is a parameter in the robot controller that needs to be set to allow the 'monitor' function. That's just the way Fanuc tends to do things. All features are present in all controllers, but you have to enable the ones you want by setting or clearing bits in control words. The CNC techs know which words control which functions, but Fanuc doesn't publish the information for the end user. The longer you've been using their controllers, the larger the list of parameter settings you accumulate.

For example, in the CNC controllers, there is a parameter named 'Parameter Write Enable' (PWE, called peewee by tech support). In order to change values, you have to set PWE. When PWE is set, you can't run a program. That one is documented in the manuals, but unless you know about its existence, it's easy to miss (knowing enough to ask the right questions).

I expect Fanuc robot controllers are similar to the CNC controllers, but having never worked on one, I don't have a clue where to start looking.

About the only suggestion I could make would be to try swapping the pendants (if there is one pendant per robot) from the robot that has the problem to one that's working properly. That might help if the problem is in the pendant, but nothing that cdlove23 has written indicates that's where the problem is.

Another possibility that came to mind while composing this: Is the ability to use the monitor function something that has to be included in the application program? Is it possible to run the program in question on a different robot to test this?

SOLVED!!!

Ok I got this finally figured out, but let me see if I can explain it. The reason that I couldn't monitor the program because it was actually running a macro. This won't make sense to anybody unless you have experience with Fanuc robots, but here goes....

The tech rep said two things that that are going on. One: "you can't monitor the program because it's really running a macro that calls up the programs" and two: " there is a bug in that old style software" Number one is the reason why i couldn't monitor the sequence of the program and two just makes me nervous.

This robot in the past has had some difficulty getting started after a fault or operator intervention. It used to be that when the robot would stop and if you wanted to restart you would have to select the program using the teach pendant and get to re home and then select the run program. Well for the tech. and operators running the line this was way out of their scope so they would have to stop the production line for a while to find someone qualified. The "quick" fix at the time was to program the robot in the startup sequence to run a macro, and the macro decided what program needed to be run. So now if the robot stops or crashes into something the operator can just restart the robot (kind of like rebooting) and the macro takes over. The long-term fix was to reprogram all the programs and points to handle a recovery operation. Something that isn't on the drawing board. We never have any problems with this robot, this was just one of those bells and whistles that should of been working but wasn't.

This also explains why I was getting the weird errors, because the macro was trying to call up programs that were not there anymore. Thanks for everybodys help and thanks Steve for the idea on the pendants, and for easing my mind about fanuc and the mystery settings. I thought i was going crazy for a minute.

cd,

A macro is just like plc logic. In most robotic applications it runs continuously in the background scanning the logic just like a plc scan. It's also called sequencer files or jobs by other manufacturers. The tech is right that you usually cannot monitor the logic in the macro, but you can monitor the addresses or I/O's that are apart or in the logic. The macro and the articulated programmed steps should be two different things. They can work in tandom or totally seperate. I have know idea what Fanuc was thinking on this one, unless your using a very old robot model.

A robot is only as smart as its teacher. If the robot was taught to do a task or job then why in the world would it "forget" where it was when it crashed or was put in a hold state? It sounds like the macro is not working properly either. The macro file should have all the what if logic in it to run properly. A band aide will get you going, but the restart needs fixed. The software problems are very common. Usually the manufacturer won't call you to tell you that they have a glitch in the software, unless its a safety issue. If you get the upgrade free, more power to ya, but usually they'll want you to buy it.

Good Luck Cd,
Tim

PS. Thanks for posting a robot question, this is one of my more stronger fields.