I see that this is and old post but I'm having the same issue at random. I see that the thread here is marked as resolved. I wanted to try here before making a new thread. Would love to hear what resolved the issue, or what could be some things to look out for. Encoder seems to be incrementing just fine, I have also not been at the robot when it throws these faults. Any help is appreciated.
Posts by ezraseth
-
-
just an update guys, so I did hear back from Fanuc. As you all have stated its "impossible" (i have my doubts about that word but Fanuc will never release that information if they have it or not) to decompile. Its between myself and the 3rd party. (Never sent in a ticket to the installer company but may in the future just to see what they have to say)
I didnt get a chance to really look into the program to see what it does but hopefully i will be able to tomorrow. I have a feeling it may be very close, if not exactly the same logic as the built in ATAN2 code...either way as soon as i figure something out i will report back.
-
You are talking about 3 parameters for your custom ATAN_2 function. Have a look on the ATAN2 example from the KAREL reference manual. This is an example with 3 input parameters. You could copy the example code and create your own function. You could then compare the result from your own function with the ATAN_2 function by using the same values for the arguments. Maybe you have luck and it is the same. It's worth a try.
That is a great idea. I just started reading into the KAREL manual. I will CTRL+F "ATAN2" and go ahead and try exactly what you suggested. That will be the best way to know if its the basic Atan2 equation or if theres some other stuff going on in there. I will report back with my findings if i get a chance tomorrow! Thank you again!
-
KAREL hast a standard built-in function called ATAN2. Your ATAN_2 could be something similar but maybe extended with some user-specific stuff.
i will take this into consideration as well. Im excited to see if this is true. Ill have to look, but i may just be able to pop in the 3 arguments the program calls into the ATAN2 formula and figure this all out. Bad thing is i know for a fact on other robots we have that sort/create layers are using .KL...those are most definitely not something generic like this one but i will cross that bridge once i get there. This robot is one of the easier ones (so i thought) so i wanted to use this one to step through the program and make note of what each line of code does.
The sorting robots for creating layers are using Line Tracking. I figure Id save those guys for later down the road. I have a very basic understanding of Line Tracking, but will have to go through the manual again once i focus my attention on those. Dont worry guys i know who to ask all of my questions to now! This forum is a beautiful thing. I have noone to rely on at work for questions like this.
-
By the name, one would assume it is the standard atan2 math function. You can Google it. It would be useful for calculating an angle to rotate a frame.
I briefly just looked and i may be able to figure it out via the arguments it calls with that program. That will be the first thing i look at come Monday. Looks like i should have just stated the program name from the beginning. Its a stretch but im going to try and figure it out. Glad to have rockstars like you guys around.
-
By the name, one would assume it is the standard atan2 math function. You can Google it. It would be useful for calculating an angle to rotate a frame.
Perfect im on it. Thank you!
-
There isn't a decompiler.
What is the name of the program?
If it's not part of pallettool slipsheet program it's 3rd party and Fanuc will not be able to help you either.
Also,the way i see it, if it can be compiled, it can be de-compiled. That's way over my head but im assuming if it can be done one way it should be able to be brought back the other.I'm just a novice but that's how it seems.
-
There isn't a decompiler.
What is the name of the program?
If it's not part of pallettool slipsheet program it's 3rd party and Fanuc will not be able to help you either.
The name is ATAN_2.PC it is not part of PalletTool...(at least im pretty sure its not). I guess thats it then. I will however let you guys know what Fanuc says if its anything worth reporting. The last very small but possible way would be to put in a ticket with the installer company and see if they would send me the source code. HIGHLY unlikely but again, i want to know that i literally tried everything to see this program and i left no rock unturned. At that point i will know i did everything i could and then just move on.
-
Couldn't hurt to ask. I would not be surprised to find out if Fanuc has a de-compilation tool. I would take it as a long shot though.
Thats about how i feel about it...what would it hurt at this point? My last point of contact said he reached out to his colleagues....(he works for the same company but a different branch) and they said that thise programs come straight from Italy where they are programmed before sending out. From looking at these programs....they are VERY generic. There are programs that are loaded mut not used at all. We have a single infeed palletizer on each line. There are programs set up for double and triple infeed lanes that we dont have. Those programs sit idle in the controller. The way i see it the guys come in and after the mechanics mount everything in place they tweek a few positions and so some moderate PLC work. They are great guys. I worked along side them to learn everything i could. But they refuse to give me the source code to the .PC programs. I guess i will try and email Fanuc tomorrow....hopefully its not a bust. The guy i was speaking of earlier assured me that if the company i work for is the end user of the robots and we have no ties with the installers that they should be able to provide the info. The guy pretty much told me that in house Maintenance and Automation work on those robots. If they had someone from the installer company there 24/7 then it would be a different story and there would be no obligation to hand over the information, because the imbedded technician could solve any of the problems....bit that's not the case.
-
Okay guys it me again, i was told by a previous Fanuc employee(early 90s to early 2000s) that as long as the company i work for is not "leasing the robots"( which we arent) and have access to CRC/hold the "license to the robots" (which we do...i think...) that any .PC programs should be able to be "de-encrypted"/ "de-compiled" to .KL. I thought it may be fishy but cant say for sure....is it worth reaching out to Fanuc for this issue? Learning these robots is my passion, and i want to exhaust all possible options before throwing the towel in and accepting that i will never see the programs... What do you experts think?
-
Perfect! Thank you. I will do just that!
-
hello again. Quick question. I cant seem to find what the meaning of Variable $EGD_CFG.$ERR_SV . Hoping someone has some insight on this one? Thanks.
-
thanks guys for all of the explanations. i have one more programmer that i may try and ask but its gonna be a long shot. Funny that they removed the .KL files. I would really have liked to see whats in that program. Its must be highly top secret. If i get access to it i will be sure to report back to you guys. Im not getting my hopes up though. Thanks again everyone!
-
The *.KL file is the source code and is written in a human readable format with a syntax similar to Pascal. The KAREL source code cannot be executed on a robot controller. For this, the *.KL file must be translated into a *.PC file (p-code).
Without having the *.KL file you have no chance to see the source code or make any changes.
So what you are saying is the *.PC is unreadable and what i need is the *.KL of that program? So just curious again. If i was to get that what does it look like? Is it similar to statement list/structured text in a PLC? I really need to do my research on this but luckily i have you guys here to help. I again want to thank each of you for your input.
-
KAREL is a compiled language and is similar to Pascal. The *.KL source file must be translated into p-code before it can be executed on a robot controller. Once the file ist translated to a *.PC file it is no longer human readable. It is then kind of a "black box". Having the *.KL file is the only way to get an insight into the source code.
So its not a "code" that needs to be entered....i literally just need the *.KL file?
-
A .kl file is a text file. You can open it in any text editor. I use notepad++
again i apologize for my lack of knowledge but at what point would you enter this "code". Is it literally like a password or what? I understand the guys wont give me the code but im still confused on where this code needs to be entered or whatever. I understand it is encrypted. But how could i make a.PC file me readable to a human with some sort of code? Otherwise,as you all know, it looks like hieroglyphics...
-
okay. Installer is not going to give up the source code. If someone would be so kind, how exactly would it work if they did? I would take a backup of the .PC file and open it up on my laptop using what? Roboguide? Some special KAREL software? Where would this code need to be entered in order to view this logic? Im just trying to make sense of all this. Sorry for the lack of knowledge...
-
You should ask for the *.KL file (KAREL source code) for the compiled *.PC file. For example, if you have a Test.pc file running on your controller you need the Test.kl file to read the code inside.
Perfect! Thank you. I hope i can reach the guys.
-
i guess im going to try and reach out to the programmer. How exactly should i word what i need? Thanks again guys.
-
You won't get it with an AoA backup. A copy must be provided from the original programmer, then store it in a safe place.
So my only option is to contact the programmer? They most likely wont give this information up. There has to be another way...if not, then i guess my studies with this particular robot is over. Such a simple job that it does. Yet i wont be able to see what links the calculations together. I may try and reach out, but the chances are slim to none that i get a response...