Posts by gpunkt

There are 2 separate MOLEX PC-applications that could come in handy.

PFN TP CT (ProFiNet Teach Pendant Configuration Tool?): which is an application that looks and works exactly as the application that's running on a teach pendant. To use it, the PC needs to be connected by ethernet to the controller's CD38A or CD38B port.

PFN CT (ProFiNet Configuration Tool): which is for creating the libraries that's needed for the I/O-controller configuration. Downloads directly to the profinet card.

Both of them are covered (or at least mentioned) in the Profinet Manual for FANUC robots.

Only tested the functionality once or twice. Basically, you save the TP-program you wish to edit as an .LS-file, then you edit it, and then you load it back to the robot and it will compile it into a .TP-file again.

How you load it to the robot can be done in a few different ways, I assume.

Quote from zentz

Hello

I'm trying to simulate in Roboguide a backup I just got from a costumer, the following error shows and I can't continue.

The robot model is LR-10iA/10, how can I add the model to my files or what can I do to simulate it? Im using Roboguide V9.40P/03.

I don't think that v9.40P/03 supports this robot model. It was released at a later stage. v9.40P/03 i probably 5 years old by now...

Also, virtual robot controller version is one thing, and Roboguide revision number is another.

I would advise you to get a hold of the latest revision of Roboguide, then you will also have the latest virtual robot controller.

The robot model must be supported by the virtual robot controller AND be available in Roboguide's library.

Quote from MoEL

Hello,

Can you please share the "hacky" workaround, I stuggle everytime during the first tests and commisionning

Thank you very much

When using "External Mode Select", the operating mode is selected depending on the state of two Safe System Outputs:

You can create simple logic in Safe I/O Connect where you manipulate the SSO[6] and SSO[5] with two NSI-signals:

These NSIs are mapped to two DOs (Rack 36, Slot 0)

Then either manually turn these DOs on/off or write a neat program that does it for you.

The physical outputs on the CRMC20 terminal is represented logically by the system at RACK 48, SLOT 1 and START 1-8 according to the following table:

Note that the terminal # are not in the same order as one would think (because of the physical layout of the terminal block).

A key (pun intended ) point is if it's only the key itself thats missing, or the entire key-switch (mode selector)...

If it's only the actual key, get a new one (they're cheap).

If it's without the mode selector, then the robot should have the necessary software to accomodate the switching of operation modes. As suggested by Erik Olsen , check the DCS-manual for reference.

You can check in orderfil.dat that is part of an All Of Above backup.

Or you can view this file in the TP: Menu->Next->Status-Version ID
Then press the NEXT-button and choose (F3?) "ORDERFI" to display this file on the screen.

The option you should look for is the one ending with R764. If that line starts with an exclamation mark (a ! ), then this line is regarded as a comment, and thus that option is not currently installed.

There are remote switch boxes available for FANUC's controllers that you can order from FANUC. They will replace the Operator's Panel in the cabinet.

But for simplicity, your solutions seems valid.

However, pushbuttons with Normally Open contacts (such as the Cycle Start button) will have to be wired in parallel, not in series.

Also, there are no red buttons, only a red lamp (for indication of errors).

Depending on what you're looking for in the vision process, determining the orientation of a round hole can be quite hard.

Are you able to correctly position the gripper at the correct position above the hole (X/Y) when only moving the part in X/Y-direction (translation) without rotating it?

You can use the HOLD-signal to pause the robot's program. But simply toggling this will not start/stop the robot. The robot will need a start command to start/resume again.

Or you can look for the request-to-enter-signal in the robot's program and if present set pulse an output from robot "ok-to-enter" and then wait for req.-signal to become OFF. While the robot is waiting, the door is unlocked and the Auto Stop will pause the robot (as well as throwing a FENCE-alarm). Once the door is locked and the safety is reset, the PLC will reset the req.-signal and then send the start-signal to the robot.
The robot can now continue the program (req.-signal is OFF) and starts by setting ok-to-enter-signal to OFF and then gets on with what it's supposed to be doing.

If you want to keep track of time while no TP program is running, I would suggest a BG Logic program that does that for you.

One simple way would be to wait for the conditions to apply, and then start a counter that will increment each cycle (8ms), then when the TP program is started, the increment is stopped and the time is calculated.

Quote from prince2409

Hi,

can I have the gsdml file, please

With J709 Profinet iDevice, you need to create your own .gsdml-file in TIA Portal. The procedure is covered in the Profinet Manual

CP1604 (FANUC option J709) must have FW 2.7.2.0 according to the Profisafe-chapter of the DCS-manual.

Try searching for "FANUC iRVision" and you should be able to find more info. Here are some interesting videos:

The term used in FANUC is "Snap in motion" where the object that is being photographed is moving in relation to the camera.

To overcome motion blur, there is a formula: when moving with x mm/s, use y ms exposure time (also it takes into consideration what lens you're using and some other factors). I think this formula is presented in the manual.
But basically the faster the objects are moving past the camera, the shorter the exposure time needs to be. And the shorter the exposure time is, the more intense the light needs to be.

Get in touch with your local FANUC rep and ask for the iRVision reference manual. Also ask them what they can offer in terms of education or workshop related to your needs.

If you are trying to cut cycle time, remember that it's not always necessary to move fast, but more important to minimize the stops. So it's probably better (cycle time-wise) to have slower but continously motion, than fast motions with FINE termination.

Hi,

Please explain what you mean with "dynamic shooting".
Do you mean that there will be a fixed camera that will photograph something that the robot is handling, while the robot is moving? Or will the camera be robot-mounted and take a picture while the robot is moving?
Or will the camera be fixed mount and take a picture of a static object while the robot is doing something else?

Your robot controller needs to have the vision hardware (Mainboard with COAX connector) and you will of course need the camera and coax-cable.

The video you uploaded didn't provide much insight into what you want to achieve. All I saw were robots moving fast and doing stuff. Cool, but not very informative.

Hi,

You state that you want to know more about the SR-6iA, but for us to comply with your request, we first need to know exactly all and everything that you already know. Otherwise, we might feed you knowledge that you already have, thus you wouldn't know "more". And wouldn't that be a shame?
Please state in detail all your knowledge about this robot, and then we can try to make you know even more about it!

All vision processes are done "on the fly" by starting a parallel task in which it's being executed.
This can be very useful, but it can also be unwanted. Depending on your application, you might want to perform the vision process (image acquiring and processing) either with the robot at a standstill or while the robot is moving.
Furthermore, if you need to snap while moving, will the robot's motion influence the picture?
If so, you will need to pay attention to the exposure time of the snap to not get a blurry picture. Probably it will have to be a short exposure time, which will require a lot of light, so make sure to have proper lighting!

I would suggest asking Google about "FANUC iRVision" and having a look there. I bet there are many videos out there that cover both the features and how to set them up.

That's basically the way robots execute the programs, line by line.

If you're used to the way that a PLC is executing (reading every line of code each cycle scan), then you better think again.

In a normal TP-program, each line is read and executed one after the other. If you have an "IF"-instruction, then the condition is only evaluated when the program pointer reaches that instruction.

If you want to monitor something "in the background", you must either put it in a background program (called "BG Logic") or you can use "Condition" type programs which are used to monitor a certain condition, and if this is satisfied (at any time), a specified TP program is called, interrupting the current program:

(Pseudo-code)
MAIN.TP
!Start monitoring
MONITOR CHECKFORSTOP
!Stuff happens in the main program
...
!Some motion instructions
....
!Something else..
...

!Monitoring ends
MONITOR END CHECKFORSTOP
END


CHECKFORSTOP.TP
!Check if Stop-button is being pressed
WHEN DI[1:Stop Button] CALL STOPPRG
END


STOPPRG.TP
!This program is being called
!as an interrupt when the stop
!button has been pressed
PAUSE
END

Or you can choose to go to Remote control state, where you map an input as an actual stop-signal through the UOP-signals

ZERO position mastering should never be used imo.

However, I doubt that it would affect the outcome of any subsequent mastering operation.

When performing a Quick mastering, each axis is set to the joint value written in REF_POS.
When performing Zero position mastering, each axis is set to 0°

Check for the original master data (from the documentation that came with the robot), or at least from a backup from when it was last ok.

Assuming there was no mechanical intervention in the drive train since the data was stored, punch in the same numbers in REF_COUNT.

Make sure that REF_POS is set to 0 on all axes.

THEN perform the Quick Mastering.

As someone stated, the actual pulse count is lost when BZAL occurs. But being within one pulse coder/motor revolution from the reference position (which should be 0° per default) , the system can calculate and apply the necessary offset to the new pulse count and then perform the mastering from that.