Problem with postprocessor

The "Commanded acceleration" errors are typical of an Axis 5 singularity issue during a LIN or CIRC motion. Interpolated motions that take A5 close to 0 must be avoided.

The "Commanded Torque" errors are a bit more unusual -- the path planner appears to believe that executing the current motion will exceed the torque limits of the axis. This is a predictive error rather than a realtime error. I would recommend checking the acceleration settings and the load data.

Have you performed a payload analysis of your end effector? Mass, CG, and inertial moments must all be entered into the LOAD_DATA array for the tool number you are using. Does your robot have the Load Determination tech package installed?

Calibration depends entirely on your situation. The Base and TCP data inside the robot must match to what is in Powermill.

As for the A5 singularity, there's only one way to address it, really: Don't make that move. If your Powermill simulation is driving the robot through a linear motion that drives A5 too close to zero, you can't make that move. The only alternatives are to reduce the robot's speed enormously (which, as A5 approaches zero, becomes less and less possible), change the move from LIN to PTP (not possible during a milling cut motion), or allow the TCP orientation to "float" during the move (also not something you want to happen during a milling cut motion).
There's nothing you can really fix in the postprocessor to handle A5 singularity -- this issue has to be addressed at a higher level. Usually, when simulation shows that a required motion will drive the robot too close to A5 singularity, the tooling design is altered in such a way that all critical LIN motions can be performed without A5 coming close to 0.

Hi vvelikov,

This is an old post and not sure if your still around or found a solution to this... if so I would love your input.

I am also experiencing problems with programs created in powermill having the same axis limit errors 1447 axis limit reached (Kuka kr30 with KRC2)

Powermill simulation is not showing any singularity problems.

The output file is a series of PTP points (which as I learn more doesn't seem to be right for a milling operation.... given the robot will not necessarily take a direct line between points as it would with Lin command.. )

At this stage I am trying to figure out if it is the Status and Turn causing a problem or a singularity.

Base and a Tool data are correctly calibrated/used on the robot as trials with manually created programs run correctly.

Some of my forum reading said that PTP movements are not affected by singularity.

I would like to hear about your experience or others with similar setup/experience.

Message 1447 has definitly nothing to do with singularities. Rather it is a tool, base or turn issue. As you said tool and base are already verfied to be the same check turn.

Fubini

Thanks guys!,
I will look at what Turn info it is producing Fubini, good to narrow it down.

Vvelikov, we are using powermill 2016 and it appears it was a custom made Post from Delcam produced way before my time. When back at work I will find out some more details.

The robot has not been running for a long time (before my time).

Another person who saw it running said it was very slow... I think this might be due to a huge amount of PTP data produced (drip fed by CamRob) instead of LIN commands... thanks for your confirmation about this being incorrect!!

I did do the calibration for Tool orientation once I had used XYZ 4 point for the TCP. Once both of these were done I then followed the process for defining the Base (thanks for checkin!)

When I get a 1447 error, axis limit reached... I did experiment by manually entering the next PTP manually minus the external axis and Status and Turn data ie PTP {X Y Z} and the robot then move without a problem to the new point!!!! Very annoying!!

It appears to want to approach each PTP from a horizontal rather than vertical orientation and I guess this is what is pushing the axis past the soft limit.....

When importing the model into powermill the axis is in solidworks orientation with Y axis vertical. I then create a workplane aligned with my Base system (z axis vertical and x and y plane creating the surface on which the part sits.... I thought this is all that would be required with all PTP data created relative TCP and this Base/workplane.

My tool axis in powermill and Kuka are not the same... kuka using standard x axis for tool direction but powermill is using z or y. I didn't originally think this would matter but maybe because of the Status and Turn information included in the PTP command it does!!! (Learning heaps... and so much more to go!!). I haven't been able to find how to change the tool coordinate axis orientation in powermill but did find in some reading tonight that the MTD file might be the place to do this and this is why I couldn't find an option within powermill frontend.

thanks!!

phillfalco

Have you had any success with powermill?
Do you still using powermill for robot milling?
Thank you

Hello

I use powermill, I had the same problems even gentlemen of delcama could not master it.I started to do the trial and error method myself, my wniosek is a well set up utility and database and typing this data well into powermilla.Attach my postprocessor that I use

Hello, i've have this problem and i've solved change the ptp xyz for the ptp a1, a2,a3... only for review if the robot reaches the position correctly the problem almost always is the base or the coordinate with the code was used can be tool, base, dynamic base can you review with the actual base and tool and position of robot for review.
Best regards