Strange PathPilot G54 Error?

[Peter Neill]

I've just upgraded to PathPilot from Mach 3 arrived about 2 weeks ago, latest version 2.0.0. I've also just moved from Sprut7 to Sprut 11, but that may be incidental.
Machine is a series 3 PCNC1100, with very low hours.

Anyway, on to the problem.
I went to make a part I've made 50 times before, but always using the Mach controller. I set up my G54 using the Haimer, ran a trial part, and found that the Y0 position had moved by -1.5mm. X0 was still OK.
Re-checked the zeros, checked the Haimer concentricity, checked G54 was still correct with this, ran another trial part. And again Y0 was off by -1.5mm. Re-set everything again, tried once more - same thing. The code is fine btw, no errors there.
All the features machined were absolutely perfectly on size and perfectly in relation to each other - just offset by-1.5mm in Y.

So today I grabbed a piece of aluminium flat bar 50mm wide and manually scribed a centreline from Y0 to the middle (25mm), then scribed lines from X0 at 25, 50, 75, 100, 125, and 150mm.
Clamped in the vice, set G54 again using the Haimer, then put a 90deg point tool in the spindle. jogged around a bit, then typed X0Y0 in the MDI, and brought the tool down to just indent into the aluminium bar. And again, it was offset! But this time in both X and Y..
Typed Y-25 in MDI and scribed a line along X using the tool in the spindle. Then did the same along Y at the above intervals.
This time Y was off by 0.677mm, and X was off consistently by 1.75mm at every marker line, in the same direction. And yes, I can measure this very accurately with the optical inspection equipment I have before anyone asks.

I'm waiting for a call back from Tormach Technical Support at the moment - but in the meantime, anyone have an idea why?
I'm pretty sure we can rule out any couplings slipping, as the features machined were perfect on every occasion, just consistently offset from where they were supposed to be.

Similar Threads:

• Pathpilot error message
• strange error code
• Help with a strange error message
• New Machine Build- Strange error in machine defintion manager (Error occoured in IOF chook: )
• Need Help!- Strange sum error on VMC40

[syscore]

How exactly are you setting up G54?

In PathPilot, you move the table/head to where you want zero to be and then press the Zero-X and/or Zero-Y and/or Zero-Z buttons, depending on which axis/axes you are zeroing. Is that what you mean by "setting up G54"?

Can you try this...

1. Put stock in vice.
2. Using Haimer, find left edge of stock.
3. Press Zero-X. (X should now read 0.0000 in PathPilot)
4. Raise head and jog around some.
5. Type G0X0 in MDI
6. X display should read 0.0000 and the tip should be right over the left edge.

Repeat this for Y.

[Peter Neill]

I'm sure you didn't mean for that to come across as condescending at it sounds
But to put your mind at rest, G54 is always at the top left corner, and all axis zeroed when set. I've been doing it this way for 40yrs...
And after it's set like this - the zero offsets itself!

Had a long and un-fruitful chat with Tormach last night, and they asked me to repeat the exercise again.
So I did, and this time both X &Y were offset by 0.66mm/0.054".
Picture below, showing hand scribed lines and then lines scribed with the diamond drag engraver 0.002" into the work and positioned using the MDI.
Note the very first line at the far left, scribed after setting the MDI to X0Y0, after setting zeros off the side edge and top edge.
Weird...

How exactly are you setting up G54?

In PathPilot, you move the table/head to where you want zero to be and then press the Zero-X and/or Zero-Y and/or Zero-Z buttons, depending on which axis/axes you are zeroing. Is that what you mean by "setting up G54"?

Can you try this...

1. Put stock in vice.
2. Using Haimer, find left edge of stock.
3. Press Zero-X. (X should now read 0.0000 in PathPilot)
4. Raise head and jog around some.
5. Type G0X0 in MDI
6. X display should read 0.0000 and the tip should be right over the left edge.

Repeat this for Y.

[syscore]

"I'm sure you didn't mean for that to come across as condescending at it sounds"

I was just trying to remove all variables. You didn't exactly say how you "set G54" and I was making sure that there wasn't something you got into the habit of doing with Mach, but that doesn't work with PathPilot. You said that the part was perfectly sized, just shifted over. My first guess would be that your Haimer is off. Seeing that everything is offset the same, would indicate that you are getting the error during the zeroing. Try setting G54 using just the 90 deg point tool and visually placing it over the top left corner, zeroing, move around, then G0X0Y0, and see if it is off. My bet is that it won't be.

[syscore]

Shorter version - it seems that your zero is holding (meaning that you can go back to it, or to any point relative to it, accurately), but it isn't over the top left corner as you expect. But that is according to the Haimer. Maybe the Haimer is off and you can test this by using the 90 deg point tool and setting your G54 manually.

[jmosley]

Maybe try zeroing g54,g55,g56 and so on in your first step to test if the offset is somehow related to Mdi confusion with another coordinate system . Or do your entire test with jog shuttle - no typing coordinates.


Sent from my iPhone using Tapatalk

[bhurts]

.66mm is .0259 not .054". Other then that it's likely the haimer needs it's run-out adjusted. That would explain why the offset varies from setup to setup. Whatever the problem I hope you get it sorted without to much frustration.

Ben

Sent from my HTC6525LVW using Tapatalk

[kstrauss]

As with the other poster I don't want to sound condescending!

Did you by chance replace your Haimer probe stylus? Using an Imperial probe stylus on a metric Haimer (or vice versa) gives about that amount of error. When testing the Haimer as described in their documentation everything will appear perfect. Don't ask me how I know about this but is there anyone in Canada who wants some, almost new, Imperial Haimer probes for a good price?

[shred]

Also check the Haimer stylus is in tight. That drove me nuts for a while but it was just a tad loose.

[Peter Neill]

Just to update you on this - the Haimer was running concentric to around a thou', no damage or loosening of the tip (same metric tip for 5 years - not changed), and worked perfectly.
The problem was exactly the same when I used a mechanical edge finder, random but repeatable offsets.
I've taken the Pathpilot controller off for the moment and gone back to using the old Mach controler for making the parts, and these were perfect.
Over Easter I'll reload the PathPilot software and run the tests again.

[mountaindew]

Did the offset in Sprutcam get moved off the place set? I recently changed cad systems and It allows a part import process that is faster and better. I have to be careful now where I place my offsets on my models in cad. The other day I noticed all sorts of funny numbers showing up on known heights in Sprutcam as I was selecting surfaces. Went and looked and somehow I had moved the offset in cad system just a tiny bit. It is easy to move and set this way but it also works both ways and its easy to mess it up.

That said offsets are not easy to change in Sputcam program and require close inspection at times when you import models and locate them. But when set your not going to inadvertently move them like I did in cad