Tool Deviates From Program

[CarbonCAD]

I need to get some opinions on this.  I do not know where to start to look for the error because it didn't do it the first part that ran but did it on the second part that was ran.  The tool goes around the fin with cutter comp.  Pretty simple, but as you can see it went through the fin and cut through the material (or lack of) designated in white.  I was not the one running the parts, so there may be more details I do not have yet.  Mori Seiki NV501FD3774 with a MSX-501III controller.  Any ideas?

 

[Newbeeee™]

If your're running comp and someone changed the tool diameter in the offset register, has that casued the machine to error on lead in/lead out?

[betts]

Lots of variables there to account for. Top guess (assuming as newbeee said that no one changed anything) would be a modal that wasn't canceled between first and second part. A modal plane change for example that wasn't canceled because there isn't an appropriate safety line spit out by the post.

 

Just a guess but if you share the code it would be easier to see what is happening.

[CarbonCAD]

It sounds like no one changed any offsets besides possibly a Z offset.  I wouldn't think that a Z offset for wear would hurt anything?

Now that we ran some more I'm more comfortable that we did not start and stop in any way to keep something modal.

Snippet of the code:

%
O0000 ( TEST )
( DATE: 10 - 09 - 14 )
N1 M01 ( 3.0MM ENDMILL-3/8 FLUTE   )
G91 G28 Z0 T3
M6
G17
G90 G80 G40 G0 G54 X-6.5154 Y-.5199
G43 H3 D3 Z.1 S3000 M3
M51
G5 P10000
G1 Z-.0787 F10. M51
G41 X-6.5596 Y-.5641
G3 X-6.5154 Y-.5824 R.0625
G1 X-5.833
Y-.6393
X-7.1978
Y-.5824
X-6.5154
G3 X-6.4712 Y-.5641 R.0625
G1 G40 X-6.5154 Y-.5199

 

G5 P10000 a culprit maybe or does it not come on until a certain rpm?

[Guest MTB Technical Services]

The code is fine, meaning it will run.

It's generally considered a bad practice to call your CRC register with the TLO call.

You are then stuck with the modal value.

 

HPCC wouldn't have caused this.

 

This looks like operator error on the TLO offset entry and they aren't coming clean about what they did.

[MIL-TFP-41]

I need to get some opinions on this.  I do not know where to start to look for the error because it didn't do it the first part that ran but did it on the second part that was ran. 

 

Someone changed something. Either the program, tool offset (height or dia). A machine does not run something fine once and then do something weird. Someone changed something & is not coming clean.

[gcode]

_{are you using wear comp or control comp}

[Joe788]

Call Ellison and ask them if that machine has a keyboard logger.

[crazy^millman]

 A machine does not run something fine once and then do something weird. Someone changed something & is not coming clean.

 

Sorry, but I have to disagree with that statement. I have seen machines do some very weird things in the past. Maybe not the case here and your last statement I have seen time and time again.  I once got on the phone to call the police about an issue in a shop was not till I got on the phone to report a theft did someone go get the scarp part they hid to not get in trouble.

[brandon b]

When we moved to a new building, we where cleaning out the old building and found all thoses parts that " that we never received...or I have no clue about those parts "thrown up in the rafters buried in the insulation on top of the office's

[brandon b]

When we moved to a new building, we where cleaning out the old building and found all thoses parts that " that we never received...or I have no clue about those parts "thrown up in the rafters buried in the insulation on top of the office's

[taperlength]

I'd venture that the comp value was changed a little before the second part.   The program leads in with a straight move of .0625 length and then a circular move with .0625 radius.

The tool is supposed to be a 3mm diameter which has an .059 radius.  If the comp was changed to be more than .0625 for some reason, like it was mixed up with a 1/16" endmill, then the comp would cause that gouge.

[CarbonCAD]

It's control comp.  I could see it getting mixed up with a 1/16.   It could still be related to the comp considering there was more than one shift working on it and they could of tweaked a little and not communicated clearly to the next guy.  We ended up with enough parts to fill the order, so I'm moving the problem to my "to do list"  I am not ruling out a machine error just yet either. 

 

What I like is when something is scrapped and we have more material magically appears (tucked away somewhere).  Hard to argue with it - sometimes the scrap is my fault.  Pot calling the kettle black kind of thing.

[Guest]

Ran a FADAL once that literally had a mind of it's own. Run OK for a dozen parts or so, then X-Axis would run-away randomly. Not just on me either.