[Motorsports-X]

Hey folks. I have a older VF2 (the one with the old control) and we are having problems biting us with the thermal comp. #1, we arent, and should be running a full warm up program in the morning. I don't know as this will help us however, since the machine might get used for an hour in the morning, and then not again till the afternoon. if it stayed on operation all the time, id just let it max its growth and call that 0, but this machine doesnt get used that way.

Right now, if you machine a long program, the longer the program, the deeper z cuts. Ive seen up to .007" after 2 or so hours in a surfacing program. At first I chalked it up to "old hass, lost steps:" etc.. but on another haas, program cut perfectly. If you put an indicator on the tool and wait, you can watch the tool move back up. This tells me the thermal comp is working, its just not working correctly.

I know there is a setting for the coefficient, and the last PPE, the tech said that "the number didnt look right" in the z axis. Its at something like -8000000 right now. I have been unable to find instructions for adjusting this number. I also believe that setting 109, (time since warm up) also is used. But I'm not sure how this number applies to the math formula the machine uses, or how it would need to be adjusted in relation to the thermal coefficient parameter.

Any help here is appreciated!

Edit: this is z axis parameter 133.
Edit 2: setting 109 is not thermal related on this old control as far as I can tell.

[Motorsports-X]

we have a vf3 manual on the shelf. it tells to set 133 at -800000, but thats for a machine with temp senors. not our machine. I wonder if someone tinkered with something not knowing what they were doing.

[Motorsports-X]

talking to myself.. for historical purposes.

running this code.


%
T4 M06;
G00 G90;
G43 H4;
Z10.0;
Z.25;
G01 Z0.0 F10.0;
G04P1.;
M99;
%


Been running for 15 minutes with an DTI on the shoulder of a tool holder, and its already moved about .0006. Letting it run for an hour or until I feel like axis only movement based on thermal comp is or isn't the problem.

when this is done, Im going to run the same program with a Z home line and let the run for a few hours tonight just to make sure its not an encoder count problem or something unrelated to thermals.

In the morning. Ill add in some spindle.

[Motorsports-X]

Well,. the longer that program Ran, The worse it got. I stopped it after 2 hours and it was almost .002" deeper. Now here's the thing. From my understanding, haas thermal comps are time based. when you stop movement the thermals reverse slowly, not all at once. after I feed held that program with tthat tool in the air, i turned around to talk to someone. then i turned back and pressed start button again, so that I could catch it during the dwell and watch the needle move back to zero as it cooled. The thing was, it was no longer .002 off. it went back to 0 with no cool down time. It was never homed.

Im at a loss for whats going on here.

I do know what will fix the problem for sure...... a new machine.

[Fletch_CNC]

I'm definitely not an expert on thermal comp, but could it be that the T Constant is too high? That's parameter 274, I think. That would mean it is decaying the applied comp too fast. So, the ballscrew grows, but not enough comp is applied. Just a thought. On my machine, the Z thermal comp is -225000, and the T constant is -63000. There is also setting 160, which is an adjustment to the Z axis thermal comp (I think you can make it +/- 30%). Don't know if any of that helps, but I thought I'd throw it out there.

[Motorsports-X]

I'm definitely not an expert on thermal comp, but could it be that the T Constant is too high? That's parameter 274, I think. That would mean it is decaying the applied comp too fast. So, the ballscrew grows, but not enough comp is applied. Just a thought. On my machine, the Z thermal comp is -225000, and the T constant is -63000. There is also setting 160, which is an adjustment to the Z axis thermal comp (I think you can make it +/- 30%). Don't know if any of that helps, but I thought I'd throw it out there.

We just ran a slow program for 2 hours. This program would put very little heat into the ball screws compared to the rapid moves i was doing yesterday. with the spindle running and the axis moving after 2 hours the z was .006" deeper than set. This was just based on touching off a part and viewing the offset amount.

In this scenario, you would think that comp would make the tool shorter, as you have less heat than anticipated but comp still active. that isnt the case however. its almost like its comping the wrong way.

I will check those parameters and compare them to mine. Which machine do you have?

On my machine, setting 160 is a fuse level parameter. setting 274 is t const. and is set to -60000

Im going to try setting my -8000000 to your -225000 and see what happens.

[Motorsports-X]

OK I adjusted that and ran it again for an hour and a half, after it appeared that it was cooled back to original zero. Over an hour and a half it grew still about .0037" Still in the wrong dirrection.

I looked through the paramaters and there is a 268 Neg Comp Dir. I thought that was negative direction but its actually Negate comp direction. This switch is on (1) So im going to let it cool once more, flip that to 0 and try again.

- - - Updated - - -

OK I adjusted that and ran it again for an hour and a half, after it appeared that it was cooled back to original zero. Over an hour and a half it grew still about .0037" Still in the wrong dirrection.

I looked through the paramaters and there is a 268 Neg Comp Dir. I thought that was negative direction but its actually Negate comp direction. This switch is on (1) So im going to let it cool once more, flip that to 0 and try again.




EDIT>.... OMG im gettin annoyed now. Walked out to the control and that Neg Comp Dir switch was already flipped on its own back to 0.

completely give up on this thing. its negated to drilling holes.

[Fletch_CNC]

We just ran a slow program for 2 hours. This program would put very little heat into the ball screws compared to the rapid moves i was doing yesterday. with the spindle running and the axis moving after 2 hours the z was .006" deeper than set. This was just based on touching off a part and viewing the offset amount.

In this scenario, you would think that comp would make the tool shorter, as you have less heat than anticipated but comp still active. that isnt the case however. its almost like its comping the wrong way.

I will check those parameters and compare them to mine. Which machine do you have?

On my machine, setting 160 is a fuse level parameter. setting 274 is t const. and is set to -60000

Im going to try setting my -8000000 to your -225000 and see what happens.

Setting 160, not parameter 160. You can use setting 160 to tweak the comp plus or minus, up to 30%.

The Negate Comp setting I think has to do with what end of the ballscrew the motor is mounted on. I think.

I guess I am unclear on which way your parts are going. If the Z is going too deep over time, I think that would indicate the comp value is too low, as it is not comping enough for the amount the ballscrew is growing. Like I said, I'm not an expert on thermal comp, I'm just trying to brainstorm a little.