Thermal Expansion!!!!!

[Zbuilder]

We had a thermal expansion of .002" also on our Leadwells vertical CNC. after all kinds of money spent with techs coming to check it, we found the solution ourselves. The machine was in front of a large window. During sunning mornings, the machine would heat up. Afternoon, nights & cloudy days, there was no problem.

[underthetire]

Originally Posted by jetski This is a common error among encoder machines. If it is a big problem you needed to order the glass scales. Even with glass scales on a 180,000.00 machine at a power drop I found I was more secure doing a re indication of zero. Just because I am a anal retentive tool maker. Just so nobody could say you dumb-#$%ss you didn't reindicate it after the power went out. Kinda like the guy who goes to the manual mill with a vise clamped down and assumes it to be square. I always run an indicator across a vise and always rezero after a power out just good tool makin. Saves time and embarasment of trying to track down an inocent rookie. He was probably only trying to get the top and bottom flat...no need to indicate.

Not all encoder machines have this problem. We have Hurcos and some top quality machines as well. All of them have spindle coolers on them, while the Hurco's we have are no where near as good as the others, we don't ever see that kind of growth. I have seen people mess with the coolers in the past, thinking colder must be better. I could put an indicator on the spindle, power up the cooler and watch the head move from being too cold. Sounds to me like it was a pre-sales issue, or I don't want to pay for that issue. Coolers should never be an "option" on any machine above 6K, especially a belt drive head, should just be a standard. I don't know if Hurco stretches the ballscrews, or even is a double support on that model, but all the builders I worked for would stretch the screws to compensate for growth.

[bloke]

I'll be sure to give 2007 a call and inform them of your findings....

[underthetire]

Ha, your right. Didn't see someone dreged this up from the past. At least I sorta stuck up for Hurco on this one !

[Zbuilder]

Not sure if ball screws are compressed or stretched. hmmm. anyone?

[underthetire]

Again-2007. Ballscrews are never compressed BTW.

[bloke]

Stretched - so thermal expansion relaxes the tension rather than changing the length.

[Rich Carlstedt]

To get .002" growth, you need a 20 inch arm and raise its "Total" temp 20 deg F.
( Actually 400 units ie 1 inch times 400 degrees, 4 inches times 100 degrees or 25 times 16 ect)

I therefore have to assume the spindle head housing is warming up that much to push the spindle away from the column ?
OR
it could be that the table and knee/cross-slide are being cooled by coolant sufficiently to produce that result.
It could be a combination of both.
Do you have refrigerated coolant ?
I would compare the two of them for differential in temps both before and after warm up.
knowing the thermal conductivity of Cast Iron and the twenty degree difference would be incredible

The comment about the belt was excellent. Some old KM3's had to have a fan placed on the head drives to prevent the plastic bushings from premature failure.

Now, after writting the above, I went back to your original posting and reread it.
I now have to add this. I think your assumption about the spindle and head is wrong
Thermal growth of the head does not move X.
The only way it could, and still be in the head is if you have a turret head ?
and only one cap screw of the spider was locked down.

How are you measuring the drift ? what procedures ?

[bloke]

The original post was from 2007. I don't think he can hear you from that far away...

[Rich Carlstedt]

Thanks Bloke !
I missed that..

We had a 5 inch 5 axis CNC Boring Mill (85,000 # worth)
at work in a non temperature regulated work envirionment and
yet worked to very close tolerences.(.001)
There is so much to thermal expansion as you know so I
thought I would join....looks like I missed the boat...by a long shot !
Years ago there was a fellow who kept loosing Y location and it turned out
to be loose Ram Clamps, and not a servo drive problem
Back to basics..

[pgf545]

I know this is a 07 post but I just had the same issue with one of our VM2's last week. I have been running a VMX42 for several years but I need to do another job so the boss said to run it on one of the VM2's. I started running it at 8000 and noticed the tool depth changing. I rechecked tool and noticed it was about .001 to .0015 deeper. My thinking was that there was a little expansion in the spindle itself and maybe some of the heat transfered to the tool holder and that was my total expansion??? I set the tool a little higher and reran the program again.