Temperature limits for Contura CMM's

[[Ri...]]

I recall being told there are limits to reliability of Contura CMM's exposed to uncontrolled temperature conditions.
What I cant recall is the temperature range.
Any one have information on this?

[[Cl...]]

I've got the spec's in front of me.

Not 100% sure if this is the answer you are looking for but,

Under "Environmental requirements" the measuring reference temperature is 18°C to 22°C (64.4°F to 71.6°F)

[[Da...]]

You may have bigger problems with the parts changing size. How big are the parts you're checking? What material are they made from? How close of a tolerance are you trying to hold? How big of a temperature range are you operating under?

[[Cl...]]

Following up on my last response, the Readiness for operation (Ambient) temperature range is 17 to 35°C (62.6 to 95°F)

[[Ri...]]

Thanks for the replies.
Our parts range from small to barely fit on a Contura (X 700mm x Y 1000mm x Z 600mm).
Currently I am drowning in 82°f at 50% humidity.

[[Da...]]

Zeiss has a temperature probe that can compensate the temp difference back to 68 degrees. You enter the materials coefficient of expansion and it applies the compensation based on the size of the feature. I have it on both Micuras and I won't run parts without it. We have run aluminum parts that get honed. The honing process warms up the part a lot . We were checking parts that were at 90 degrees. The next day after soaking overnight in 68 degrees the parts checked only about .0001" different. We were sizing hot parts that checked.001 over sized with the in-process gauging at the machine. We used to run the parts then let them shrink overnight, then run them again to try to hold the size. The probe and compensation works really well.

The larger the parts are the more they will grow and shrink with temperature variations.

[[De...]]

The CMM has scales that are very thermally stable, so its possible its giving correct measurements even over a wider range of temperatures, however its important to note that though the CMM may not be greatly impacted by the variation, the same cannot be said of the parts themselves.

The movement of the part is determined by the material the part is made from, and the size of the part.

Every material has a CTE (coefficient of thermal expansion) which is expressed as follows.

For our example I will use 300m (which is basically fancy 4340) it has a CTE of 11.34 µm/m-°C

This means the material expands 11.34 microns or .01134mm per meter per degree Celsius over 20 degrees Celsius.
(or alternatively it shrinks for every degree under 20C)

82 Fahrenheit = 27.7778 Celsius
27.7778 - 20 = 7.7778

7.7778 * .01134 = 0.0882mm or 0.00347 inches

So if you measure a part made from 300M which is one meter square and it is 82 degrees Fahrenheit you can expect it to measure 0.00347 inches larger than it will when measured at 68 degrees Fahrenheit

If your machine has temperature sensors for the part, you can activate thermal comp which will do all of this for you and report the corrected measurement based on the CTE that you put in. This means that your 1 meter part will report as correct even though the CMM 'knows' it measured it .00347 inches too big because it will compensate for the higher temperature based off the CTE provided.

If the machine does not have temperature probes and thermal compensation, then your measurements are most likely 'as measured' which means they are not compensated for the higher temperature in the measuring environment and your 1 meter part that measures the correct size now.. will measure .00347 inches smaller once it is cooled to 68 degrees.

On top of all of this is the accuracy of the CTE value itself, all materials are different and in fact the CTE varies at different temperature ranges and to a certain amount in different lots of material.

So some additional uncertainty is added by the variance in the CTE and the greater the variance in temperature from nominal the greater the uncertainty in the measurement even with thermal compensation turned on.

So in answer to your original question, the only real way to know what the temperature needs to be is to determine the amount of acceptable error in your measurements, then using part size and CTE determine how much error is likely being created by the temperature variation.

[[Ow...]]

I wish they had the like button activated because Dereks post on CTE is well written.

Without temperature compensation probing or wires, one thing you have to watch out for is the amount of temperature change the CMM is absorbing through out the day.
I MAY BE WRONG (please correct if so) but the way I interpret this is:
While it can be run in up to 26°C (78.8F) temperature, what ever the temperature is after calibrating the probe can not change more than 1.5K (kelvin) per hour, which is like 2 something degree's Fahrenheit.