Co-efficient of thermal expansion and NIST 1

[aftconsultaps]

Hi,
I'm working on some tolerances after cool down, and have up until now been using CTE's; typically 17,3um/mK for 304SS, 16um/mK for 316SS, and 16,7um/mK fro BeCu

I've been refining my design, and have been double checking with the NIST cryo materials page, where I have worked out expected shrinkage, based on their linear expansion equations.

Firstly, on NIST there is the same equation given for 304 & 316, which doesn't fit the with the CTE's.
Secondly, when I evaluate the expressions then I get a higher shrinkage for BeCu than for 304 (&316) at the temp (100K). If I use the CTE for 304 vs BeCU, then I'd get a lower shrinkage for BeCu than for 304.

Can anyone shed any light on this? It's rather important due to the tight tolerances I'm working with.

Thanks in advance,
Chris

 

[EdStainless]

First, you need to pay close attention to the CTE data, is it at a temperature or is it for a specified temperature range. The CTEs are very non linear.
Secondly you have to cycle the material through the full range a few times before you can expect to get good repeatability. Residual stresses (and their non-uniformity) are a big deal.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[aftconsultaps]

Thanks Ed; do you know of any good (reliable & accurate) database where to get the CTE data from?

 

[EdStainless]

The problem that I have found is that the data depends on the method used for measurement.
We would heat samples to about 600F (depending on the alloy), hold that temp and cool very slowly.
Then cycle the samples across the temperature range of interest a few times before we started measuring.
Even doing this we would have samples of 304L that would be different by 20% or more.....

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[iainuts]

Interesting info from EdStainless. Thanks for that.

I've always used the Cryogenic Materials Data Handbook:

http://www.dtic.mil/dtic/tr/fulltext/u2/438718.pdf

See page D.11.p for 304

 

[aftconsultaps]

Thanks Ed & Iain for your comments; I also found this

http://www.eckelsengineering.com/

, although I haven't paid for it yet, so anyone that's used it would be good to hear from.

I guess it is a computerised version of all the data mentioned in your link (which I have also used by the way!) , plus more.

 

[iainuts]

That's interesting... I know Phil Eckels. We have a patent together from January 2000. I enjoyed working with him.

I haven't purchased his computerized database. Instead, I have spreadsheets to do the same thing.

 

[EdStainless]

IF you are really concerned about tolerances at low temps you need to be using material with low CTE.
I know that there are not many options, butt hat would greatly improve your results.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[aftconsultaps]

Good piece of advice, well noted!

 

[georgeverghese]

Invar is a popular material of construction for the inner membrane of large LNG storage tanks operating at approx 100degK.

 

[EdStainless]

The only problem with all of the low CTE material is that they will rust because they do not contain any Cr. It is common to use a Cu flash platting (very thin) to protect such parts.
Invar (Fe 36%Ni) is the only metallurgical discovery to ever be awarded a Nobel Prize.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube