Achieving Low Temperature for Charpy Impacts 1

[Lanti]

I am trying to determine a way to achieve a testing temperature of -238F specifically for low temperature Charpy impact testing, on a weld procedure. This is a design temperature per a customer requirement.

Right now, my last option was just going with liquid nitrogen and testing at -320F but the weld metal was too brittle at the temperature and did not meet the impact requirements.

I am looking for any ideas of getting a sample bath closer to that -238F temperature, but not as cold as pure liquid nitrogen. Help?

Thanks!

 

[metengr]

Yes. Look at one of these or similar;
WDC-196T Charpy Impact Test Cooling Chamber

Otherwise, have a competent metallurgical lab perform the testing and certify the test results.

 

[metalman8357]

Get your hands on some liquid Krypton...

 

[Lanti]

Thanks for the help!

 

[fmanzanilla]

This might be a silly idea, but would it be possible just to cool the sample using N2 and wait for it to absorb heat until it reaches specified test temperature? Keep track of T with a pyrometer or some other way.

I mean, the material would mostly recover it mechanical properties after it warms up, right?

 

[metengr]

fmanzanilla;
Yes, the impact properties will recover because of temperature. However, you are measuring the surface temperature and the sample needs to be tested at -238 deg F. Assuming the sample surface temperature reaches -238 deg F, the core temperature can be below this. Any competent materials/metallurgical testing lab does this testing routinely and has the equipment.

 

[redpicker]

To add to what metengr said, ASTM E23 has some specific requirements for acceptable cooling methods for low temperature Charpy testing, with specific holding times and cooling media. Without a properly designed chamber, such as mentioned above, it will be difficult to achieve the requirements of E23.

From a pratical standpoint, however, I don't think it is going to help. That is, if you don't meet the requirements at -320F, you aren't likely to meet them at -238F. Assuming that the material exhibits a ductile/brittle transition with respect to temperature, either temperature is going to be on the lower shelf of the transition curve so the effect of temperature isn't going to be all that great; you will likely see more variation from the statistical nature of the results than you will see from the temperature difference. If the material does not exhibit transition behavior, even less. I guess, depending on what your material actually is and what results you are trying to achieve, it is possible that you might gain enough to meet the requirements, but I would doubt it.

rp

 

[redpicker]

Just to add, I found that R-503 refrigerant (Chlorotrifluoromethane/Trifluoromethane) boils at -128F, so if you get some of that and cool it with your LN2, you might be able to make it work. I have no idea about the flamability or toxicity of the stuff, and I can't believe it would be worth dealing with (taking you samples to a lab that had the necessary equipment would be the easiest thing to do), but it would be a possibility.

rp

 

[Lanti]

Once again, thanks for the additional information. I appreciate the help!

 

[Mett87]

Isopentane (C5H12) and Liquid Nitrogen slush can be taken down to -256F (-160C) if thats any help?