View Full Version : Liquid Nitrogen - really broken

08-14-2004, 01:09 PM
OK I've done 7 more blades using my test results. 3 S30V and 4 CM154. Still no response from the LN! :eek: :eek: :eek:

The latest batch were CM154,

soak at 2000 for 20 minutes, 61, 61, 61, 60
temper 225 for 1 hour, 62, 62, 62, 61 (no quess, why 62)
LN soak for 23 hours, 61, 61, 61, 61

The S30V blades didn't have any change either. And yes Roger some of the tests had 45 minute soak times.

Some one have an explaination?


08-14-2004, 01:26 PM
I am only guessing but it may be that you are decomposing most or all of the retained austenite to pearlite/bainite with your tempering at 225 for 1 hour. There may just be no RA left to transform by the liquid nitrogen. I am a believer in a short, low temperature draw prior to cryo treating but maybe you need to reduce the time to 15-20 minutes or less? :confused:

08-14-2004, 02:09 PM
Beats me Steve. I think the first thing you'll need to do is get a good handle on the oven temperature. That is the only variable in the process. It started going squirley after repairing the oven door. I'm sure you used the correct brick on the door and that you layed them with a bed of refactory cement and that you patched all cracks or chips with refactory. I am also sure you checked the oven at about 500 F using a oven thermometer.

For sure, I believe 45 minute soaks are too long and that 20 minutes may be too short. I would think that tempering at 225 F would only hold the steel and not temper at all. You are not allowing the steel to achieve room temperature before snap temper(?). Whether you snap or not I would expect an increase in Rockwell after deep cryo and that a absolute max of 10 to 12 hours should transform all austenite it can.


08-14-2004, 02:49 PM

Really I have tested all sorts of oven temps (from 1950 to 2000) and soak times and steels. Plus, quenchcrack, I did some tests with and without snap tempers including short tempers. I did get various as quenched readings and different responses to tempering. ( from 60 to 63)

3 differents steels (ATS34, CM154, and S30V) never did the LN change the hardness. I'm confident of my readings because I had multiple blades in one batch and they are all pretty much going thru the same hardness changes.

I've had 2 samples in one pack and one sample per pack. They all come out the same.

The only thing I can think of is my quenching process is not leaving retained austentite. I find that hard to believe.

You know Roger the oven might be off, but I'm getting as quenched readings consistent with the Crucible sheets - if you subtract 30 degrees from my oven setting.


08-14-2004, 03:09 PM
You are either perfect or something is off. Score and snap one of your tempered plate quench samples to examine the grain (place a rag over the steel before striking. I have had fragments stick in the wall 15 feet away). Try one more sample by rapid air quenching (no plates). See if the cryo shows a difference on it. Score and snap it. Compare the two samples for grain fineness.

Soak the air quench sample for 25 minutes at about 1960 F. When it becomes hand warm but no less than hand warm place in preheated 300 F. snap temper / 1 hour.


08-14-2004, 05:32 PM
"Score and snap one of your tempered plate quench samples to examine the grain"

OK, looks excellent (I compare to a snapped nickelson file looks same). I mean really good. Better than my carbon steel tests that's for sure.

"Soak the air quench sample for 25 minutes at about 1960 F. When it becomes hand warm but no less than hand warm place in preheated 300 F. snap temper / 1 hour."

Been there, done that.

I would think that the temp might be off by a full 100 degrees, except that the S30V resonds to tempering just like the data sheets - without the jump in hardness from the LN stage.

Go figure.

I'm gonna try and find some firing cones and test the temp.

Open to all explainations......


RJ Martin
08-14-2004, 07:54 PM
Steve: I'm not all that surprised by your results, and, it sounds like you've got good results. Just out of curiosity, what kind of hardness tester do you have, and, what is the condition of the surface of the blades when you test them? How many tests are you doing on each piece?

Remember, you're converting the majority of the structure to Martensite on the initial quench, and, if that's where your diamoid point hits, you're going to gat a hardness that reflects that reading. And, quench plates cool the steel at a tremendous rate-much faster than required for conversion.

30 minutes at temp is a fine soak time for knife blades, 5 minutes either way isn't going to have a big effect, either, unless your blades are really big.

I frequently don't see a noticeable hardness change with the LN2. But, that LN soak is still improving your structure, reducing grain size and helping with toughness (once the steel is tempered, that is). BTW, the snap tempering recommended by Crucible is largely to prevent shattering when the blades go into the LN2. the 400F recommended for S30V takes enough stress out of the Martensite so that the tendency to crack is minimized. This is straight from Scott Devana of Crucible.

08-14-2004, 08:31 PM
Thanks RJ, I guess I won't worry.

Yeh the plates are fast. It doesn't take long at all to go from the oven, betweent the plates and blast with air. Definately under the 'nose' for these steels.

Within 30 seconds (never counted, maybe 15) they are cool enough to cut open, pull the blades and check for straight. They are just too hot for the gloves at that point. It gives me plenty of time before they are too hard to 'adjust'. All in all I'm very happy with the quench process and really don't want to change it.



08-14-2004, 09:34 PM
Steve, that is exactly why I was hoping you would compare the grain structure of rapid air quench vs. plate quench; not to compare it with a broken Nickolson file but to compare the samples alike except for type of quench.

I thought grain structure was developed during soak and that cryogenic treatment ONLY transforms retained austenite to UN-tempered martensite.