Why a quench during thermal cycle vs air cool?

B

BFrancios

Joined
Jul 9, 2012
Messages
65
I have seen recommendations to include a quench during thermal cycles prior to the hardening step. I even saw an explanation for why once.....
Now that I have progressed enough to be dangerous, with a working understanding of the heat treat, I have been searching for that thread and can't find it.

Would like to work this quench into the heat treat for my current project but want to know the reasoning behind it.

I AM NOT TALKING ABOUT TRIPLE QUENCH during the hardening but about including a quench during thermal cycling.

For example, here is one recommended thermal cycle I found during my recent search;

Simple Steel, 10xx or W2
Heat to 1650 - cool to black
heat to 1550 - cool to black
heat to 1450 - quench
Then anneal and proceed with hardening.

I have also seen recommendation to quench at each cycle instead of letting the blade air cool.
Does anyone have the rational behind this quench. What are we trying to accomplish within the structure with a quench vs air cool?


Barry
 
The purpose is grain refinement. Martensite from the quench provides good nucleation sites for the formation of multiple austenite crystals. Each time you quench you convert the (now smaller) austenite to the (now smaller) martensite crystals. I've seen some amazing grain refinement pictures posted for No quench vs multiple quench. A finer grain structure gives increased toughness at a given hardness.
 
I'll be watching this one, great question.
The way I THINK it works, is that the higher temp dissolves and distributes the carbides, and fairly quick cooling (I often see the opposite of what you describe- quench from 1650 and 1550, cool from 1450) stabilizes the structure.
So often my mental picture of what's happening needs straightening out...
 
Kevin Cashen is one good source to talk to about this. He has gone back and forth, IIRC, from just air cool during normalizing cycles vs quenching during normalizing cycles. I think at this moment, he is leaning towards air cool, don't quote me though. I don't quench myself, strictly air cool. If there were hard evidence that grain is THAT MUCH finer with a quench, I will change my methods for sure. Maybe just use the canola for normalizing quenches and save the commercial stuff for austenitizing. Great question indeed! I hope this produces some great info.
 
what you want to do is to freeze your structure without pearlite precipitation and not give anything that has gone into solution time to aggregate into large particles

-Page
 
Stuart(samuria),

Kevin's site was actually the last stop in my search, and his silence on the subject in his heat treat pages led me to make this post. I am fairly certain he contributed to the thread I remember on this subject but can not locate. Would love to find his take on the subject.

Looks like the quench, if used, should be done at the higher temp cycles.

Barry
 
A lot depends on the steel. Take a number of test pieces, clean them up, get rid of an areas for stress risers, and run them through the different grain refinement steps. After you quench for the last time, break them and see if you end up with a finer grain and if you do - how much finer. Then evaluate the cost/benefit of the additional steps.
 
Stacy might be off duty for the weekend- he has described thermal cycling where you quench from the 2 higher temps and cool from the lower one, several times, but I don't recall a stated reason for the quench vs cool- I probably just forgot what he said about it...
 
OP, Are you forging or are you doing stock removal? If you get speroidized steel for stock removal you will not have to anneal the steel. You will go through the grain refinement steps just before the final quench. Also, you just need to reach austenizing temperatures. Getting the steel too hot contributes to grain growth, defeating the final goal. If you are forging, you will anneal the steel just before the final shaping. There are two pathways I can think of; normalize-->anneal-->shape-->austenize-->final quench, or anneal-->shape-->normalize-->austenize-->final quench. With the normalizing step being the grain refinement/homogenization steps. There are other pathways, but these are the two that I have memorized.
 
Barry, I sure hope there isn't a rule about mentioning another forum on this forum, but go to hypefreeblades.com and search within that forum for threads Kevin has reponded to concerning normalizing/grain refining/quenching. Mods, if mentioning another forum on this site is verboten, please delete.
 
Last edited:
You're freezing a structure so there will be more grains thus smaller ones thus more nuclei for subsequent grains.
 
Samuraistuart - thnx!

That is not a site I had yet located. While is have not yet found exactly what I was looking for in my original question it is probably answered somewhere in that site. Looks like the time searching the site will be well spent. I have always found Kevin's Posts here on bladeforums to be invaluable.

Bo T- yes I am forging.

I will post back with what I find, if others have info on a quench during the normalize/thermal cycle process, please chime in.

elementFE - Stacy's process IS one I was able to find, if I remember he quenches at each thermal cycle. What he did not say in my reading is WHY and what is happening to make a quench preferable over air cooling.


Barry
 
Thanks, METE.
Once again, it's great to have knowledgeable people "on call' for questions like this- you're greatly appreciated.
 
I think the biggest concern with quenching during normalizing/cycling, especially at the higher temps, is that it can cause microscopic stress cracks where the plate martensite interact, whereas the air cool negates that problem. The air cool being just fast enough to do the job without the worry.
 
Mete- Thanks for the reply, I appreciate your input. We posted at about the same time and I missed your reply.

Here is what I have found on this subject.

A quote of master yoda from a galaxy far far away;
"Finer grains result from austenitizing martensite because there is vastly more strain energy to drive the rate of nucleation, thus there will be more points of nucleation and more grains, more grains in the same area means they must be smaller. However as you touched upon one can also increase nucleation through duplexing the phases with tight heat control, or limiting grain growth and staying ahead of the decreasing grain coarsening temperature- once again with tight heat control. But some of the bladesmithing crowd, you gotto love em, attempt heat treating with forges or torches, so reheating martensite is probably their best bet."
by Kevin R. Cashen » Tue May 21, 2013 10:41 am. Taken from a thread on heat treating W1, with a torch!

To paraphrase, releasing the energy from those tightly pulled cubes of martensite can result in smaller grain size and hopefully smaller carbides in the final hardening. But the same result can be reached,with less shock to the steel, using good heat control and thermo cycling.

It so happens that this hobby blade mangler fits into the group that Kevin mentions at the end of his statement!
My home built, PID controlled forge holds temp +-10 degrees in that 2 cm spot where the thermo is located. Temps outside that tiny spot could be hundreds of degrees higher.

Since I could be completely misquoting Mr Cashen, I am going to take 2 small pieces I had to cut off my current project. No comments please on what that says about my forging skills.:) Both were forged and normalized to the same process at the same time and are from the same bar of W2. I will 1450 quench and lightly anneal/temper one and then harden both and snap them to compare grain size.

If I can detect a difference in grain size, I'll use the process with the smaller grain. If a can't detect a difference, hobby maker here, I'll probably add the quench step to future normalizing/thermo cycles anyway.

Barry
 
You must log in or register to reply here.
Back
Top