Refining grain structure through tempering...

Rick Baum · Feb 15, 2003

I've been doing some reading about cryogenics and there is a chapter in the book that talks about heat treatment. In this chapter it talks about double and triple tempering steel. It says.. "It is often advantageous to double or even triple temper to increase wear by refining the grain structure." It goes on to state.. "lower the tempering furnace temperature 25 to 35 degrees F from the original temperature and proceed with the second temper. Tempering again at the same temperature as the first has little effect on the grain structure refinement. The 25 degrees F difference causes the optimum change and assures the best rockwell hardess level is maintained."The author goes on to say that "the higher the alloy content, the better off you are."

It makes sense to lower the temperature to preserve your RC hardness, but to refine the grain???

Does anyone out there lower their tempering temperature for the second and third tempering cycles? If so, have you done any comparisons with knives that did not have the temperature lowered to see if the grain structure was in fact finer?

Rex, Are you out there? Can you shed any light on the subject?

Rick

Epsilon · Feb 16, 2003

I'd have to read the section you're describing to get a better idea of the surrounding subject matter. However, interpreting what you have quoted from that book both does and does not make any sense.

In a carbon steel, say 1084, this just doesn't sound right. Carbides in fact will coarsen and grow larger as time and temperature increase. The end extreme being the large, spheroidal carbides in spheroidize annealing.

In a higher alloy, martensitic stainless steel like ATS-34 the statement may be true. If considering cryo, the change from retained austenite to martensite for example would result in an overall smaller percentage of fine grain. This is because austenite grains are much larger than martensite grains. Thus %-wise you have a finer grain. As far as "tempering to refine the grain", perhaps they are referring to the formation of alloy carbides in secondary hardening. This may be true if considering the carbide sizes. Mo, and V for example are smaller that cementite (iron-carbide).

Some ideas to ponder.

Jason

tmickley · Feb 16, 2003

Rick, what book is it you are reading this from?

Rick Baum · Feb 16, 2003

tmickley, The book is "Cryogenics" by William E. Bryson. It's not exactly a technical manual, but it does explain things so that even I can understand.

In a carbon steel, say 1084, this just doesn't sound right. Carbides in fact will coarsen and grow larger as time and temperature increase. The end extreme being the large, spheroidal carbides in spheroidizing and process annealing.

Epsilon, The author does mention that "The higher the alloy content, the better off you are." I'm obviously no expert on the matter, but isn't 1084 a relatively simple steel? (I honestly don't know which steels are considered high alloy and low alloy) Would 1084 be considered a low alloy steel and therefore wouldn't apply to the benefits that the author is talking about? Wouldn't the grain growth that you are talking about be at the higher temperatures required for annealing and quenching not the relatively low temps required for tempering? I hope I don't sound like I'm being a smart A$$, I'm just tring to understand. I really do appreciate your input. I think you hit it on the head as far as grain size being refined due to the transformation of austinite to martensite. What is secondary hardening? Is that what the author is refering to when he talks about fresh martensite being converted from austinite during the tempering cycle?

Thanks,
Rick

Epsilon · Feb 16, 2003

No, you're not being an ass.

You're asking really good questions. I was using 1084 to kind of explain what kind of steel would NOT see this effect. ATS-34 would qualify as "high alloy" basically. 1084 is a carbon steel and is the most simple steel type (alloy amount) as far as industry standards are concerned. It would certainly fall into "low alloy" class. High and low alloy is just a generic term. Its not specific. Basically when we talk knife-worthy steels, low alloy would be 10xx, 5160, L6. High alloy would be stainless steels, D2, M2. You get the idea by studying their chemical compositions.

Wouldn't the grain growth that you are talking about be at the higher temperatures required for annealing and quenching not the relatively low temps required for tempering?

I use the idea of "range" in tempering. "Annealing" in the sense you are describing has to breach critical first and then slow cool. Spheroidizing and process annealing occurs below Ac1 or critical temperature. Technically this is within tempering range. Although, to be practical, most steels are tempered within 300-1100°F. Most spheroidizing and process annealing occurs around 1100-1250°F, just below Ac1 for many steels. The steel grain doesn't yet recrystallize. Breaching critical changes everything. The relatively low tempering temps you are speaking of will show change but, subtle change. Oh and just FYI, spheroidizing gives the MOST machinable condition.

What is secondary hardening? Is that what the author is refering to when he talks about fresh martensite being converted from austinite during the tempering cycle?

No. That is simply converting 'retained' austenite to martensite. But did he say this occured due to tempering? Was it BECAUSE of the tempering, the conversion to martensite took place, he said?

Secondary hardening is when carbon precipitates out of solid solution (usually after enough holding time/temp) and it begins to form carbide with the alloys in the steel. For example, on a graph, it might show a "dip" or drop in hardness, as time/temp changes (carbon precipitating) and then rising again as the alloy carbide forms. In other words iron carbide sorta goes bye-bye and some alloys get their turn.

Jason

Rick Baum · Feb 16, 2003

Epsilon,

But did he say this occured due to tempering? Was it BECAUSE of the tempering, the conversion to martensite took place, he said?

The answer is Yes. The author is saying that during the initial and all susiquent tempering cycles, there is a transformation of retained austinite to "Fresh martinsite" that occurs in diminishing quantities with each tempering. He goes on to talk about secondary hardening... "What happens in a steel that has secondary hardening is this: the initial quench is enough to create a high percentage of transformation. But because these steels are so highly alloyed, there is a reluctance for all the carbides to precipitate and complete transforms. The second temper raises the temperature well above the matrensitic grain formation point, and, when cooled, causes this secondary transformation process to continue."

I guess my question about all of this is geared more towards the type of steel that I am using and others that I want to use in the future. Those being 5160 now and 52100 later. I know that Ed Fowler has done a lot of personal testing and has come up with forging and heat treating formulas that are creating extreemly fine grain structures, this has been verified by Rex Walter. I guess I'm just trying to understand why the triple steps of heat treating work so well for him. Is 52100 steel considered to be a high alloy steel? I guess the steel doesn't need to technically be "high alloy" the author is stating: "the higher the better."

Thanks for your help on this one Epsilon. By the way, you seem to be quite knowledgable on this stuff. Do you have a background in metalurgy?

Rick

Epsilon · Feb 16, 2003

I am seeing what the author is talking about now, but perhaps I feel it is being somewhat generalized. Not uncommon. For a book of this nature, that is the audience it is reaching. And it makes sense. This makes things a bit grey though and seemingly contradictory if you take it literally without no further investigation. Especially, since you yourself are interested in the finer mechanics of ferrous metallurgy. Which I am greatly encouraged to hear. Its a fun and fascinating field to pry into. It would take up so much time and space to explain the subject you are addressing though and no doubt to the boredom of many here. But, if you want more detailed answers just keeping asking or you can even e-mail me. Will be a lengthy, on-going session though.

Best advice I can give on his book is to follow the general practice itself, but make the quotations simply a ground rule. Metallurgy can be broken down into such tiny pieces of study. Often, I found myself not answering some metallurgy questions because a "basic" explaination would be incorrect. Does that make any sense? Does to me. Hmm.

My metallurgy background is simply books and blacksmithing. No schooling.

I guess my question about all of this is geared more towards the type of steel that I am using and others that I want to use in the future. Those being 5160 now and 52100 later. I know that Ed Fowler has done a lot of personal testing and has come up with forging and heat treating formulas that are creating extreemly fine grain structures, this has been verified by Rex Walter. I guess I'm just trying to understand why the triple steps of heat treating work so well for him. Is 52100 steel considered to be a high alloy steel? I guess the steel doesn't need to technically be "high alloy" the author is stating: "the higher the better."

No. 52100 is not classified as a "high alloy steel". To be more specific, it is an "alloy steel" with a high-carbon content. Same with 5160, though it is a hypoeutectoid steel and 52100 is a hypereutectoid steel.

I guess the steel doesn't need to technically be "high alloy" the author is stating: "the higher the better."

Again, don't take it literally, just apply it to the subject at hand.

Jason

Rick Baum · Feb 16, 2003

Thanks Jason. I think I will try to keep from boring everyone for a while. It sounds like I need to do some more investigating and keeping things general will probably suffice me for a while. My biggest problem is time. I never seem to have enough of it to do the testing that I want. So far it's taken me more than a month to get my fist forged knife ready to quench.:grumpy: The problem is that the more I learn, the more questions I seem to have. Hopefully they will become more intelligent as time passes.

Thanks for all of your help,
Rick

rlinger · Feb 17, 2003

It should also be noted, as explained by the same author, that tempering times are or should be stated as: minimum tempering times, as tempering time can not go too long. Example: temper at 'such a temperature' for 30 minutes can be elongated at the operators option to 'overnight'. Proceeding tempers and preceeding snap tempers may differ in temperature but only specified temper time is important as a minimum time to temper and going over that time (at the specified temp.) is not harmful. Doubling temper time does not count as two separate temper sessions.

Now, to stray just slightly backwards from your questions and observations - I am, I think, learning that double and triple quenches (austinitizing/hardenening sessions) is very good in developing good grain structure also. And that cryo after snap temper and multiple tempers should follow as well.

Roger

Epsilon · Feb 17, 2003

Baumr,
Your questions were intelligent! I hope you hadn't taken my responses the wrong way. I know exactly how you feel. Its frustrating. I wanted to know 'why' all the time in the beginning. Still do! Sometimes more so than actually completing a knife. It was a great learning experience though. Still is. We all keep learning. I will be on this path forever. The more you learn, the further away you get from understanding.

Keep asking questions. Always. I encourage it! Work at your own pace. If it takes you another year to finish that knife than so be it. In the end, it was supposed to be that way. IMHO.

Best of luck, always happy to help out.

Jason

Rick Baum · Feb 18, 2003

Jason, Don't worry, I didn't take your answer the wrong way. I think I didn't do a very good job of expressing myself in my sleep deprived state yesterday.:yawn:

Roger made me think of another question. Do you guys know why the author (and others) recommends to temper the steel within an hour of quenching or you would be better off to anneal the steel and start over with the quenching? (Not an exact qoute) Again, this is one of those statements that causes a need for further ivestigation. I can't figure out why this would be necessary and the author doesn't say why either. I'm beginning to understand that this book is very superficial. But hey, you gotta start somewhere, right? If nothing else, the book is fueling my curiosity.

Thanks for all of your help guys!

Rick

rlinger · Feb 18, 2003

Baumr,

It is important to temper or snap temper before the steel falls below 125 F.. Not doing so will spell short life to your steel (especially, as I think I understand, tool steels). Grain structure can enlarge as the steel falls below this area of temperature before tempering. I try to catch my steel between 140 and not less than 125 F. before placing in the pre-heated temper oven. And again, I am only concerned that the temper time is for at least the minimum specified and do not worry a whit if it stays in the oven longer - so long as the specified temper temperature is correct.

Roger

Rick Baum · Feb 18, 2003

Roger, Thanks for the advice. This brings up another question though. If I am doing 3 quenches and they are a day or more apart, and I am putting the blade in the freezer between quench cycles, wouldn't the steel have cracked by now if it was going to? Do you think it's because I'm using 5160 that I haven't had any problems? I'm getting a good hard edge after each quench so far and the spine is nice and soft. Maybe the differential heat treat is helping to prevent problems.

Rick

Epsilon · Feb 18, 2003

Baumr,
The reason to temper "immediately" is to avoid the risk of cracking. An hour is simply a rule of thumb. It just means ASAP. Some steels will bust in your hands seconds after a quench, some the next day. Dimensions and cooling rates on those dimensions play a VERY large role. Cracking is MOST susceptable in parts cooling at different rates from surrounding areas. This is one big reason why air cooling tool steels are so desireable in complex parts for machinists. Air cooling is less severe and tool steels are deeper hardening, meaning a better chance at uniform grain sizes. Thick or thin. Comes down to that old adage, 'use the right tool for the job'.
Fun stuff.

rlinger · Feb 19, 2003

I know you're not doing a cryo by putting the blade in the freezer but ..... If doing multiple quenches; and I mean by that you are heat treating and quenching multiple times, it would be a waste of time to cryo between heat treating sessions because the grain goes through dynamic changes during heat that negates any benefit from tempering or cryo. Cryo should be done upon cooling to room temperature after the immediate snap temper or immediate full temper following the final quench.

So long as the steel is not worked and handled carefully between quenches, waiting a day or two between quenchings should be okay in my limited experience but a snap temper would be good insurance. I would hope though that someone here with more experience would correct me if I am wrong about that.

Epsilon nailed it on the fracturing. And it is important for that reason to quench all sufaces of the steel equally and as close to the same instant as possible.

I will repeat my conviction that it is important to temper or at the least snap temper before the steel reaches ambient room temperature. When it says "temper immediately upon cooling to such and such a temperature" it is because grain structure will change for the worse if not.

Roger