Jump to content
Sign in to follow this  
DFogg

how about 52100?

Recommended Posts

The L6 thread has been really interesting. Could anyone offer some similar advice on heat treating 52100? I've got a coal/charcoal forge, a propane forge, and a kitchen oven to work with, so it's not the highest of technology. I'm hoping it will be enough though.

Share this post


Link to post
Share on other sites

I won't claim to be the expert, but I use 52100 a lot and have figured out a couple of things (if I'm wrong please correct!!). Forge at lower and lower temps as the blade is formed. This helps to keep grain growth down. Thermal cycle three times at a minimum after the blade is formed to prepare for heat treating.

To heat treat, bring the blade to 1550 degrees and oil quench. If you have the grain fine enough you can get away with an interupted water quench, but you'll crack a few blades until you get the feel for when to interrupt. If the blade cracks instantly, the grain was too large or the water too cold.

Tempering: 400 degrees, half an hour, twice=60 HRC

500 degrees, half hour, twice=58 HRC and a really tough blade.

 

I usally edge-quench my blades. If you polish and etch them you can see a definate temper line. Nothing like the hamon's I've seen on this forum, but there none the less.

Share this post


Link to post
Share on other sites
Thanks Tom! 52100 is tricky stuff, and as I found out, it burns at just orange heat. I then made one piece that turned out ok but the second piece cracked and the tip fell off, hence my questions. The tip fell off after tempering, so I'm betting it may actually have cracked in the quench (motor oil, yuck, I'll try veggie oil next time). How hot are you going up to on the thermal cycle part? And is your quench water bath temp, boiling, or somewhere in between? I have been heating quench oil to a couple hundred degrees before quenching, but will happily try something different if anyone can suggest a more suitable procedure.

Share this post


Link to post
Share on other sites

While I'm forging, I quench the blade in oil every couple of heats to keep from overheating it. The actual thermal cycling process I use is to take the blade up to about 1600 degrees the first time, quench, then up to 1500 and quench. I do the 1500 quench at least twice. For the actual hardening phase, I heat the oil to around 150 degrees. I have a five gallon bucket of mineral oil with a shelf in it to set the quench depth. I quench the edge and rock it back and forth until all the color has left the back, then quench the whole blade. You should temper immediately after quenching. I had blades crack on the way to the kitchen, so I set up a deep frier full of oil to do the first tempering heat.

As far as water goes, figuring out when to interrupt the quench seems to be more important than the water temp, at least in my experience. I probably wouldn't try ice water, though.

The best thing you can do is test, test, test!!! Get some 52100 bearing races and break one before heating so you can see the grain. Keep that piece as a reference. Then, try any heat treating method you think might work, break the blade and compare the grain. You have to have that fine grain structure or you'll have nothing but problems.

Again, I'm not the expert, just my experience.

Don, Howard, Mike, Larry, Mr. Graham, someone jump in here!!! :)

Share this post


Link to post
Share on other sites

I discovered 52100 to be "problematic" for me... meaning it didn't like me much...

it seems to be picky in forging temps, to hot it has the whole crumbly thing going on, and it don't like to be forged at low temps at all... so it's a tough one for me, I like to forge on the cooler end of things.

 

What worked more or less for me was to keep the forging time short, to keep the blade at temp, I tried to keep it just up over aus temps, like around 1500, 1550, a pretty narrow range.

I normallized by going just barely non-mag, after a soak, say about 1450, let it cool to a full "bloom", the visible re-calecence, and I quenched it just as it was dropping temps again, to try and make fine-grained peralite, and this worked ok.

 

Generally though, it just didn't suit my methods well... I ended up going to 1086M as my main high-performance cutting steel, and still use W-2 if I can find it.

 

and quit callin me Mr....

:angry:

Share this post


Link to post
Share on other sites

52100 is a peculiar beast, and I do not reccomend it to people as a rule. I know it is several well known individual's "pet steel", but remain of the opinion that many who use it do not understand it well at all, and will go to my grave believing that it is Al Pendray's private joke on the rest of the bladesmithing world (at least the majority who don't understand it).

 

The grain grows rapidly at high temperatures. It is forgable at high temperatures in spite of this. I have made pattern welded steel out of it with nearly everything else in the shop, just to prove it could be done. This includes 52100 and L-6, 52100/1095, 52100/1086, 52100/4340, 52100/1045, and more. All using quite high welding temperatures.

 

It is good quality steel, and it can make excellent blades. IT can also make crap blades if not done well, they can end up brittle at any hardness level.

 

If you don't have temperature controls, and do not understand the metallurgy of the way it works, my advice is to not mess with it at all. In spite of whatever Ed may say. I think he is confused about many things.

Share this post


Link to post
Share on other sites
A shot across the bow  [applause]

Share this post


Link to post
Share on other sites

I don't know the metallurgy yet but I'm fairly confident that I can learn it with some help (ok, maybe a lot of help :P) from everyone here. Would anyone care to elaborate on chromium carbides and their care and feeding? And some questions specifically for Howard: if I did ever attempt to pattern weld it to something lower in C like 1045 (how about 1018? or is it too different?), I would expect the carbon to migrate toward equilibrium somewhere between .45 and 1.00 (would this happen before the grain got too large?) but what would the other 52100 elements do?  The pieces I have are not large, so if I could get the stuff to forge weld I could do san mai or a low layer count pattern weld. Do you have to add fluorite, or is just borax a sufficient flux? Is it any easier to work with the 52100 after it's 'diluted' in layers with a lower alloy?

 

At this point I'm mostly concerned with making a blade that is functional and adequate, as opposed to wringing every drop of hi-tech potential out of a particular alloy. That is to say, I'd rather be able to reliably get, say, 55-58 HRC and relatively durable than to occasionally hit 60+ and lose a lot of work to cracking etc. I'd like to heat treat it myself if at all possible, rather than paying someone else even if they might get slightly better results. I'd like to use a forge whenever I can in the process, because fire is fun  :) If there is no good way around it, I'll have to use these pieces for some little stock removal blades and go with a commercial heat treater, but I'd rather not.

 

I forsee a lot of experimentation coming up, I'd just like to minimize the dead ends by not repeating what others have already done. So, thanks all, and keep those thoughts a-comin.

 

Michael

Share this post


Link to post
Share on other sites
Guest daryl meier
I have experienced no problems in forge welding 52100 to itself and to other steels.  However; when it comes to heat-treatment, 52100 marches to a different drummer. Certain clad and patternwelded combinations that include 52100 can experience extreem internal stresses during hardening(cracking occurs). Reread Howard's last paragraph several times.

Share this post


Link to post
Share on other sites

There's a nugget of info on 52100 on Howard's site, under the salt pot section:

 

http://www.mvforge.com/salt%20pot%20info.html

 

"Temperature control from this method is unparalleled, and allows one to manipulate the microstructure a great deal more effectively than "eyeball" methods. It is especially good to use if you intend pursuing 52100 as a blade material. What that steel does in heat treatment is very dependent on austenitizing temperture, and the results can be dramatically different with as little as a 100F change in the temp prior to the quench. If too much of the available 1%C is dissolved into the austenite solution, you will get "plate" martensite, which is more brittle, always, irrespective of actual tempered hardness as measured on the rockwell scale. Austenitizing temperature for 52100 should normally be held to 1550F or less. At 1550F, there will be .55-.6% of the available carbon dissolved into the austenite, with the remainder present as retained carbides in the martensite that is formed in the quench. This is good. Over-heat the blade to 1650-1700F (real easy to do by the "eyeball" method, and you may put as much as .65-.8% of the carbon into solution. This is bad. Plate martensite results, rather than the more desireable "lath" martensite, the retained carbides are fewer, or not there at all (which reduces the materials wear resistance or edge holding ability, and IMO reduces the potential "sharpness" of the blade as well.)"

 

There were a couple other posts on SFI from Howard on the subject, way back when, that I saved somewhere.  Note that those are old posts, and it's possible that they don't reflect Howard's most up to date knowledge on that, so if there are inacuracies, bl;ame me, not him.  Here's a copy:

 

From: Howard Clark

Where: SFI Discussion Forum

URL: http://swordforum.com/forums....5&part=

 

"The grain will not grow unacceptably with a long soak at 1550 with 52100. As part of one of the experiment series I did for/with JDV (my metallurgy professor friend), we compared grain sizes for two sample heated at 1550f, one soaked at temp in the salt pot for 3 minutes, one held at 1550f for 9 hours. There was no difference. 52100 has very strong carbide formers (high carbon and chrome for it to bond to). Until you heat it hot enough to dissolve all those complex carbides into solution in the austenite, the grain does not grow. The temperature required to do that is over 1700f. The only thing critical about soak time is to allow enough to bring the whole piece up to the temp you want, rather than have a differential between the body and the edges (unless that is what you want. The simplest way to get hard edges, and a not hard body, would be to heat it up in a very hot forge, quickly. Theedges will come up fast, the body not, and you can selectively austenitize only the edges, leaving the body soft. The negative side of this is that the body will be very soft, not just a little bit, unless the whole thing was through hardened first."

 

 

From: Howard Clark

Where: SFI Discussion Forum

URL: http://swordforum.com/forums....b=&vc=1

 

"The reason for mutiple quenches is to refine the austenite grain size, get it uniformly small, and to set up a preferred microstructure for the final hardening operation. It is not so critical on pieces that are not forged, and have not been subjected to intense localized heating that was uneven, which happens in all forging operations. The reason it works so well (and it does, trust me on that one) is that martensite, upon heating, precipitates epsilon carbides, which then impede the growth of the new austenite grains once you reach a high enough temperature to form austenite (critical temperature, aka A1).

 

So, the first time, you get a hardened blade, but the grain size may not be even and consistent throughout the piece, due to the aforementioned uneven heats. The second time, you are closer to all the same, and the grain size is smaller. The third trip, the grain is smaller yet again (provided you control the austenitizing temperature and don't over-heat and cause grain to grow again, undoing the whole process), and bu this time it is more or less evened out so the grain size is the same throughout the piece.

 

Make sense now ?

 

You can get almost all the same benefits from multiple normalizing cycles, but it may take more than three to get you there. There is also no hard, fast "one size fits all" rule here. Everything is variable, depending on the steel, how hot you got it in forging, and the rest of your shop practices. But, in general, it will produce a finer grain size, and better performance in the finished product. I suggest doing the multiple quenches for grain refinement BEFORE the grinding and finishing, then doing only one hardening operation on the finish ground blade, it saves a lot of trouble."

 

A follow up question from me was "why is it you should let the steel rest for 24 hours in between the quench cycles?"

 

Howard's answer: "There is no need to do that. It matters no thow long one waits, 24 hours, 24 minutes, it makes no difference. [...]

 

Forge the blade to shape until you are happy. The bevels are where you want them, and you have a staright blade. Then, heat it up to some temperature above critical, bya margin, then quench it. Repeat with a lower austenitizing temperature, and then repeat again, also from a lower temp. After the third quench, then anneal it (the method I prefer for many steels is to place them in a pre-heated oven at 1300f and hold them there for one hour minimum). Cooling rate at that time is unimportant."

Share this post


Link to post
Share on other sites

Howard, Daryl, Mr... I mean Randal, thank you for your advice.

I have noticed the bad points, but I think I'm getting around them. Howard is entirely correct about good controls. I think you'd have to have a very good eye indeed to optimize 52100. After I got my salt up and running I found out how bad my perception of the right temp is.

The carbon distribution dynamics in Joss's post make alot of sense. I've had complex forgings go short and start to crack on me after a long forging session. Is this when the carbon distributes itself evenly and all the carbides dissolve?

Given all its bad points, is properly treated 52100 that much better than any other type of steel, or am I wasting my time?

Another question that might clear up some "confusion": Given that the smith can control the grain size through thermal cycling, what benefit does beating a three inch ball into a knife hold over starting with 1" round stock and just forging to shape? You aren't going to make the grain any smaller, does something else happen or am I confused?

Share this post


Link to post
Share on other sites

There is no advantage to one stock size over another for starting material, except convenience for the operator.

 

I do not think it is worth the extra trouble compared to the 1086 that I use all the time, for myself. It 'can' be a little better in terms of edge holding, but the extra work required makes it a poor trade, for me. I will not knock the material, it is in fact made to very high melt practice standards, and the quality of it is always good, as a bearing material, it has to be so. It can make very good blades that are as good or better than most simple and low alloy steels can be if it is heat treated well, and there was no other major processing mistakes made before the last heat treatment. The set-up processing is more ciritical than the final heat treatment in my experience with it.

 

Here's my unabashed opinion.

 

It is very good quality steel. It can make the very best quality blades. It is versatile, and NOT air-hardening no matter what anyone says to the contrary. You can watch it change from austenite to pearlite right before your very eyes in the dark if you pay attention, it is a visible phenomenon, just like with simple carbon steel, bright and dramatic. It is not for the amatuer without temperature measurement, and preferably, controlling devices. The pitfalls are many.

 

She is a hard mistress. Treat her with the exact methods required, and it is a wonderful thing. Mess up (and there are oh soo many ways to do it) and you would have been vastly better off with something else, less demanding.

 

Learn to read and understand the isothermal transformation diagram, then study it in detail. Observe careful, controlled shop practice, and work carefully, and it will make great blades. Do it not carefully and with understanding, and they will not be good blades in the end. At least not nearly as good as they could have been. It is a steel with tremendous potential, but learning to tap that potential to it's greatest advantage is not a beginner's road. Other steels, with simpler rules should be learned first. Then if you want to delve into the mysteries of heat treatment and metallurgy where tiny differences in temperatures and times can make big differences in the end result, this is a fine way to explore those areas of study.

 

Is it worth it in the end ? Not for me, unless the customer really , really wants it, or for some reason I have valued edge holding over all other considerations for the project, or I really need abrasion resistance in some other kind of part I am making. Otherwise, I'd rather use some other steel, most of the time.

 

I have made three katana out of 52100, and heat treated them so that they had tempered martensite edges and pearlite bodies, just to prove that it could be done. The hamon never showed up well, with any of the methods I tried, and though they are excellent swords (I destroyed the first one intentionally beating it to death to find out) the material is just not worth the trouble for me again.

 

All the while bearing in mind that I am in fact a complete nutcase, and usually pick the harder road just because. Ask the  guys that know me, I am not kidding.

 

But hey, if you like 52100, it's ok with me !  :D  Life is good, be happy, have fun.

Share this post


Link to post
Share on other sites

Thanks for the reply, Howard.

I may be thick headed, but I'm going to keep using this steel. Looking at the progress I've made with it in the last year, I think I am on the road to somewhat controlling it. Besides, if it's easy, why do it?

On the CCT/TTT diagrams, do you or anyone else know where I can get a copy of the diagram for 52100. I know there is an atlas out there with diagrams for many alloys, but I haven't found a copy in my price range. So far I've been able to find a few specific numbers from the bearing makers, but I've never seen the actual diagram.

Another question (stop answering if you get tired of this): Dissolved carbon, if you keep this low and keep much of it in carbide form, does this explain why some smiths can water quench this steel consistantly. Joss's post said only .55 to .6% of the carbon will be dissolved in the austenite, doesn't this put the hardenability into the water quenching range? Please correct me if I'm wrong, but I believe it does. I've found that the knife will not be as hard (file test) after a series of multiple oil quenches as opposed to a single oil quench.

Does a fine grain size mean that carbides are present? I guess I'm looking for a visual clue that I'm on the right track. Can you have a fine grain and still have too much carbon dissolved into the austenite and ultimately the martensite?

One more and then I'll stop (for a while anyway). Bainite vs. Martensite: Without a microscope, can you tell what structure you have?

 

Sorry to keep bugging you, but I love cold hard facts, as opposed to stories and myths. (Although I do like the one about the mysterious spontaneously softening knife blade.)

Share this post


Link to post
Share on other sites

Yes, fine grain and low austenitizing temperature means you can water quench it, though I do not suggest one continuous cooling,  but rather interrupting it after a couple of seconds. All you really have to do is get the steel to a temperature below the nose of the curve on the left side of the IT diagram. This is over in three seconds in thin sections, and four or five in thicker pieces in water, IF the austenitizing temperature is not too high.

 

52100 will harden from as low a temp. as 1380f if the setup is right. It will be relatively low carbon martensite, with loads of retained carbides. By carefully controlling the austenitizing temperature (salt pots kick but here) you can control the amount of carbon in solution, and thus control the "as quenched" hardness very closely. Add variables quench rates to the equation, and the variations are almost limitless, and measureable with small changes. This is why the old "eyeball" thing doesn't work all that well with this material. A change in 100F can make a significant difference in how much carbon is in solution, which affects the hardenability, the amount and size of the retained carbides, and also affect the response to tempering. Changing the austenitizing temperature and the prior microstructure changes EVERYTHING else to some degree.

 

Once you learn the dance, you can make this material do nearly anything you want, but you have to have measurement and control instruments, and you have to pay attention. My data, measured with good instruments, disagreed significantly with "The Heat Treater's Guide" on this steel, in several circumstances. Tweak the routine, and you can generate data that matches exactly. Change it a little more, and you get new numbers again.

 

Starting to see a pattern here ?

 

The commonly available IT diagram actually has two samples with different austentizing temperatures over-lain on one diagram, and it is very confusing. I have not seen a single sample one anywhere. That does not mean it doesn't exist, just that I am not aware of it. Perhaps someone with a better grasp of Photoshop than me could remove one of them from the dual diagram, I do not know.

 

I am not trying to discourage anyone from using the steel. I AM trying to discourage unreasonable expectations of how it works. :)

Share this post


Link to post
Share on other sites
Guest daryl meier
Howard et al:  Here's a picture of a page from "Atlas of Isothermal Transformation and Cooling Transformation Diagrams" ASM 1977 no isbn# shown.  One set (aust temp 1550) crudely removed.  This shows aus temp 1950f  WOW.

Share this post


Link to post
Share on other sites
Guest daryl meier
Here is the single diagram for aus temp 1550f.

Share this post


Link to post
Share on other sites
Guest daryl meier
And here is the continous cooling diagram with jominy end quench lines superimposed--- it would be toooo much effort for me to remove them.

Share this post


Link to post
Share on other sites
Guest daryl meier

Sorry about the quality of the last image.  Lots of difference in the lines between th iso and continous!!!!  Also please note that the nose of the pearlite curve moves about 100f, and the Ms temp line also moves when the aus temp changes from 1550 to 1950f on the iso's. ( aus grain size 9 and 3)  Wonder where the pearlite nose is with #14 aus grain @ 1380f ????

certainly not a steel for childs' play.

Share this post


Link to post
Share on other sites

Yeah, thanks a bunch Daryl!!!

 

Here's a test I did yesterday. The left-hand pieces are from one blade, the right are another, both sets were from the same stock. The left hand ones were heat treated by eye and a magnet in an open forge. Number one was just heated and oil quenched (one cycle), after forging. The next piece was oil quenched after two cycles. The next was three cycles and oil quenched, the fourth was four cycles and water quenched, no interruption (you can see the quench cracks). No tempering was done on any of these pieces, I just snapped them off when they were cool enough to handle. You can see the grain size reducing in each section for each cycle. You can also see how it became tougher and tougher to break the blade. It started to tear instead of snap cleanly the more cycles I conducted. The bad news is the third piece shows larger grain in the thicker section, meaning I must have overheated it and ruined some of the previous work. My eye is not good enough!!!

 

The last two sections were heat treated with my salt pot. I did one normalizing cycle of 1600 degrees and air cool until the color left the blade. Then three grain reduction cycles of 1550 for the first, and 1500 for the second two quenching the whole blade in room temp oil each time (actually the oil was slightly warmer each time, but never over 100 degrees). For the final cycle, I heated to 1500 and edge-quenched the blade until the color left the back, then quenched the whole blade. The oil was at 150 degrees at the start of this quench.

After the blade was cool, It bent to about 30 degrees before snapping. This piece is second from the left. I then tempered twice at 500 degrees and bent it again. It now went ninety degrees one way, back straight and about 45 degrees the other way before cracking at the edge. I had to flex it twice more before the back tore off.

One thing apparent that I haven't noticed before is a series of curved lines across the cross-section in the tempered piece. These lines are in the transition zone between the martensitic edge and the softer (pearlite?) back. Are these upper and lower bainite? Something else? More mysteries!!!

 

I know this test is far from scientific, but I really want to know I'm on the right track. Please help if I'm not. Any comments extremely welcome.

 

Thanks in advance,

Tom

Share this post


Link to post
Share on other sites
Guest daryl meier
Here is another attempt to share the continious cooling diagram posted earlier.  Hopefully this image is more readable thanks to Don.

Share this post


Link to post
Share on other sites

Tom,

 

Thanks for sharing the samples. That is the way to approach learning the steel, very dramatic.

 

Daryl,

 

I have trouble reading charts initially and this one is baffling me. Could you walk me through what you find notable? Thanks

Share this post


Link to post
Share on other sites
Guest daryl meier

Don:  Spent the evening erasing the jominy lines from the cooling diagram, no energy left to talk about what it shows.  Maybe tomorrow.

I uploaded the diagram to one of my "aol hometown pages".

 

52100 continuous cooling diagram52100 continuous cooling diagram

Share this post


Link to post
Share on other sites
Sign in to follow this  

×
×
  • Create New...