-
The BladeForums.com 2024 Traditional Knife is ready to order! See this thread for details: https://www.bladeforums.com/threads/bladeforums-2024-traditional-knife.2003187/
Price is$300$250 ea (shipped within CONUS). If you live outside the US, I will contact you after your order for extra shipping charges.
Order here: https://www.bladeforums.com/help/2024-traditional/ - Order as many as you like, we have plenty.
1095 Cro-Van and 50-110B Types of Steel
- Thread starter D-weaver
- Start date
- Joined
- Oct 14, 1998
- Messages
- 2,155
up to about 13 inches, but the end of the tool (long chisel at that length) is the important part, so i would have to be accurate enough where the end would be.
I've eyeballed a few used pottery kilns here but most of the inexpensive ones are top door, which I'm assuming is not so great for heat treatment. they've gotten my attention because of the pencil lead thing - pencil leads are graphite and clay mixed and need to be fired.
I've eyeballed a few used pottery kilns here but most of the inexpensive ones are top door, which I'm assuming is not so great for heat treatment. they've gotten my attention because of the pencil lead thing - pencil leads are graphite and clay mixed and need to be fired.
My pottery kiln does a pretty good job. You can often find the round ones that are around 14-15" ID. It's top door, with the heating elements in the top lid, but goes to 1715 with a PID controller. You lose heat opening and closing, but it's back up to temp pretty quickly. Mine takes about 90 min to get to 1500, but its a 110V model and has a lot of extra space to warm up. I will probably make my own HT oven down the road, longer/narrower cavity, front door, etc.
I sent you an email or someone at AKS about not needing to normalize O1 and 26c3 and that I thought they were easier to do in the open atmosphere than 1084, which a lot of woodworkers will try and fail with.
Spheroidized O1 is especially easy to do and get good results with if one heats it quickly well past critical and quenches. 26c3 needs parks.
I pointed it out because I constantly run into people on the woodworking side who read but don't do, and they'll ask for my advice and tell me it's wrong. You mentioned needing a thermocouple, but it's not needed for those two steels.
I unintentionally proved my comments by providing 1095 and 1084 samples to Larrin that weren't as good.
The method that i use for 26c3 and O1 takes about 10 minutes for several thermal cycles and a quench - it's a practical thing for someone who wants to make a higher hardness knife with a simple steel.
Results - Chart
(I wrote the 1095 result in, thus the scribbled look - some guy asked me if I would make him chisels out of 1095 and just with routine testing of stuff I've made, it seems a bit hard tempered and under tough - Larrin confirmed that for me)
I overheated one O1 sample (I cycle them in pairs so that I can see and handle them by eye/color and then quench everything individually) - you can see that. I threw the bum O-1 sample in, anyway - no way to prove that it's the flyer in the O1 samples, but the odds are in favor of that.
Where I made my mistake after getting the O1 samples and 26c3 samples was believing that there might be some interest in the cycle that I used - that's not really been the case. I'm perfectly happy to not make waves as I buy most of my steel from you. ....Actually, the only reason I sent 26c3 and O1 samples for testing was constant goading from people who said that it's not possible to come up with something that will match commercial results. "Leave the steel to the experts, you're wasting your time", something like that.
400F temper for 1095 and 390F for O1 and 26c3 in the results above.
26c3, because I only have large tongs, was a little to a couple of points softer on one end of each sample, which may help explain toughness a little better than expected.
Last edited:
- Joined
- Oct 14, 1998
- Messages
- 2,155
I'll send it to you PM.
- Joined
- Oct 14, 1998
- Messages
- 2,155
My Dec 27, 2021 email response to D-weaver:
D-weaver, based on my email response, I do not understand why you wrote I do not "like forge heat treating that much". Repeating what I wrote in the Dec 27 email, "When austenitizing O1, it is important to equalize the temperature and hold for 15 to 30 minutes.". I like heat treating done properly. Not following heat treat processes results in variable and inconsistent results.
We frequently heat treat with forges in our shop, especially when testing alloys that will be heat treated in forges by our customers.
Chuck
D-weaver, based on my email response, I do not understand why you wrote I do not "like forge heat treating that much". Repeating what I wrote in the Dec 27 email, "When austenitizing O1, it is important to equalize the temperature and hold for 15 to 30 minutes.". I like heat treating done properly. Not following heat treat processes results in variable and inconsistent results.
We frequently heat treat with forges in our shop, especially when testing alloys that will be heat treated in forges by our customers.
Chuck
Ahh...I assumed that you were describing something that wasn't practical for someone who is heat treating by hand and eye (in a forge, which to most people in woodworking will mean a forge with no ability to soak anything accurately for any period of time), and assumed that you preferred limiting heat treatment to a furnace.
I do something different, and have not had any issues with O1 or 26c3 - never had a poor performing sample of either outside of maybe the second attempt to heat treat O1 when I had too many BTUs of heat that wasn't hot enough, of that makes sense - a large flame without much focus.
What I am doing in the samples that I sent to Larrin is three quick heats that never become nonmagnetic but that get very close- the trick is training your eye in a constant light environment to see just about exactly what the color of steel is when it changes from magnetic to nonmagnetic. After that, very quick heat (as fast as possible) to a point well past nonmagnetic and then a quench in a fast oil (O1 is fine in vegetable oil) and getting whatever is being heat treated finished in cold water and then dumped in the freezer.
I found the samples, because I usually do this with larger items, smaller and harder to control temperature on. That is, I would expect when I do a chisel or plane iron, the results are more consistent with less risk of overheating while cycling.
If a steel needs to be normalized like 52100 suggests to me (from experience with results) that it would like, then I just don't use it.
it's clear from my results that I need to come up with one alternative method for steels that don't seed carbides as well as 26c3, or that require less process. I'm experimenting with those, but slowly -I don't use them.
I read the data sheets, of course, but don't follow them as forge heat treatment can be done differently. My goal is to do the heat treatment as part of the toolmaking, and not separate from it. to have the item being finished in the freezer within 10 minutes including thermal cycles is ideal. The results are reliable. I can't make generalized statements for anything else, as I specialize in the two steels and those are the first samples I ever had analyzed. I'd made a couple of hundred things over the years by that point, and since woodworking generally pushes the limits of edge stability, I've had a feedback loop that many wouldn't have the luxury of having. As in, if I find a mid 1800s chisel, pick the best one and then see if I can match its performance, that's a pretty high bar to meet before anything ever gets sampled. If steel is 60 hardness, it'll never pass that test. If it's coarse grained, it'll never pass that test.
I hand finish the tools after quench and many are long and thin, so managing warp is sort of a different topic, but I haven't had to make much of an adjustment to the above - or any. The controlling of warp is done by heating as evenly as possible, even though the heat is very quick, and then quenching as evenly as possible. I don't do heavy forging, just heat steel and shape it, which also helps avoid grain growth before the thermal cycles.
What I was posing with this is that most people get stuck in the mud thinking they need to soak steel in a can forge and try to do it for 10 minutes to 30 minutes and when someone is new and they try to do that, nothing good happens. I think it's possible to get results in line with published results for those two steels doing no more than what I described, and more practical for someone with no thermocouples, no furnaces and limited skill. It takes some initial learning with a magnet and then paying attention while doing the cycle, and perhaps snapping samples to see what really results in finer grain (which is what I did) and then testing the tools to confirm.
if i flatly say the soak isn't required, it will cause a lot of people to get upset. But you can see my results with test coupons without soaking, so I think it's not required.
What got me to this point is that some of the best chisels I've ever used were made in the mid 1800s. I could be wrong that they didn't soak steel back then, but I doubt they soaked steel. The fineness and hardness of the steel is stellar and was probably done quickly by a small number of skilled workers.
I do something different, and have not had any issues with O1 or 26c3 - never had a poor performing sample of either outside of maybe the second attempt to heat treat O1 when I had too many BTUs of heat that wasn't hot enough, of that makes sense - a large flame without much focus.
What I am doing in the samples that I sent to Larrin is three quick heats that never become nonmagnetic but that get very close- the trick is training your eye in a constant light environment to see just about exactly what the color of steel is when it changes from magnetic to nonmagnetic. After that, very quick heat (as fast as possible) to a point well past nonmagnetic and then a quench in a fast oil (O1 is fine in vegetable oil) and getting whatever is being heat treated finished in cold water and then dumped in the freezer.
I found the samples, because I usually do this with larger items, smaller and harder to control temperature on. That is, I would expect when I do a chisel or plane iron, the results are more consistent with less risk of overheating while cycling.
If a steel needs to be normalized like 52100 suggests to me (from experience with results) that it would like, then I just don't use it.
it's clear from my results that I need to come up with one alternative method for steels that don't seed carbides as well as 26c3, or that require less process. I'm experimenting with those, but slowly -I don't use them.
I read the data sheets, of course, but don't follow them as forge heat treatment can be done differently. My goal is to do the heat treatment as part of the toolmaking, and not separate from it. to have the item being finished in the freezer within 10 minutes including thermal cycles is ideal. The results are reliable. I can't make generalized statements for anything else, as I specialize in the two steels and those are the first samples I ever had analyzed. I'd made a couple of hundred things over the years by that point, and since woodworking generally pushes the limits of edge stability, I've had a feedback loop that many wouldn't have the luxury of having. As in, if I find a mid 1800s chisel, pick the best one and then see if I can match its performance, that's a pretty high bar to meet before anything ever gets sampled. If steel is 60 hardness, it'll never pass that test. If it's coarse grained, it'll never pass that test.
I hand finish the tools after quench and many are long and thin, so managing warp is sort of a different topic, but I haven't had to make much of an adjustment to the above - or any. The controlling of warp is done by heating as evenly as possible, even though the heat is very quick, and then quenching as evenly as possible. I don't do heavy forging, just heat steel and shape it, which also helps avoid grain growth before the thermal cycles.
What I was posing with this is that most people get stuck in the mud thinking they need to soak steel in a can forge and try to do it for 10 minutes to 30 minutes and when someone is new and they try to do that, nothing good happens. I think it's possible to get results in line with published results for those two steels doing no more than what I described, and more practical for someone with no thermocouples, no furnaces and limited skill. It takes some initial learning with a magnet and then paying attention while doing the cycle, and perhaps snapping samples to see what really results in finer grain (which is what I did) and then testing the tools to confirm.
if i flatly say the soak isn't required, it will cause a lot of people to get upset. But you can see my results with test coupons without soaking, so I think it's not required.
What got me to this point is that some of the best chisels I've ever used were made in the mid 1800s. I could be wrong that they didn't soak steel back then, but I doubt they soaked steel. The fineness and hardness of the steel is stellar and was probably done quickly by a small number of skilled workers.
Looking for what else may have led me to believe you didn't approve of forge heat treatment, from the O1 page on your site:
"Many people think the alloy can be properly heat treated in a coal and/or propane forge. This information is incorrect."
All of the samples above were done in a propane forge. You can see from my order history that I don't drop down into too many steels below the eutectoid limit - they're just not good for woodworking. I tend to get better results with steels that are assumed less suitable for quick no-soak heat treatment, but I've done a fair bit of this and made incremental improvements - mostly in getting higher hardness out of the quench and understanding what not to do if straightness is important (and it always is).
"Many people think the alloy can be properly heat treated in a coal and/or propane forge. This information is incorrect."
All of the samples above were done in a propane forge. You can see from my order history that I don't drop down into too many steels below the eutectoid limit - they're just not good for woodworking. I tend to get better results with steels that are assumed less suitable for quick no-soak heat treatment, but I've done a fair bit of this and made incremental improvements - mostly in getting higher hardness out of the quench and understanding what not to do if straightness is important (and it always is).
- Joined
- Feb 16, 2022
- Messages
- 615
Well alpha knife supply puts the exact elemental make up of the batches of steel they sell. So that's the place I would buy w1 or 2 if you are concerned about that.
Also I believe 52100 is tougher than 26c3. I could be remembering wrong, but I'm fairly sure I remember seeing 52100 having a higher toughness at a given hardness.
- Joined
- Feb 16, 2022
- Messages
- 615
Honestly you can heat treat anything with either setup, if you're willing to put in enough effort, and sacrifice some performance.
I guess my point is that for the two steels I'm showing, you can compare the results to Larrin's write ups. I'm not sacrificing anything with those two.
Without working up a routine and snapping samples with other steels, I don't know if I could match the book on anything else. I think I could, but sending samples to larrin let me know that my routine isn't a cure-all for everything with below O1 and 26c3 on the charts.
I also don't have machine tools, so creating endless samples that would fit in a standard machine requires me to hand dimension the samples, which is a pain.
I wasn't setting out to prove that you could match results in a forge, though, but rather to put together a routine for a bunch of people who claim they want to make their own tools including the metal parts and heat treatment. As in, they're doing something as a hobby and $200 of equipment is doable for a side item, but $2500 isn't. I use about $200 of equipment, maybe $300 - my HT forge is small with very focused heat, and a two burner larger stainless forge that I have is sitting idle in the shop. It makes a lot of heat, but it lacks the control that I really want, too.
- Joined
- Feb 16, 2022
- Messages
- 615
I wasn't being sarcastic or facetious btw. I really meant it. You can heat treat anything If you're willing to put in enough effort. Theres a video on youtube of a guy doing aebl in a forge with a 15 or 20 minute soak at 1950, and hes doing it in a muffle pipe with a thermocouple, and at least the thermocouple is saying he kept it withing 10 degrees difference the whole time. It definitely took a lot more effort though.
I feel like the biggest thing with forge beat treating is to mostly disregard the temperature of the atmosphere around the knife, and solely worry about the temperature of the knife itself, and doing whatever you can to get it even, and to keep the fluctuations of the temperature to a minimum.
I ended up making a kiln though, because I just wanted to be more sure I would have consistent results. , and with the steels I would like to work with mostly are ones that need fairly long soak times.
I've definitely gotten air hardening steels hard by just using a forge though.
Larrin is my oldest son, I am involved in most of the testing that he does. I did the heat treating of the 52100 samples talked about in his book. I’ve been forging for more than 36 years and making knives for almost 45 years.
52100 tested higher in toughness and edge holding over O1 and 26c3, and many other steels.
Larrin told me about the testing he did for you and how poorly it performed.
I’m not trying to insult you but you seem to be looking for validation more than answers.
52100 has the finest carbides and the smallest grain size of any steel that I know of. It is also cleaner than most steels because it usually ends up as a bearing.
I have used a lot of 50100, 1095cv, and 19c5va. It’s a good steel but it’s not as good as 52100.
52100 requires precise heat treatment.
Hoss
- Joined
- Feb 16, 2022
- Messages
- 615
If that is the case. Can I bring up, that you mentioned you keep your steels below critical for anything you do prior to austenizing in a different thread D D-weaver ? If you are forging your steels, or getting them from different suppliers you really aren't going to be able to get a proper normalization, or any kind of thermal cycling in a forge. by keeping them below critical, and that could compound any other problems that are already happening during your ht.
Devin - the second set went poorly. The first set, the results are shown. If they are poor, I wouldn't know why - the O1 samples match Larrin's tables. The 26c3 samples match larrin's hardness but the toughness is higher than larrins.
I sent the second set, and communicated to larrin that they were done with the same method as the first, and probably incorrectly assumed the second set would do as well as the first.
I could repeat the results in the first set (O1 and 26c3) 100 times and the results would be about the same. I couldn't make any realistic comment about 1084 and 1095 because I've made 3 items out of 1084 (tested one as a plane iron, but it was a few years ago) and the 1095. I made one chisel out of (poor results - before parks 50) then after getting parks 50, three plane irons. I never followed up further on those two because I don't use them.
Put differently, I paid my dues on working up 26c3 and O1. I guess with the others, I was hoping I could use that for a free lunch. The results from larrin on those were a good thing to learn - I can't think of anything that provides a better lesson than poor results.
But I can't emphasize how disappointing it is to hear that the conclusion after the first set, as shown, that I only had very poor results.
I could do one of those per day on different days 10 days in a row and send them to larrin and they'd be the same - but for what conclusion - well the results are crap because they were for others? I don't have a great deal of incentive to come up with something for 1084 or 1095 and while I'd say I could do any number of samples of O1 and 26c3 and the result just make more dots on those charts, I wouldn't even go so far as to say "I know I could come up with good results" on the other two, I haven't done it. who knows.
What was I really looking for here? I was hoping someone would perhaps be able to point me to a retailer who had 50-100B or 50-110B and just didn't have it listed. Or anything somewhere close to that.
The problem with 52100 having too much toughness for woodworking tools is a legitimate problem. It caused problem in chisel tests for me - 26c3 next to 52100 and the fact that the edge will hold deformation instead of letting it go leads to significantly more effort being spent with a chisel to do something like cut half blind dovetails in furniture. Before that, I questioned why stanley and others didn't use it early in the 1900s - they used water hardening steel, oil hardening steel and what seems like something with a little bit of tungsten in it. I think I know why now.
Not sure what triggered the discussion of the second set being poor in the first place - was it being surprised by the good results on the first one? I don't know.
This whole lump it all together and the results were bad is annoying, though, but it doesn't change outcomes - I still have only two steels that I use and I'm looking for a third. The fact that nothing else shakes out other than maybe W2 is more disappointing. If you had a pile of 50-110B and you wanted to tell me that my shoes were dirty but you'd sell it to me - I'd be perfectly happy with that.
Lastly, the fact that I can nail the two steels that I generally work with has some to do with things i learned from larrin's site (how much time do I have, to chase hardness at the tail end by getting the steel as cold as possible as fast as possible) - I guess it is what it is. I'm more pleased to learn it and less concerned that larrin's more interested in the bad results than the good ones - I'm not a sentimental guy.
Good luck
