Humbled by Heat Treating Results. Questions/Discussion.

D

Don Nguyen

Joined
Oct 4, 2011
Messages
1,043
I recently started working with 1084 from Aldo.

I forged and ground a 9.5 inch Chef's Knife (my 5th knife); didn't grind bevels, just the distal taper. 2.5mm thick at the heel, about 1.5mm thick near the tip. Overall height at the heel 2.5 inches. Heated above non-magnetic as even as I possibly could in the forge, and quenched with back-and-forward agitation in 140 degree F canola. Tempered at 400 F for one hour, twice.



By nature of the quench container that I used, I quenched the main body of the blade first, then after the color faded from the spine and tang, quenched the whole blade itself afterward. This means that I had to quench with the blade angling downward, where less of the heel actually made contact with the oil initially; however, the heel DID contact the quench, about halfway up towards the spine.

I tested for HRC today in my metallurgy class (completely separate from the knifemaking), after cleaning up the scale and whatnot. At the heel, 30 HRC. One inch forward, 45 HRC. Another inch forward, 60 HRC. The rest to the tip was 60 HRC. I was quite sure that my angled semi-edge quench caused the heel to be softer.



My metallurgy professor, who has a Ph.D in the field and had taught at the University for decades, asked me details about the knife etc, and I told him it was 1084 quenched in oil. He said I went wrong quenching it in oil, and that it should have been quenched in water. Even if I had gotten 60 HRC along the blade, it was probably only on the surface he told me.

So, I'm kind of thrown off by this. Thrown off first of all that the heat treat I thought succeeded was actually not that great, and also by the case my professor made, mainly because for our craft, it's preached by the majority of knifemakers to quench in a fast oil rather than water for a 10XX steel.


Really looking forward to responses. Thanks for your time.
 
As to your hardness results, I would think you experienced an "auto" temper where heat left in the thicker portion of the spine/ricasso/tang that was out of the oil and bled into down into the initial part of the edge near the plunge, slowing its cooling and resulting in pearlite or even tempering down the martensite that you may have formed. Also, the limited amount of space in your quench tank may have reduced the effectiveness of your agitation. I'm curious; how much oil does your tank hold? I don't feel comfortable without at least a gallon to disperse the heat.

As to your professor, he has to realize we're dealing with very thin cross sections. If we were quenching large, blocky pieces, he'd likely be correct regarding surface hardening. But with opposite surfaces only 1/32" to 1/4" of an inch apart from each other, even surface hardening on either side results in through hardening. Canola may not be ideal, but it is pretty much fast enough with thin blades of 1084. As for quenching in water vs. oil, even industry quenches in formulated oil vs. water, so I'm not sure where he's coming from there.

--nathan
 
Thanks for the comment Nathan.

Hmm, I hadn't thought of that heat bleeding thing. I was using one gallon, but it can probably hold 2 gallons. I just don't have 2 gallons on me, however. Rather than using the horizontal troft (sp?) that I was using, I might go with a vertical quench next time to avoid such issues again. I had trouble getting clearance with the tongs and the walls of the troft, thus the angled quench.

I did show him the knife, and he got a good look at it. He is pretty old fashioned, but maybe that has nothing to do with it...
 
Do you know the grain size? I recently read Verhoeven and he referred to a study on 1086 quenched in oil. It seems a courser grain size will move the pearlite nose giving more time to lower the temperature. 1086 with a grain size of 11 did not develop pearlite to 0.22" while 1086 with a grain size of 15 started loosing hardness immediately. The minimum thickness tested was @ 0.08". tests were done on untempered steel starting @ 66 HRc.
Are you going to do another HT on the blade?
 
I always take the "surface quench" comments from non knife folks with a grain of salt. They tend to be thinking in terms of large parts or complex shapes for other industries rather than our thin flat profiles. Yes, if the object was a 6" cube it only got surface hardened, but a knife thickness object was not as thick as that surface and had no hot core to keep spreading heat around so it cooled even faster.

I think you hit your situation spot on, you didn't get the heel in early enough and the heat from that area spread down a bit to reduce the quench rate slightly further than just the spot that didn't make it in the oil. I'm actually not sure how it acts, whether it reduces the transformation and you never get fully hard or is acting as an over heated tempering effect as the heat diffuses from the unquenched area to the area that did make it in.
 
also, Canola isn't exactly a fast oil...

I use Parks #50 and it's great, I used water for years before that with good result
 
I have to ask what you hoped to accomplish with the edge quench? If it was to get the beautiful temper line, thats fine. If it was simply to achieve a hardened edge and softer spine, I would have gone a different route that involved a full quench with differential tempering. Much easier on the blade and your confidence.

As for Professor Ping... just smile and wave... smile and wave.
 
Thanks, Rick. I needed a good chuckle this morning, "Professor Ping" did it for me. :)
 
When you say, ".....heated above non-magnetic", that statement is pretty vague.
Please consider that non-mag is only around 1414 or so, and even if you DID! get above non mag in the area you chose to check with your magnet, which was probably somewhere in the center of the blade due to its ease of access, you may not have had a full austenization in the heavier mass area of the heel.
If you DO get a full austenizing of 1084, you can quench it in spaghetti sauce and you're gonna get a LOT! harder than 30 Rockwell.
You should have file checked that right out of the quench BEFORE you tempered. At that hardness, it would have been easy to discover the non-transformation.
In that case, you could have done a thermal cycle or two and just re-heated and quenched again.
 
Bo T said:
Do you know the grain size? I recently read Verhoeven and he referred to a study on 1086 quenched in oil. It seems a courser grain size will move the pearlite nose giving more time to lower the temperature. 1086 with a grain size of 11 did not develop pearlite to 0.22" while 1086 with a grain size of 15 started loosing hardness immediately. The minimum thickness tested was @ 0.08". tests were done on untempered steel starting @ 66 HRc.
Are you going to do another HT on the blade?
Click to expand...

I don't know the grain size, but I will be requenching it again.

stephanfowler said:
also, Canola isn't exactly a fast oil...

I use Parks #50 and it's great, I used water for years before that with good result
Click to expand...

Definitely was not aware that canola wasn't that fast. I got the impression on this forum and other internet sources that canola was actually quite fast (at least, faster than motor oil, slower than water).

Rick Marchand said:
I have to ask what you hoped to accomplish with the edge quench? If it was to get the beautiful temper line, thats fine. If it was simply to achieve a hardened edge and softer spine, I would have gone a different route that involved a full quench with differential tempering. Much easier on the blade and your confidence.

As for Professor Ping... just smile and wave... smile and wave.
Click to expand...

I did not exactly choose to do an edge quench, I just didn't notice that I didn't have enough clearance from the tongs and quench tank wall. For the future quenches, I'll make sure that I can actually dunk the whole blade in before I try... :rolleyes:

Karl B. Andersen said:
When you say, ".....heated above non-magnetic", that statement is pretty vague.
Please consider that non-mag is only around 1414 or so, and even if you DID! get above non mag in the area you chose to check with your magnet, which was probably somewhere in the center of the blade due to its ease of access, you may not have had a full austenization in the heavier mass area of the heel.
If you DO get a full austenizing of 1084, you can quench it in spaghetti sauce and you're gonna get a LOT! harder than 30 Rockwell.
You should have file checked that right out of the quench BEFORE you tempered. At that hardness, it would have been easy to discover the non-transformation.
In that case, you could have done a thermal cycle or two and just re-heated and quenched again.
Click to expand...

I'll be sure to check with the magnet along the whole blade in the future. It's very strange though, because the whole blade, excluding spine and tang, skate the files. I tried to do the hardness test as close as I could to the edge, about half a centimeter away, and it still showed lower HRC results. Maybe it's only hard on the very edge itself? I'm not sure.



Overall this was a good experience for me, because now I learned a lot of the issues I have with my processes that I would have left unchecked. I was about ready to finish this poorly HT'ed blade and call it done! Even though I'm impatient, I'll give it another shot with everything I learned in consideration. Thanks for the responses everyone.
 
mike nelson said:
did you normalize the blade?
Click to expand...
I was think more along the line of .... "Did he spherodize the blade, then not get the heel into full solution during HT" It would raise even more questions if it was 1095. I take it that he did normalize. Did you happen to stick it into vermiculite or wood ash after normalizing?
 
Last edited:
I normalized 3 times, all progressively lower temperatures from slightly above non-magnetic to non-magnetic. I didn't anneal it otherwise, however.
 
Note to self: get some spaghetti sauce and try to quench a 1084 blade in it, just to see if Karl is full of crapola. ;) :D
 
Vegetable oil works pretty good and my college level machine shop practices book recommends it over most stuff DEPENDING on what you are doing. You need to get a bigger quench area even if it is just a small metal garbage can with a couple gallons in it. It does heat up quickly sometimes and having some water nearby can help boost the quality of your quench.

I second the file test right after quench as it is a relatively simple thing to re do it.
 
The file test is quite unreliable as to the acual condition of the steel. If you have a mix of pearlite and martensite, it still tells you that everything went well, when it really didn't. Unfortunately, the file is all many of us have to work with, but especially with 1095, don't bet the farm on what it tells you. Sheep and files lie!
 
LRB said:
The file test is quite unreliable as to the acual condition of the steel. If you have a mix of pearlite and martensite, it still tells you that everything went well, when it really didn't. Unfortunately, the file is all many of us have to work with, but especially with 1095, don't bet the farm on what it tells you. Sheep and files lie!
Click to expand...

I didn't suggest that the file would tell him how hard it was, I suggested the file test would have told him how SOFT it was before he wasted time tempering it.

I said, "You should have file checked that right out of the quench BEFORE you tempered. At that hardness, (which would have been the 30 RC he was referring to) it would have been easy to discover the non-transformation."

At 30 Rockwell, it would have been DEAD SOFT and there would have been no confusion about being any combination of pearlite, martensite, or any other phase. It would have shown it to be dead soft.
 
Last edited:
You must log in or register to reply here.
Back
Top