52100 Ultra low hardening/aust temperature

BluntCut MetalWorks

BluntCut MetalWorks

Knifemaker / Craftsman / Service Provider
Joined
Apr 28, 2012
Messages
3,475
I will harden 4 paring + a 24cm gyuto 52100 3/32" thick blanks with ultra low temperature. gyuto: 1440F; paring: 1435F, 1430F, 1425F, 1420F with 10 minutes soak (evenheat oven - calibrated temperature). My target working(after temper) hardness gyuto: 64-65rc, paring: 62-64rc. I've tried (probably around 60 blades by now) all hardening temp from 1440F to 1550F with good result. 1475-ish is well proven for many, including myself but I want to explore whether 'not-enough-carbon in solution' is applicable to my ht or not.

Recently I hardened 2 blades (a petty & bread knives) aust at 1440F 20 minutes soak. Grain ballooned up to blades aust at 1480F. So this time, I cut the soak time to 10 minutes. I destruction tested the bread. Kept the 63rc petty.

Let's assume that these blades are well prepared(normalized+grain refinement+fine spheroidize+etc...) according to your best ability/knowledge/conjecture. What do you think and advise before I proceed?

Thanks,
Luong.
 
Luong, I am absolutely intrigued by you, and your unique sense of experimentation and exploration. I read your posts, and am amazed by how prolific you are, as well as how driven. I hesitate to comment on your posts, mainly because I think you'll take my position as that of a nay-sayer. To a certain extent, I am, in point of fact. The single biggest reason is that, unless I'm mistaken, you're basing many of your conclusions on assumptions, whether they be educated assumptions or not.

Have you had any of your work examined through modern metallography? Are you partnering with anyone that might be able to examine your samples and affirm your conclusions? For example, you mention that the grain 'ballooned' when you austenitized them at 1480, but other than you taking that as a known effect due to previous studies by others, are you certain that's what is happening? There are so many mechanisms involved with this, and perhaps I'm just stuggling to follow you due to the brevity of your posts, but there seems to me to be too many assumptions made in your conclusions for me to view your results as meaningful. Does that make sense?

Also, is the book in your signature written by your wife, and about you? If so, I would love to get a copy of it, and perhaps have it signed by the author and the subject...:)
 
if we assume the blades are well prepared, then i won't definitely expect 52100 aus-grain to balloon up at 1480F.
That said, if your temperature measurement instruments are ok, i suggest to stick with 1475F; this very temperature has been kindly offered by people who spent a lot of time and efforts in finding the sweet spot for the object steel, backed up by laborious metallographic examinations and sound testings. Less then this and you won't get maximum martensite strenght, more then this and you'll start to loose hardness due to RA. It is really simple as that.
52100 is also a steel which benefits of very tight alloy specs, hence less surprises during ht then other steels which comes in many "flawours".
is 52100 so boring in itself that many like to add spices in it's ht? ;)
 
About 9 months ago, I did a series of tests with 52100. I normalized, then grain refined, then tried 7 coupons austentized in 10f increments from 1445 to 1605. I started at 1475 and went out from what was considered the sweet spot. All I found was that as quenched Rc#s dropped and became less consistent as I moved away from the sweet spot. I got Rc66/67 at 1475. It's all I use now for this steel.

What is it you are trying to accomplish by rejecting what is known about this steel? Like Matthew, I don't want to be negative, but I don't get it. To be honest I read every one of your posts, and always find them interesting. :thumbup:
 
Bluntcut, don't let me say the wrong thing here, correct me if I mis represent.

He suspects grain growth at the 1480F mark. He is after super super fine grain in 52100. In order to avoid the grain growth he suspects at 1480F, he would like to try a lower austenitizing temperature. But he realizes that the lower austenitizing temperature will not reach max hardness with the usual quenching medium, so he tries an accelerated quench medium. If this quench medium were to be used at the normal austenitizing temps, it would cause unnecessary strain and stress.

I "think" that is the underlying hypothesis. I may be all wet there, tho, but I think that is what he is after.
 
I derived & innovate my ht based mostly public research w/ SEM & TEM images. Along with a few private SEM pics. My results are mostly interpreted (tried to be non-bias) using combination of readily avail tools: triangulation, macro pics, sharpen & test, 400x usb microscope, hardness tester. Recently (after a long BO) I got Nital 3%, sadly can't resolve much with my 400x (~50x linear) usb.

Stuart, thanks for correctly rephrased my terse 'ballooned' statement. Here is a macro pic (where 804C = 1480F)
rHBQ9DL.jpg

Macro for 1440F look like 1480F. Ditto after nital etched. Martin/HeavyHanded generously offer to take 2K optical pics for me. I suppose to send out samples but decide to hold up until I complete this experiment. I told Martin if with good result, I will acquire a 2.5K optical, otherwise my leaky theory need serious re-examination.

Warren - thanks for sharing your experiment result. My 52100 samples hardness drop below 67rc when aust temp around 1450F(7 minutes soak), however got 67rc with 1440F with 20 minutes soak. Good reason for me to try 1440F & below with 10 minutes soak.

stezann - I hear you on the saying "don't fix if ain't broken". To spice ht life, I goes for the saying (Red Green Show) "if it's not broken, you are not trying". Could be that, I learn how & why things break, then fix, might register as skills. I forgot most of my theoretical math & comp sci. because lack of applied/context.

Matthew - Your open minded professionalism is well appreciated. You are right to assessed my tendency to draw conclusions and worse with on possible falsely assumptions. Am I suffering from a grandiose disease? OK, don't answer that, just give me med:p Now looking at my (sample of 2) 1440F 20 minutes soak 67rc and macro structure look like 1480F. I concluded that plenty of carbon in solution at that temp + soak, and possibly too much. So I will cut soak time to 10 minutes, I aim for AQ hardness be around 65rc. Which might indicate (again assumption/guessing) just barely sufficient carbon in solution. 1435F, so on would reveal -via interpolation- the amt of carbon in solution. 3 key benefits/reasons from this data provide:

1. I want non-solution carbon stay in particle/carbide/etc form to nucleate grain and at the same time provide more wear resistant.

2. Minimize carbon & other elements diffusion, avoid aus grain size regression from prepared state.

3. Higher percentage (95+%) of Lath martensite to gain extra toughness.

*. (edit added) obvious but worth considering. A 0.6%C martensite instantaneous goes into aust at temp say - 1440F for a minute or 2. Temporary precipitation & dissolution cancelled itself out on the ramp. It's reasonable to expect 0.6%C in solution, agree? Spheroidize structures (fine & large) are lower potential energy which require higher temperature to break them apart to release carbon. then, is 1475F optimal?

***
Catching Shrimp with Bare Hands book - if anyone curious and time permits, while on Amazon click 'Look Inside' and free preview up to 40 pages. It took Michelle over 8yrs (~4 fulltime years) to write it, time spent is not always equate to quality, prudently/thereby my preview recommendation. Thank you.
 
Last edited:
That's interesting that you had such high Rc#'s with the lower austentizing temps. How did you prepare the steel prior to heat treat? Different initial conditions could change the heat treat results quite significantly. You have me curious :thumbup:.

Just to be clear, I do not mean in any way to dissuade you from what you are doing. I try some out of the box things too, but I think you not only go out of the box, but to a different dimension :thumbup: :D. We need people to ask questions and test the results. Have you ever considered getting in touch with Kevin Cashen or Roman Landes regarding your steel? Kevin was quite helpful with me at a hammer in last year and let me know about the equipment I will need to look into the steel in more detail in the future, which I cannot afford this year. Might be worth an e-mail or two to see where it goes.

Warren
 
Last edited:
I view a blade as an energy system. The lower the potential energy the higher the temp/energy need to release particle bonding. With that, I prepared for a system of high potential energy small particles - smallest possible cluster of Fe3C and alloy form, plus high stress but organize lattices. I am aware of: spheroidized cementite dissolution curve, thus I avoid carbide in this form. So as soon as aust exceed curie temp, carbon from lattice(perhaps some precipitated, then quickly dissolve) is in solution right away, take a few minutes for particle dissolution. Ideally, I want localize Carbon diffusion and just enough particle dissolution - leaving the rest to pin/nucleate the grain. With this setup, Parks50 quench gives me mid's 50rc result, brine ~61rc, accelerated (super quench) 67rc. I suspect a pearlite nose/gap probably 1second or less. Quench crack appears when aust higher than ~1550F. Grain size(ok whatever bumps size on the macro view) at 1500F = BF std ht* with 1475F aust. I actually made a 15" 58.5rc bladesports chopper harden aust at 1500F. All this just pure garage experimentation without lab instruments to verify, hence fwiw.

* BF std/recc 52100 ht (norm, (therm - black)+, optional: fine spheroidize anneal, aust/harden 1475F 9secs oil). This preparation is loaded with fine spheroidized cementite before harden. A sphere carbide shape is strong but still larger than small aggregate. Small aggregate will have more surface interface with matrix and sharp edges. Fine spheroidize prep probably works better for UHC (1.4+%C) simple carbon steels, whence there be enough fine spheroidized carbide to nucleate grain.

Willie71 said:
That's interesting that you had such high Rc#'s with the lower austentizing temps. How did you prepare the steel prior to heat treat? Different initial conditions could change the heat treat results quite significantly. You have me curious :thumbup:.
Click to expand...

Thanks for your suggestion. Kevin & Roman probably have better thing to do than entertain wild ideas & proclamations from many metallurgist-wanna-be out there, myself included. Yeah, once Kevin setup his new SEM machine, hopefully he would enlighten a few mis-guided metal souls. At the least, wow us with 50K res images. I could re-rent a SEM or scatter SEM for a few more hours - spell pricy. Probably cheaper for me to grab a 2.5K optical microscope to examine at grain & carbide size+orientation+mis-alignment angle, beyond that is SEM/TEM realm.
Just to be clear, I do not mean in any way to dissuade you from what you are doing. I try some out of the box things too, but I think you not only go out of the box, but to a different dimension :thumbup: :D. We need people to ask questions and test the results. Have you ever considered getting in touch with Kevin Cashen or Roman Landes regarding your steel? Kevin was quite helpful with me at a hammer in last year and let me know about the equipment I will need to look into the steel in more detail in the future, which I cannot afford this year. Might be worth an e-mail or two to see where it goes.

Warren
Click to expand...
 
It didn't rain today so I ht-ed 5 knives according to temp in OP (gyuto: 1440F; paring: 1435F, 1430F, 1425F, 1420F with 10 minutes soak). Well, my most complex ht today, took me a whole long day and still tempering...

Pre-emptive info: TC in my Evenheat KO 18 is +/- 5F with Auber external TC probe. Auber TC is within 10F from 450F kitchen oven. With all that, I might buy another TC to triple check. I verified my hardness tester with test 61.5rc block.

After 15 minutes tempered 275F in kitchen oven. Hardness reading: 66rc@1420F, 65@1425F, 66@1430F, 66.5@1435F, 67@1440F.

Even for an out-of-the-box guy, I am in shock with the result. Look like I need to reproduce similar result for at least 2x more times.
 
Before repro my experiment. I'll check my ht oven temp using coupons aust around Curie point.

5 coupons (surface ground 1/8" thick, 1"x2" size, numbering = 0 to 4 rounded corners)

Normalize(fast): clay coat, normalize 1850F 5 minutes, air cool to black

Reset grain: 1520F 5 minutes, air cool to below 1000F => water dip => grind clean => hardness test (N_rc normalized hrc, expect less than 43rc)

Hardening: 10 minutes soak and parks50 quench. C0: 1390F; C1: 1404F; C2: 1414F; C3: 1424F; C4:1438F => grind off decarb => hardness test​

Hardened coupon rc should be greater than 43 when austenite=>martensite phrase taken place.

Have anyone done/know similar test for simple carbon and have a hardness curve around Curie point? thanks.
 
This thread probably doesn't offers much value to most. Nevertheless, here is my follow up.

I bought a 2500x microscope (1000x optical, 2.5x empty mag).

My oven TC is almost spot on with few degrees flux. 1420F =65rc; partially failed to harden at 1415F (some 58r spots, most 40s rc). Completely failed to harden at 1410F & 1405F (~40rc).

After a couple weeks played with the microscope. I don't have to poke in the dark as much ... until SEM becomes a 'need' :p

My latest experiment: SQ and hardness test (no temper). I broke off an attach coupon to a paring knife to see grain *as is/quenched* surface (no etch). 400x optical color negative, because of uneven surface the microscope only able to captured a thin strip in focus.
20150314_52100_K0_paring_break_off_02.jpg


With eyeballs & a 15x Peak Loupe, I can't discern individual grain. I have 7 coupons need to polish & etch - hopefully I can see grain boundary along easy to spot carbides.

ASTM grain size quick ref

5: 63.5um
10: 11.22um
14: 2.8um
15: 1.98um
16: 1.4um​

On average standard ht 52100 ASTM grain size is between 9 & 10.
 
If what you're shooting for is super fine grain maybe you should do a couple more refining heats. Maybe something like 1550, 1500, & 1450. Just a thought. I find your testing & experimentation interesting. I definitely give you an A+ for thinking outside the box.
 
Thanks Darrin! I have a Darrin-Sander section in my 52100 ht file (dated 9/24/2013). Thanks very much for being very generous sharing your ht recipe. Top section listed Stacy Apelt and then there are KevinCashen + others. Tips to new ht tinkerers should search & read Apelt+Sander+Cashen+(controversial EdFowler - keep open mind)+etc..

I went through many combinations normalize + 0-6 steps thermal cycling + 0-2 sub critical steps => Made way too many good experimental knives, always thinking the last batch was the one:o. And there were quite a few regression/going-backward knives & blades = into recycling bin. W/o a doubt, thermal cycling indeed will refine grain but don't push beyond point of diminishing or declining return.

darrin sanders said:
If what you're shooting for is super fine grain maybe you should do a couple more refining heats. Maybe something like 1550, 1500, & 1450. Just a thought. I find your testing & experimentation interesting. I definitely give you an A+ for thinking outside the box.
Click to expand...

If you 3x zoom in the 400x image above, you see interconnecting reddish web. They look like grain boundaries with average size less than 2um, hence ASTM grain ~15+. Given my current grain size assessment/read is correct. It will be very difficult achieve sub-micron grain size with 0.94%C 52100. A couple weeks ago, I bought a bunch more of 52100 from Aldo. Now, I need to find some of Carpenters' 1.05%C 52100. Blue#2 could be a decent but expensive substitute.

From my last weekend 20150308 experiment - most complex ht (a whole long day worth of work) with various normalize + 3 to 5 thermal cycle and some with sub critical steps. Micrograph below belong to best or 2nd best sample. I can see individual grain from break off with a 15x Peak Loupe. Grain size around 4-5um (ASTM grain between 12-13) is very good but ... It took a few days for me to swallow the pain&pride from a few months of going backward tried to reconfigure carbide w/o a microscope.
sq52100_1Kx_20150228_02.jpg
 
I agree that an ASTM grain size of ~15 is excellent, especially with an AQ hardness of 66-67. I may be wrong but I think that if you were able shrink the grain size much more then the AQ hardness may drop.
I completely understand and agree with your statement about balancing the point between diminishing returns and performance.
 
My regressed ht produced numerous large carbides & larger grain.
sq52100_1Kx_20150228_02.jpg


*********
Below is my corrected & improved ht 52100. fyi - those tiny almost spherical shape/bump are carbides.
Micrograph of the break-off coupon in prev post. Almost no carbide > 1um in size, most are around 0.25 - 0.4 um in size. At grain size 15+, I would need at least SEM 3K resolution to see/trace grain boundaries.
20150314_sq52100_K0_03.jpg
 
I believe you spoke from experience when heat-treating with extra fine grain steels. Yes, AQ hrc will decline for 2 main reasons (OK, conjecture & experimental-data by me). 1st. pearlite nose gap is probably sub 0.6 second - brine is not fast enough, hence I use super quench. 2nd. nano size carbides & homogeneous grain yield additional strength & toughness but doesn't reflect/register correctly/meaninfully against hardness tester diamond indenter - it flows up(crater ring around) the indenter more than compressive displacement.

3rd is about rapid loss 2-4rc when temperature > 225F, due to low energy (thermal) require for carbide precipitation and some misc conversion. The paring knife where the coupon above taken from - tempered 275F, current hardness is 63rc. Actual AQ hardness for this knife after LN2 dip was 66rc; however stated 65rc let's me buffer 1rc. This knife edge rolls during hard use tests (i.e. chop into seasoned oak)... Doh, over tempered for a kitchen knife - was targeted for microchip hardness.

darrin sanders said:
I agree that an ASTM grain size of ~15 is excellent, especially with an AQ hardness of 66-67. I may be wrong but I think that if you were able shrink the grain size much more then the AQ hardness may drop.
I completely understand and agree with your statement about balancing the point between diminishing returns and performance.
Click to expand...
 
I'm still working through these images but this is my 52100 at 1500° F. I posted this in another forum which some of you may have already seen; couldn't hurt to show it here either. Prior thermal cycles included 1650°, 1500°, and 1400°. 15 minute soaks for all.

jgLuR14.jpg
 
Don - Thanks for sharing:thumbup: I (Luong) replied to your thread on HF - http://www.hypefreeblades.com/forum/viewtopic.php?f=3&p=7095#p7095

Your x5.00K pic (http://i.imgur.com/b7objEr.jpg) clearly shows grain boundaries. Look like(eyeballing) it has aus-grain dia between 5-8um (astm size 12-11). RA 15-20%, ~3% martensite, bainite balance. Large aggregated cementites taken placed at grain boundary. So on... I envy SEM/TEM.

Don Nguyen said:
I'm still working through these images but this is my 52100 at 1500° F. I posted this in another forum which some of you may have already seen; couldn't hurt to show it here either. Prior thermal cycles included 1650°, 1500°, and 1400°. 15 minute soaks for all.
Click to expand...
 
Mr La, may I ask how you get the pearlite nose of 52100 to be sub second, requiring that faster quench medium? I think I know but was hoping to get more info from you. Something that caught my eye....your sequence of 1850F, 1520F, then austenitizing....1850F seems awfully high, and with only one thermal cycle (1520f) between that 1850F and austenitizing temp (1430F or what have you), the grains would be larger than wanted?

Really curious about the 0.6 second PN and how you achieve this! This is VERY interesting for sure. I wish we all had access to the equipment needed to do these cool experiments! Keep it coming!
 
You must log in or register to reply here.
Back
Top