Heat Treatment - Crystal Weaving Foundation

I trained by a Material Scientist on the SEM machine. Did captured SEM, Hires SEM, EBSD images for O1, W2, 26C3. Some up to 80K mag. EDX/EDS (x-ray) is time consuming to run - 20 to 45 minutes per map. Samples were polished to 0.25um diamond powder, etched with nital 5% for 50 seconds.

These images are still being analyze. Thus far, (imo) results matched ht 3.0 prediction well.

I plan to schedule 4 hrs on SEM instrument for imaging 36 hrs version of ht 3.0 on O1, 1095, W2, 52100, T15, .. in mid July.

When using external/3rd party metallography services $250-275/hr is reasonable (common/standard). EBSD and EDS often cost extra$.

Any plans for polished and etched SEM Micrographs?

I'd love to see some 52100 btw.

That Lisin Metallurgical place has SEM.

But they charge $250 per hour to run.

Yikes
 
I will be cutting SEM coupons on/off for the next few days: 1095, W2, 52100, O1, 26C3, 10V, 204P, CPM T-15. Follow by ht control/baseline using std/popular protocol and BCMW HT3 and ht3+new ideas.

@DeadboxHero, Hey Shawn - If you make SEM coupons (5-10mm x 10-15mm ~2-3mm thick) for 52100, W2, 1095, .. with your or std ht. Untemperred + decarb-free + polished to 0.5um. And one edge has fractured surface <= I cut & ht a 10mm x 60mm stick with scored (partially 6mm cut one end at 10mm) and after polished, I break off the 10mm x 10mm piece = SEM coupon. So fractured surface is around 4mm long. I will etch and take SEM images your coupons along with my pile of coupons. Term: Images of your coupon are BCMW's IP however you've full access.
 
Here is a comparison of carbon distribution/density(via EDS/EDX x-ray) between W2 (from njsb 2019) vs O1 (from AKS)

W2
9zAJ6QO.png


O1
hMkJd3U.png


W2 was normalized at 1900F for 2 hrs and obviously some spheroidized cementite didn't dissolved. Carbon% listed (by vendors) for O1 and W2 are about ~0.95%. Density; granularity; particle uniformity are ideal conditions. EDS images show surface elemental mapping of 1-2 micron depth.
 
Edge Stability Test

CPM S90V, 64.5-65rc HT 3.0, 15dps, 0.008" behind edge thick

Test cutting materials: Paper, Sisal Rope, Rosewood, Zip tie, Chicken leg bone, skinning - 8 gauge outdoor electrical cable.


I've been experimenting with HT 4.0, where 4.0 adds grain refinement step. High alloy steels will (projected/supposed) benefit more than low alloy steels. 1095, 440c, 52100, w2, 26c3 exhibited finer grain comparing against control/baseline samples. Results of latest ht experiment with T15, 10v, 15v, 204p are currently being test. Each experiment include ht with std/industry params as control/baseline, ht 3.0, ht 4.0.

Test ht 4.0 blades in 10v, 15v, t15 end up with hardness ~65rc, where maximum grain refinement ht params were used. All samples have same temperature for austenite and tempering. ht 4.0 of 204p coupon and blade failed my minimum hardness criteria. Exception: control coupons with no-cryo and employed high temper temperature (975F).
 
Carbides wear resistance performance is highly correlated to strength of their binding interface with the matrix. In other word, how strongly the matrix hold-fast onto surface protruding carbides against external dislodging force.

Standard/industry vs BCMW HT

* 32 minutes video *


A summary micrograph
ec5X8sH.png
 
HT 4.0 encountered fluctuation/variation of results for 1095, mostly due to super quench timing. This test chopper and stubby blade have near optimal quench timing, thereby test results should provide good qualitative performance for them.

1095 (1.0%C, 0.2%Cr, 0.45%Mn, 0.05%Mo,0.02%Ni) 1/8" thick

65rc HT 4.0

Chopper: 10" blade, 15.9 oz, 16 dps, 0.015" behind edge thick
Stubby: 18dps, 0.016" behind edge thick

Test materials: Wood, bone, 16D nail.

20m:39s video
 
I need a distraction from ht too many rounds of coupon for SEM and yes another one is in progress... time to parallel a productive(hopefully) procrastination.

For a few years now, I wanted to test edge shock resistance/stability by cutting 6.4mm dia nail (60D) and 8mm diameter mild steel rod. Somewhat relevant and fun ... Frank Richtig metal cutting demo looks cool however I am skeptical ... umm I meant - good to have a friendly setup - image below. With a similar setup, initially/pilot a few cut attempts with ht 3.0 blades. If result is promising, will test ht 4.0 for these blades - .180 to .210" thick, 9" edge, 1" saber height: *control/baseline, 60-63rc* [8670, A8Mod, 80CrV2], *HT 4.0, 62-64rc* [O1, 52100, 3V, D2]. Will adjust edge thickness depends edges performance.



t0jQhB2.jpg
 
You are right - 45rc (~cold chisel, 25dps) is ideal for cutting (mainly hot) metal. However at that hardness, the edge will not pass printer paper slicing test after cut 6-8mm dia mild steel rod.

Frank Richtig knives - From published research paper, edge hrc around 56. I tested one knife, it was around 57. I am guessing, in order for afterward edge to pass printer paper slicing, edge hardness need to be at higher strength. Of course, demo practice and experience are big assisting factors to pull off this trick (ok, test). First attempt will be using a broken W2 HT3 63rc.

From what I remember of Richtig, you may need to tweak the HT a bit. Maybe go from target hrc 65 to hrc 45?:p
 
With W2 HT3 63.5-64rc edge, I did 6-8 attempts, got one to cut through 8mm diameter mild steel rod. My hammer striking skills is lousy, so will practice some more and lower hrc of this practice blade to 62-63rc, then hopefully got a few successful cuts. Also will fix up the stubby 1095 ht 4.0(prev post) + give 400F tempered (est hrc will be 62.5-63rc), then give it a few attempts.

8qDgtPC.jpg


edit: nah, I will keep w2 & 1095 at their current hardness and try a different strike technique. The pic above showed the apex is smushed/crushed-in, so increase ductility would be counter productive.
 
Last edited:
Edges after best cut of W2 63.5rc and 1095 65rc look promising. It seem 63.5rc and 65rc aren't strong enough to avoid flattened apex. Even if increase bevel to 25dps, sharp/keen apex would still smush/flatten to similar apex radius below, which won't pass printer paper slicing test. Demo person often quick with hands (trick/cheat) skid pass the flattened and cut paper with unaffected sharp edge (such as the upper part of the edges in pic below).

PBgqDAJ.jpg


45 minutes data video captured attempt 10th onward (didn't track, maybe last count to 21st or 22nd). The last 2 of 3 attempts yielded best results (pic above), where entry cut started on anvil face then finish on the anvil horn.


edit: took a higher magnification closeup of afterward edges
jD2ZKc8.jpg



Edit 8/13:

Today I made over 15 attempts using Cfv 63.5-64rc HT 4.0 and with bevel angle 18dps, 20dps, 25dps and BET 0.018", 0.025", 0.021". Pic below is the last and best outcome. *Rod size comparison/perpsective - Area/cross section of 8.1mm is 3.72 times over 4.2mm diameter 16D nail.
WuOVLN8.jpg
 
Last edited:
A while back, I bought my first and last Randall knife. It was a used knife in stainless steel, but when I used it, I could tell immediately that it's heat treat was poor. It bent like taffy. Being worthless to me at that point, I did some destructive testing to get to know its limits.
2v2EGTHcyxAWtWs.jpg


After seeing my knife after destructive testing, Luong ( BluntCut MetalWorks BluntCut MetalWorks ) generously made me a new knife in PSF 27, which is D2 steel processed in a way superior to ingot steel. The spray-form processing gives better distribution and size to the carbides, making PSF 27 more wear resistant and tougher (resistant to chipping) than ingot D2.
2v2EGTH68xAWtWs.jpg


The knife is beautiful in form and function. Luong ran the heat treat up to 64 Rc and ground the blade into a high-performance geometry. The edge is 15 thousandths along the flat and 6 thousandths along the belly, where the blade is used to penetrate. In my opinion, this is the proper way to grind the edge. The blade has a nice distal taper, from 0.125 inches at the base of the spine and 0.03 inches a quarter inch back from the tip. This is a slicer that can penetrate with ease. Luong even took the time to round the spine to make the thumb positioning more comfortable.

I've been using it for a while, and its performance is killer.

The knife's design is simple, elegant and classic: The blade is pointy, the belly is gradual. There is a highly functional sharping notch. The handle is rounded, comfortable and fills the hand, with just the right amount of curvature fore and aft to prevent your grip from slipping in either direction. It does not want to twist in the hand. There is no jimping because none is needed. I love this knife.

But I promised Luong that I'd test it properly to its limits. I didn't want to do any destructive testing because I like and use the knife so much. But I also wanted to carefully test its limits.

In normal use, the knife is nimble. It cuts through normal material with ease.

To test the edge stability (resistance to chipping/cracking and denting/rolling), I batonned through an oak dowel. This is not a super tough test, but it did dent a 1075 chopper I have. The PSF 27 took no damage.
2v2EGTHa5xAWtWs.jpg

Next was chopping through 12 gauge electrical cord.
2v2EGTHQjxAWtWs.jpg

The apex suffered a tiny bit of micro-chipping, which you can see in the photo below.

I did both of these tests with the only undamaged portion of the Randall left. Its edge was a much more robust 27 thousandths, and it took only a tiny bit of micro-rolling, which you can see in the photo below, marginally less than the PSF 27 edge. Its cutting performance was far inferior to Luong's knife because of its thicker geometry.

Finally, came the acid test, which no knife I've tested has passed: Cutting through a tough piece of bailing wire. I hammered each knife equally until I could detect damage -- and no further. Neither blade would cut through the wire. The damage was about equal with the same force on batonning, but remember, the Randall had a much thicker edge and was softer, so the odds were heavily stacked in its favor.
2v2EGTHbnxAWtWs.jpg


The Randall dented (dent at left of center on edge). You can also see some micro-rolling from the copper wire test.
2v2EGTHw5xAWtWs.jpg

The PSF 27 chipped, with some micro-chipping from the copper wire test to the right:
2v2EGTHKqxAWtWs.jpg


The aggressive geometry on the PSF 27 blade was not designed for this kind of abuse, but I did find its limits without causing serious damage. I won't be abusing this knife again. It's such a joy to use. I've never tested a knife with this aggressive of geometry that took so little damage from the bailing wire test. Pretty cool.

Awesome knife with classic design, aggressive blade geometry, comfortable handle and excellent edge stability.

And a really nice heat treat.

Thanks, Luong.
 
T Twindog - thanks much for an excellent details write up with great pics on testing your 64rc(ht 3.0) psf27 knife. Cutting a wire or nail on soft wood backing is much more challenging than harder backing such as aluminum block or iron anvil or super high janka hardwoods. I will test the twin psf27 blade against electrical cord to see when & why micro chipping - did you baton or chop the cord?

I am to hear you are happy with the psf27 knife. I found ht 2 & 3 psf27 and d2 performs very well in normal tasks around 64rc hardness.
 
T Twindog - thanks much for an excellent details write up with great pics on testing your 64rc(ht 3.0) psf27 knife. Cutting a wire or nail on soft wood backing is much more challenging than harder backing such as aluminum block or iron anvil or super high janka hardwoods. I will test the twin psf27 blade against electrical cord to see when & why micro chipping - did you baton or chop the cord?

I am to hear you are happy with the psf27 knife. I found ht 2 & 3 psf27 and d2 performs very well in normal tasks around 64rc hardness.

Yes, Luong, it is a great knife. I did baton the blade through the heavy electrical cord, but it cut so cleanly that I didn't have to. It was just a couple light taps with a thin piece of pine.
 
I don't have a 12 gauge electrical cord to replicate your test, at any rate did some baton-cut 14, 10 and 8 gauge electrical cord. This blade supposed to be the twin of your blade however this edge would ripple quite a bit before chip out, oh well, clearly it isn't identical twin :D

fPcHYk4.jpg


Yes, Luong, it is a great knife. I did baton the blade through the heavy electrical cord, but it cut so cleanly that I didn't have to. It was just a couple light taps with a thin piece of pine.

Condensed HT 4.0 proof of concept with main goal of producing good toughness at 63rc

Steel 440C ESR (bought from https://www.alphaknifesupply.com/shop/440c-stainless-steel#info)

Hardening temperature and plate + cryo quenches : peak hardness
1600F : 60rc
1800F : 63rc (My guessed/projected temperature for 63rc, so beside a coupon, there is a test blade also)
1850F: 63.5rc
1900F: 64rc

CTS 204P 1600F : 57rc

With this data, floor/minimum hardening/austenite temperature is around 1750F for steels with more than 15% of Chromium. 1825-1850F is a good range for ht 4.0

*Note - conventional ht uses listed aust temperature for these steels will yield 2-25 rc lower hardness than my HT*

edit 22:18 - 62rc 440c test blade able to cut through 16D nail and 8mm rod only suffered minor rolls with edge 20 dps, 0.020" BET. Too bad, due my error the blade loss 1rc (was at 63rc).
Coupon break test: 60rc, 63.5rc were hard to break. 64rc moderate hard. 63rc splitted too easy. All coupons have notches cut by band saw (pre-ht), so stress riser does negatively affected break strenght/resistance.
 
Last edited:
A preliminary round between A2 Jungle knife/chopper vs O1 Bolax/bolo. Other choppers (diff steels) are there for comparison context.

A2 63.5rc HT 4.0, 18dps
52100 63+rc HT 3.0, 18dps
D6 65.5rc HT 3.0, 18dps
1095 65rc HT 4.0, 16dps
O1 65rc HT 4.0, 18dps


8dAu6c4.jpg
 
Back
Top