Heat Treatment - Crystal Weaving Foundation

Published cwf ht for aebl 64rc is tougher than std ht at 64rc. It ripples vs instant brittle chip of std ht.

For general users (i.e. not among us knuts) - I usually leave edge geometry as 0.012"BET, 15-18dps (depend on knife type), sharpen with 2K grit finish and then dull/rounded back a little bit to reduce apex width to around 2-4um. Aiming for long last working apex about 4-5um wide, also able to withstand ceramic(~1K)&dia(220) sharpening rod; most people smack back/forth at about 25dps, so all the apex has to endure is the first time that micro-bevel get created.

Extra ductility would be nice, so

smu2VEw.jpg

^ if you look super close, something aren't normal about those test surfaces.

I successfully replicated for aebl & 3V and got elmax+zwear+M4 with 80% level of toughness. Gained: toughness. Loss: Some wear resistant. Without carbide refinement D2 at high rc remain very chippy due to large dia carbides.

This new technique is one-way ticket ht - can't re ht at all. So for those gyuto&petty above, planned to make into knives, I only lightly applied this new technique.
 
Chào Quân,

Có thử ht 4V nhiều lần rồi.

Please gmail your phone# to flexdog2008 - we can talk and avoid mix-language posts in thread.

==Lượng

nguyenquan said:
Em chào anh !

Em là Quân ở Việt Nam trước cũng từng tham gia trại lính cùng anh. Anh ạ anh đã thử mác CPM 4V chưa ạ ?, em đang vướng mắc với món CPM 4V về cách nhiệt luyện, anh có thể trao đổi cùng em chút được không ạ.

Em xin cảm ơn Anh trước ạ !
Mong hồi âm của Anh.
Click to expand...
 
Hi Mr Lương

Cháu đã email tới chú ( Email của Chú có phải là flexdog2008@gmail.com không ạ ? ) Cháu mong chú sớm hồi âm ạ

Cháu xin cảm ơn Chú :)
 
There is an area 74 ft x 74 ft filled with loose 5476 (74*74) tiles sitting next to each other. You want to concrete them together, so each tile now 13'' x 13" (title + concrete). Given 10 hard working robots, which each will cement 5 consecutive titles together at the time. Robot can push loose titles to expand but no gap is allow because the robot wheels will get catch & it falls.

Hey boss, which way should we/robot proceed?

1st. Random independently each robot can cement which/where with requires least push/expand force (but can't separate cemented titles)

2nd. All 10 robots cementing from outside in (onion ring)

3rd. All 10 robots cementing from inside out (start in the middle then work outward)

Instantly - you want a set of smarter robots because why asked the obvious - 3rd way is a no brainer!

** Jump to the world of HT **
https://en.wikipedia.org/wiki/Atomic_packing_factor

Packing factor of austenite/fcc is 0.74, while ferrite&martensite(bcc&bct) PF is best around 0.68. (ah now see how this connected to 74x74 and cemented tile increase size to 13" - from only 68 cemented tiles can fits in 74ft)

Default ht/metallurgy robots matrix transformation is mostly using the 1st combine with 2nd ways.

Boss - what were you thinking ^ about us/robots? At any rate, us/robots really want to use the 3rd way (or partially of that route), please tell us how?

How the heck, can you make transformation from inside-out when thermomechanical taking place from outside-in (think heating & cooling - always from outside in).

** CWF HT V1.x **
Differential change one or more elements composition mass% and partition Cr and more <= maybe this or not or that or just a big steaming pie :foot:
 
Last edited:
bluntcut said:
There is an area 74 ft x 74 ft filled with loose 5476 (74*74) tiles sitting next to each other. You want to concrete them together, so each tile now 13'' x 13" (title + concrete). Given 10 hard working robots, which each will cement 5 consecutive titles together at the time. Robot can push loose titles to expand but no gap is allow because the robot wheels will get catch & it falls.

Hey boss, which way should we/robot proceed?

1st. Random independently each robot can cement which/where with requires least push/expand force (but can't separate cemented titles)

2nd. All 10 robots cementing from outside in (onion ring)

3rd. All 10 robots cementing from inside out (start in the middle then work outward)

Instantly - you want a set of smarter robots because why asked the obvious - 3rd way is a no brainer!

** Jump to the world of HT **
https://en.wikipedia.org/wiki/Atomic_packing_factor

Packing factor of austenite/fcc is 0.74, while ferrite&martensite(bcc&bct) PF is best around 0.68. (ah now see how this connected to 74x74 and cemented tile increase size to 13" - from only 68 cemented tiles can fits in 74ft)

Default ht/metallurgy robots matrix transformation is mostly using the 1st combine with 2nd ways.

Boss - what were you thinking ^ about us/robots? At any rate, us/robots really want to use the 3rd way (or partially of that route), please tell us how?

How the heck, can you make transformation from inside-out when thermomechanical taking place from outside-in (think heating & cooling - always from outside in).

** CWF HT V1.x **
Differential change one or more elements composition mass% and partition Cr and more <= maybe this or not or that or just a big steaming pie :foot:
Click to expand...
If I follow you, you are suggesting that fiddling with alloying elements may make it possible to begin the martensite transformation from the center and proceed outward to the surface? How does this relate to the super slow cooling aspect of CWF?
 
Weaving (slow cooling) works in single grain - so best scenario in this case is the tile (a grain) boundary grew minimal & uniformly to 13" x 13". Heavy twining of martensite across grain will distort that boundary, thus tile would turn into a weird shape and most likely be larger in dimension.

To promote transformation core to surface - My idea is to distribute element with gradient intensity - core2surface <=> high2low. Partitioning (captured) certain aust stabilizing element to amplify the transformation gradient. If cool & heat from core2surf then it would has strongest response for core2surf transformation <= not easy to rig up something to do this.

There are additional benefits core2surf gradient element distribution ...

Challenge - precise temp & time per work piece thickness. Due to thickness is highly varying in a blade (thanks grind bevel).
Downside - one way ticket ht, thus very expensive to conduct experiments (even with some formulaic elem diffusion rate).

^ all this is still within the realm of thermomechanical transformation. My other versions added variables outside of that realm.

Bill1170 said:
If I follow you, you are suggesting that fiddling with alloying elements may make it possible to begin the martensite transformation from the center and proceed outward to the surface? How does this relate to the super slow cooling aspect of CWF?
Click to expand...

edit to add: mapping tile analogy to a blade 3mm thick x 20mm tall and 10um grain dia, would equivalent to a wall of cubic titles: 300 across by 2000 tiles tall and whatever Length * 1000/10.
I have pointed out in many occasion about thermal mass issue - basically, dislocation/stress higher in core than surface. Which could lead to more intergrain fracture and worse case a global structure failure (blade snap in half).
 
Last edited:
A chain is only as strong as its weakest link

In ht optimization/problem-solving, I

1st. weaving - to shore up weakness inside a grain => next

2nd. transformation sequencing - to shore up weakness global intergrain => next

3rd. carbide volume gradient - increase impact load & elastic range => next

4th. grain refinement - added strength and some toughness => but

5th. regain impact toughness by direct/align impact diffusion to along the blade rather than edge to spine

How far will I get by doing this solo? As far as that bump between my eyes :p
 
I hear you on the weakest link thing. The ancient Japanese swordsmiths beat the brittleness problem by marrying a hard edge to a softer core/back to gain edge holding without losing all impact toughness. This approach reminds me of what I think you were driving at in suggesting compositional changes to affect hardening response. I may have misunderstood your statement, but sometimes the conversation is helpful in sparking fresh ideas.
 
Yes, San-mai (& similar multi-layering) technique indeed helped with impact toughness (when doing it right). These technique involves forge welding 2 or more of steels with large different in composition - e.g. white#1 core, mild steel or iron outer sandwich layer. Problems: 1. weld affected area/interface is weaker than core and shell 2. potential large hardness different between core & shell 3. core transformation is still stuck with default (random + outside-in) 4. delamination

My idea ^ calls for gradient/differential element distribution to promote more transformation from inside-out. Envision gradient akin to rain & roof-top model.

Beyond improve crystal formation sequence, is gradient in carbide volume (lower = higher toughness). I think people would be excited to get something like these:

1. M390 core with gradual/gradient outward morph into sort of 12c27Mod ... into sort of 420J.
2. CPM M4 core ... sort of cruwear ... sort of 3V ... sort of 1050

'Sort of' because diff in undissolved primary carbides & other elem composition%

edit to add: Simple thermal sequencing + cv gradient large M4 core to 3V shell

dUy2V75.jpg


Bill1170 said:
I hear you on the weakest link thing. The ancient Japanese swordsmiths beat the brittleness problem by marrying a hard edge to a softer core/back to gain edge holding without losing all impact toughness. This approach reminds me of what I think you were driving at in suggesting compositional changes to affect hardening response. I may have misunderstood your statement, but sometimes the conversation is helpful in sparking fresh ideas.
Click to expand...
 
Last edited:
So you're talking about changing the actual molecular structure of the steel. Where certain molecules gather in a specific place and others don't leaving, in essence, a San mai steel made of a single piece of one certain steel. I can envision it but the technicalities are far over my head. If possible then you'd be doing something no else is on a very high level.
 
Sure, you could think it as a san mai from a mono/single steel. However san mai is about sandwich/layer, whilst I try to diffuse certain element with gradient distribution. Maybe clearer if you think in term of painting stripe/band of grey color vs gradient/gradual grey shift from black-to-white. I also try partition Cr to improve grainular toughness - I am not certain whether partition is a net gain process or not.

bodog said:
So you're talking about changing the actual molecular structure of the steel. Where certain molecules gather in a specific place and others don't leaving, in essence, a San mai steel made of a single piece of one certain steel. I can envision it but the technicalities are far over my head. If possible then you'd be doing something no else is on a very high level.
Click to expand...

All this stuff sound like mumbo jumbo. Oh well :yawn:
 
It sounds good to me. In essence, the gradient thing would mimic how nature engineers living structures. The strength and toughness are apportioned according to where they are needed most. Consider the mix of collagen and elastin in our skin, or the varied properties in different spider silks from the same spider.

Case hardening is a crude example of using a gradient of carbon content to effect differential properties in an otherwise mono steel. Being outside-in, the effect is mostly very close to the surface, while I get the sense Bluntcut is going for a gradient all the way through the material, so the hardening proceeds like his analogy of tile being laid from the center of the room. So my question is: how can the alloying gradient be established? Continually variable spray forming? Layered powder metallurgy? Something like a dipped candle?
 
Probably ms temperature is higher in the core then on the out side. So transformation will start from the inside out.

Sent from my SM-G900F using Tapatalk
 
Bill1170 said:
It sounds good to me. In essence, the gradient thing would mimic how nature engineers living structures. The strength and toughness are apportioned according to where they are needed most. Consider the mix of collagen and elastin in our skin, or the varied properties in different spider silks from the same spider.

Case hardening is a crude example of using a gradient of carbon content to effect differential properties in an otherwise mono steel. Being outside-in, the effect is mostly very close to the surface, while I get the sense Bluntcut is going for a gradient all the way through the material, so the hardening proceeds like his analogy of tile being laid from the center of the room. So my question is: how can the alloying gradient be established? Continually variable spray forming? Layered powder metallurgy? Something like a dipped candle?
Click to expand...

Yeah, case hardening seems more similar than San mai. Case hardening, but in reverse, and where more than just carbon gets moved.
 
I guess (extra wild kind) - adaptive spray form with variable composition to thin-filming. Also quite certain, steel producers;metallurgists and others are already thought about this and more. Cost could make this and other expensive methods uneconomical/impractical.

For inside-out transformation:
My attempts are mostly involved - thermal sequencing (for diffusion & partitioning) in conjunction a non-thermomechanical force. Looking at atomic radius - Carbon can zip around fast, Cr (and elem next to it in table) aren't ready to move about much unless very energetic and or path ways in aust matrix are wider. Particles/carbides aren't/won't move at all, unless dissolved them first :)

Bill1170 said:
It sounds good to me. In essence, the gradient thing would mimic how nature engineers living structures. The strength and toughness are apportioned according to where they are needed most. Consider the mix of collagen and elastin in our skin, or the varied properties in different spider silks from the same spider.

Case hardening is a crude example of using a gradient of carbon content to effect differential properties in an otherwise mono steel. Being outside-in, the effect is mostly very close to the surface, while I get the sense Bluntcut is going for a gradient all the way through the material, so the hardening proceeds like his analogy of tile being laid from the center of the room. So my question is: how can the alloying gradient be established? Continually variable spray forming? Layered powder metallurgy? Something like a dipped candle?
Click to expand...

Good thought/insight, Oren:thumbup:

Best bet for this to taken place like this - would be rapid core cooling. Which need to be faster than Sum(Ms delta + deactivate exothermic). Gradient carbon in solution (higher near surface - super saturated, lower core - eutectic point) might back-fire with massive plate martensitic matrix near surface, thus end up with higher brittleness than core.

Please post more and often...

orenj said:
Probably ms temperature is higher in the core then on the out side. So transformation will start from the inside out.

Sent from my SM-G900F using Tapatalk
Click to expand...
 
CPM M4 0.150" thick, 10.5" blade, 15.6" OAL

CWF HT 1.3 - 64rc targeted for strong and extra tough

Stabilized Curly Mango Handle

Chops: 2x4, oak dowel, beef rib bone, pork thigh bone, 16d nail, red-brick.

Thanks for watching & comment.

15 minutes
[video=youtube_share;px730woigh0]http://youtu.be/px730woigh0[/video]
 
Last edited:
ok, I snap a pic to show post-test edge (I washed with simple green and ~10 rub with 600 grit AlO sand paper to clean up)

PPKZR7w.jpg


btw - today, I made 3 chops to 16d nail with a cwf ht 1.0 64rc D2 chopper = 3 chips. Largest chip ~0.26" wide x 0.07% tall and broken piece is still embedded in nail. Also did one 16d nail chop with cwf ht 1.0 61rc d2 = a nail shape dent ~0.06" tall. I expect cwf ht 1.3+ d2 performs better ... it will be awhile before I make another 3/16" d2 chopper.
 
Last edited:
bluntcut said:
ok, I snap a pic to show post-test edge (I washed with simple green and ~10 rub with 600 grit AlO sand paper to clean up)

PPKZR7w.jpg


btw - today, I made 3 chops to 16d nail with a cwf ht 1.0 64rc D2 chopper = 3 chips. Largest chip ~0.26" wide x 0.07% tall and broken piece is still embedded in nail. Also did one 16d nail chop with cwf ht 1.0 61rc d2 = a nail shape dent ~0.06" tall. I expect cwf ht 1.3+ d2 performs better ... it will be awhile before I make another 3/16" d2 chopper.
Click to expand...
It is absolutely nothing. this small damage can be fixed in no time.
Really amazing.

Sent from my SM-G900F using Tapatalk
 
Just wanted to give a short update. I won one of these in the GAW bluntcut did, again thank you. I'm not a metallurgist, so a lot of the intricacies of heat treat go right over my head. All I can say is that the knife I received is crafted very well, and it came ridiculously sharp. I've put it through moderate use, warehouse type chores, kitchen duty, things of that nature. No edge chips, edge hasn't rolled, and its got a decent edge after a month of good use. Between my mediocre ability to sharpen and thr hardness of the steel I struggle to sharpen it, but I have the same issue with d2, so operator error, not a blade problem.

Can't speak on bluntcut's heat treat method besides that I received a knife he made and its really sweet, and he's a friendly guy to-boot!

Thanks again bluntcut!
 
You must log in or register to reply here.
Back
Top