BCMW's ht chopping impact tests

[BluntCut MetalWorks]

D2 (was 64rc) 63.5rc@475F. Reprofiled back to 0.015"BET/15dps. Passed all tests with intentional mis-hits.

Aebl 61.75rc@350F (5 readings) 0.017"BET/15dps. Done twice - edge rippled when baton AB. Quite low macro hardness and micro hardness isn't that far behind macro but sub micron hardness must be quite low. Definitely a case of edge steered by nano weakness - yes, start small but lead to ripple up to 0.035" height (up from apex). This blade is perfectly fine for normal hard usage. I like it able to withstand abby-normal usage

Ht-ing 4x paring size D2 .110" thick blades with slight ht param deviation per each blade. Target testing hrc at ~65rc.

 

[BluntCut MetalWorks]

Heat treated

0.110" thick D2 working hardness: 3 blades at 65rc, 1 blade at 64.5rc.

Baton AB tested 2 blades (65 & 64.5rc) 0.015"BET/15dps - both consider passed with minor microchips (visible with 15+x loupe).

Once all 4 blades are tested and examined nital-etched surface ... I will pick best ht params, then tentatively, I will go directly to making a test JB in 3/16" thick D2 (63-4rc) <= hopefully, there won't be a thermal mass problem like W2&52100.

 

[Chris "Anagarika"]

Luong,

It just dawned on me that if you put 'buffer' in the geometry like Chiral suggestion, it'll be same as other makers, i.e. excellent knives at custom level but not much differentiation.. I'm guessing here: BCMW aims for no buffer hard use. That's why you keep pushing the limits and variables to find that matrix knitting HT. :thumbup:

 

[BluntCut MetalWorks]

Thanks, Chris!

Chiral's suggestion 15/20micro geometry is actually quite aggressive/thin compare to other makers choppers. Cutting wedge volume strength is much lower than a flat 20dps bevel. Hardness need to be > 62rc to support such geometry in hard use, otherwise some mode of failure will occurs. Basically it's a 15dps with 20dps to give additional lateral support.

99.99% cutting/chopping involved one-two physics: fracture + penetration. Take less energy to fracture with thin apex radius and also clean cut. Slender cutting edge (wedge) penetrate more efficiently due to lower wedging coeff. (however binding in certain material can be a big disadvantage) <= aaah but when you've an option, it's easy to tune it per intended tasks.

My D2 jb initial edge geometry probably be 0.02"BET/18dps.

btw - a quick glance at 64.5rc D2 blade (barely polished and no-etch) under 400x optic - wow, what a jungle of ~1um dia carbides :thumbup:.

Luong,

It just dawned on me that if you put 'buffer' in the geometry like Chiral suggestion, it'll be same as other makers, i.e. excellent knives at custom level but not much differentiation.. I'm guessing here: BCMW aims for no buffer hard use. That's why you keep pushing the limits and variables to find that matrix knitting HT. :thumbup:

 

[Chris "Anagarika"]

No doubt it's thinner. I'm just thinking on the common/usual engineering approach of 'beefing up' to avoid failure in making stuff sometimes can go too far

Keep us informed on your progress!

 

[BluntCut MetalWorks]

I just chopped - as hard as I can with 2 small D2 above (64.5 & 65rc) - to a 3/8" dia steel rod. Their edge dent & rolled -> According to my hardness calculation: 65rc consists of ~61rc matrix & 4rc from carbides. Basically, quite of ductility remain from possibly a max pure matrix hrc 67rc. Which mean loss 6rc to ductility. Due to D2 high alloying element% in composition, extra high dislocation is unavoidable. I think, 66-66.5rc D2 is probably in max hardness working range (cut & chop)... of course, need a lot of fine tuning to get there.

I will target D2 JB (16" OAL) for 65rc (61rc matrix + 4rc carbide) @300F tempered. Its edge should be fairly strong at 61rc/matrix and 5rc loss to ductility which could compensate for brittleness from ~12% carbide volume. This take care apex stability. For bevel stability - I think, volume (thick) is required to counter steering issues. So far, damages are localize, so blade probably is quite stable.

 

[BluntCut MetalWorks]

Heat treated

1x aebl 0.11" thick - 64rc@300F. Tested at 64rc 0.015"BET/15dps against AB -> FAILED (practically lost 1/3 front blade). 63rc/400F -> passed AB test.

1x d2 0.11" thick - 67rc@300F. Tested at 67rc 0.010"BET/14dps against AB -> FAILED. 65rc/400F -> smaller chips. Sort of a sacrificial lamp for D2 Jest bolo settings. I will grind it down to 0.015"BET/15dps and re-test.

1x d2 0.187" thick, 16" OAL, 14.1 oz(just the blade), guess based on smaller D2 above - 65rc/375F. Edge geometry 0.025"BET/16dps. Chopped for 30 minutes (swing with all might strength): pin-knotted pine, 2x4, oak, AB, olive, cumaru, 3/4" dia aluminum tube => passed.

Blade hard-binded to dried olive - so not sure, I want to thin this jb any further. Your thoughts?

 

[Chris "Anagarika"]

Rather than thinning down, will a more convex profile help reducing bind?

 

[BluntCut MetalWorks]

Blade grind is too thin, thus deep penetration conducive to binding. A thicker convex edge geometry will reduce penetration, thereby less binding but rebounce will be much higher when chopping much harder woods like AB & cumaru. That rebounce could translate into higher impact psi as well.

Rather than thinning down, will a more convex profile help reducing bind?

I cleanup-grinded the jb blade & edge and did more chopping. *note edge bevel face length ~0.047". Easy to cut through staples but lateral/deflected hits caused most of the damages to the edge.

edit to add: put a more consistent geometry 0.02"BET/15dps, 2K grit finished. It did very well chopping a lot of seasoned hardwood (includes: cumaru & other hard knotted woods). A mis-chop to AB resulted a 1/4" dia ripple along the apex. That indicate the edge is not strong enough prevent steering but strong enough not to fracture. I will sharpen it up, albeit there will be a slight dip along ape for about 1/8" long (where the ripple taken place). I will hard test this jb but won't do intentional destruction.

Once bevel is all set, it's easy to put a good edge on this jb. I used waterstone 600, 1K and 2K (a couple times 5K and 8K ShaptonGlass). 2K seem best: plenty sharp while apex is still strong enough. If apex radius is too low and steer is tough enough, that could initiate steering. Steering is very bad for the edge & blade

 

[Chris "Anagarika"]

Luong,

Can you explain more in last part about apex radius vs steering? Seems contradicting to me, sharper should be better?

 

[BluntCut MetalWorks]

Edge strength weaken at cubic rate, so thin apex can easily flex/steer by material. Once an edge part is diverging from the rest of the edge. This divergent can magnify by penetration distance & force, unless thicker behind edge+blade stop the penetration. If divergent rate & distance exceed blade yield point (due to bending), voi'la damage could be as big as half-moon

Luong,

Can you explain more in last part about apex radius vs steering? Seems contradicting to me, sharper should be better?

Finally grinded enough flat space to rockwell test this jb - it's 65rc.

A quick 3:28 runtime - D2 jb chops some pork rib

[video]https://youtu.be/23M9liJU0rc[/video]

 

[BluntCut MetalWorks]

I did a pilot ht batch w/ working hardness (5 readings per steel):

W2* 67.5rc
52100* 65rc > target 67rc next time
Hitachi Blue#2* 67.5rc
Cfv* 65rc > target 66+rc next time
O1 67rc
1084 68rc
1095 68rc
Hitachi White#2 san mai - guesstimated 67+rc

* = grinded with edge geometry at most 0.012"BET/12dps. Passed carve & hard whittle Katalox & Buloke wood.

 

[BluntCut MetalWorks]

Finalized ht recipes for i) W2+1095+1084+O1 ii) 52100 + CruforgeV. (i) working hrc range: 65-67. (ii) 63-65rc.

Yesterday, I got 2x ~12" long log chunks became my new chopping station ... hahaha of course, here is a video(4m29s) to show them off

D2 Jest Bolo 3/16"/4.76mm thick
65RC
0.024" / 0.6mm behind edge thick
13-15 degrees per side

https://youtu.be/31QhteR1SMA

 

[BluntCut MetalWorks]

A question for all - Should the blade area near spine stays at full hardness or soften it with a torch?

If leave at full hardness, easier for user to Rockwell Hardness test - satisfy those curious about high rc. Torch softened would lower a chance of the blade split into 2 - when blade is subjected to destruction force (LOL a big half-moon chip instead, EOL). Also softened thicker cross section widen the elasticity range to compensate for thinner cross section ability to bend further than thick (in context of Young Modulus).

 

[Chris "Anagarika"]

For user experience, I'd suggest soften.

Edit to add:
On second thought, can it be offered as option? BF member who is curious might want to HRC test it. NKP might not even know what HRC is, just that the performance is great.

Edit 2: the D2 JB is scary. No parrying with broom handle :thumbup:

 

[BluntCut MetalWorks]

Thanks, Chris,

I will soften/toughen the tang neck and leave full for hardness for blades with spine less than 0.8" thick at midpoint. It's fairly easy to get a rough idea of hardness by spine-to-spine whittle BCMW knife against known production knife, such as Spyderco's zdp189 knives where known hardness - for current products - is around 63rc. Deeper & rougher damaged corner = softer.

I chopped dried & green pine+burgenvillia branches with D2 JB for 2hrs = d2 jb laugh at tired Luong

For user experience, I'd suggest soften.

Edit to add:
On second thought, can it be offered as option? BF member who is curious might want to HRC test it. NKP might not even know what HRC is, just that the performance is great.

Edit 2: the D2 JB is scary. No parrying with broom handle :thumbup:

 

[Natlek]

A question for all - Should the blade area near spine stays at full hardness or soften it with a torch?

If leave at full hardness, easier for user to Rockwell Hardness test - satisfy those curious about high rc. Torch softened would lower a chance of the blade split into 2 - when blade is subjected to destruction force (LOL a big half-moon chip instead, EOL). Also softened thicker cross section widen the elasticity range to compensate for thinner cross section ability to bend further than thick (in context of Young Modulus).

bluntcut , what do you think about this ? Quote text is for tungsten carbide , but how this apply about the steel ?

The majority of grades we machine are made with standard size grains varying between 1 and 3 microns in size. Using larger grains of 2 - 6 microns will greatly increase the strength and toughness of the material because the larger grains interlock better. The trade off is that larger grain materials do not offer as much resistance to wear as finer grain sized materials. Sub micron materials that vary between 0.4 and 1.0 micron grain size are harder than standard grain materials with the same cobalt content. The sub micron grains are much more uniform in size and hence give improved hardness as well as increased carbide strength. However, as specs show the transverse rupture strength is perhaps 20% improved on 15% sub micron compared to 15% fine grain material but this can give a false impression as sub micron carbide is not as resistant to impact and may chip more easily.

 

[BluntCut MetalWorks]

In WC context, grain is WC particle and Cobalt(Co) is binder could corresponding respectively to poor ht steel where grain (matrix sub unit) is stronger & tougher than grain boundary(think weak binder). Then true, smaller sub-unit or particle and more grain interface/boundaries for same amount of binder/Co, which would be stronger and tougher.

https://en.wikipedia.org/wiki/Grain_boundary_strengthening

Good ht steel with clean grain boundary shift the weakness to a next least strong/tough aspect, such as plate martensite and or retained austenite and or coarsened precip carbide and or carbon lean martensite, etc..

Awesome ht designate lath martensite as its weakest and happen to be what give steel its strong attribute. Straight long lath is much (up to 2.5x) weaker (strength & toughness) than super short highly woven/inlocking lath. Lath itself is just an offset in lattice, which give steel a given baseline(from straight long lath) strength.

Carbide volume affects/degrades toughness linearly from 0 to https://en.wikipedia.org/wiki/Percolation_threshold, when drop rate is near exponential before flatten out to linear again.

bluntcut , what do you think about this ? Quote text is for tungsten carbide , but how this apply about the steel ?

 

[BluntCut MetalWorks]

The last 2 days, I hardened blades prepared in 8/2015 and ~12/2014 using my latest hardening step. Quite happy with the results and here are their working hardness

Cfv 65rc; 1095 67+rc; 52100 65rc; W2(12/2014 11" gyuto) 66rc

Hardened that W2 gyuto today

W2 0.125" thick, 11"/28cm stick tang, 66rc, 0.015" edge thick, sharpened 15dps

In early stage of post-ht grind this gyuto, I decided to hard use test this edge for stability.

This knife is fully hardened, with possibility of a breaking (into 2 pieces), hence capture in video as a data point.

Thanks for watching (10 minutes video) & comments.
https://youtu.be/rO3HMOB9yT8

Chipped edge after chop bone

 

[chiral.grolim]

...W2 0.125" thick, 11"/28cm stick tang, 66rc, 0.015" edge thick, sharpened 15dps...

Those are nice clean chips, you can see how the bone steered the edge. I hazard a guess: when you chopped the far end of the bone unsupported, you catapulted the other end with enough force that the bone twisted (steered) the edge with it as it went, probably wouldn't have happened if you could hold the end of the bone like you can with the longer dowel-rod or broom handle. That said, it demonstrates the fragility-threshold of the edge geometry and matrix...