K294 Mule Team - anybody KNOW the "probable HARDNESS" of the release?

[Traditional.Sharpening]

Agree with the science but wonder why the HIGHER RCs sure seem in great STEELS to out-preform the same steel with less hardness by doubling the cuts...

Well, in order to be sure of that truth I'd want to design a comparison myself to be sure the data you are providing is actually correct there. The data I've found from Cliff Stamp's sources suggests otherwise, that even something like 10 point difference in RC value is not anywhere near twice as much performance. To put it plainly, the metallurgical data from other sources do not agree with Ankerson and this is a very-long standing war between Ankerson's methods and 'metallurgical science'.

[Bolster]

...Sure, there can be correlations but generally speaking correlation does not equal causation....

I'm following your overall argument, but the above sentence is like nails on the chalkboard to a quantitative researcher. Correlations are precisely the first place to go looking for causality. It's true that correlation doesn't equal causation and there are many entertaining examples of that. But to establish causation, you first need correlation. I would say you'd be on solid ground by saying: "generally speaking, correlation is a prerequisite for causation, but not necessarily evidence of it."

Now to the hypothesis that hardness is not an independently important factor in performance (however performance is defined). Here's how you'd test it. You first decide on a dependent variable...such as CATRA cards cut, or rope cuts, or feet of cardboard cut, or whatever quant measure you decide best determines "performance" (ideally, a combination of dependent variables.) Then you load a regression equation with all the independent variables you think would impact "performance": carbide volume, types of carbide, BTE, dps, % of austentite/martensite, and so on. (Alternately you hold the variables constant that you can, and don't enter them.) You run your regression with Rc both in, and then again not in, the equation, and watch the R-squared 'proportion of variance accounted for.' If the R-squared drops considerably when Rc is out of the equation, that supports the idea that hardness contributes an independent amount of performance...above and beyond all the other variables you entered. The regression will even tell you its percentage of contribution. If the R-squared is similar regardless of Rc being in or out of the equation, you have evidence that indeed, Rc doesn't add additional 'performance' above and beyond the other variables entered. In that case, if there is a correlation between Rc and performance (which there unambiguously is--clearly visible in Larrin's CATRA graphic), but the R-squared difference says it's not an important independently contributing factor, then you can conclude that other elements in the equation are both making high Rc and accounting for performance, but it's not the Rc itself that's directly accounting for performance. Or you could run a mediation test (such as structural equations modeling) and determine that directly.

I could perform the analysis easily enough if I had the dataset. It's getting the dataset that's onerous. My guess is that Larrin has the dataset to run an analysis of this type, if he were so inclined. He also has the chops to run the regression; I've seen his regression equations.

Sorry for the wall of text, and not trying to start a flame war. Just making IMO an epistemological point that needs saying. Without a test of some sort, we can't really nail down the independent contribution of Rc. Given that correlation hints at causation, certainly one could be forgiven making the hypothesis that Rc has an impact on performance.

[Traditional.Sharpening]

Yes, RC has an impact. No. RC is not a causative factor but it can certainly improve performance in general assuming the micro-structure is well balanced. This is not something I'm going to get into as I have no need to prove this but again there are many important factors. Many of which are related to heat treatment cycles and how they affect the micro structure.

Many of the 'junk' steels have earned said reputation due to manufacturer's giving them 'junk' heat treatment and therefore people simply label them as junk because of a correlation that's been noted. 8CR13MOV could be considered junk by many here but given the heat treatment Spyderco gives them I have zero issues with it and don't consider it a terrible down-grade.

Does it score high in RC or CATRA as Spyderco heat treats it? Not really. Does that mean it's a poor choice of steel for most people? Not at all but many here still consider it a low class steel regardless of this fact because it's cheap. Cheap knives correlate to low-quality heat treatment, therefore cheap knives can never perform well if we're going by correlation.

See what I'm getting at here? It's something that's not easily understood deeply unless you're willing to study it at great length and carefully design a study to prove/disprove your hypothesis. Most everyone here will not do such things and rather than admit they really do not know much about what makes a suitable steel/heat treat they prefer to indulge a bit of fantasy.

[CLOK]

Nothing to add to the above discussion…just wanted to add that no hardness test dimple on this one like there was on the previous mule from Taiwan. May be something the Taiwan factory started and may do going forward.

[Bolster]

See what I'm getting at here? It's something that's not easily understood deeply unless you're willing to study it at great length and carefully design a study to prove/disprove your hypothesis. Most everyone here will not do such things and rather than admit they really do not know much about what makes a suitable steel/heat treat they prefer to indulge a bit of fantasy.

Yes, I think I'm following your arguments reasonably well. The point I must have made poorly is that "Does Rc have an independent effect on performance" is an empirical question...it doesn't require argument or conjecture...just data and the correct statistical tests. I'm guessing that the data to answer the question empirically probably exist on somebody's hard drive. If it were on mine, I'd have it tested by tomorrow morning!

So if any of you happen to see an errant dataset wandering about, that includes the variables I mentioned above (and other variables that Trad Sharp may be able to specify), please be so kind as to send it my way for analysis. It's a very interesting question, the independent effect of Rc on performance. Having such a dataset is my fantasy.

[Traditional.Sharpening]

See what I'm getting at here? It's something that's not easily understood deeply unless you're willing to study it at great length and carefully design a study to prove/disprove your hypothesis. Most everyone here will not do such things and rather than admit they really do not know much about what makes a suitable steel/heat treat they prefer to indulge a bit of fantasy.

Yes, I think I'm following your arguments reasonably well. The point I must have made poorly is that "Does Rc have an independent effect on performance" is an empirical question...it doesn't require argument or conjecture...just data and the correct statistical tests. I'm guessing that the data to answer the question empirically probably exist on somebody's hard drive. If it were on mine, I'd have it tested by tomorrow morning!

So if any of you happen to see an errant dataset wandering about, that includes the variables I mentioned above (and other variables that Trad Sharp may be able to specify), please be so kind as to send it my way for analysis. It's a very interesting question, the independent effect of Rc on performance. Having such a dataset is my fantasy.

Independent of what exactly? I think if you can just test knives that are all the same RC measurement and note the differences you'll see quite easily how much those examples vary in performance. So if the RC value is the same for all and the performance varies wildly, have you not disproved the theory that RC holds some sort of magical/mythical level of importance?

[CLOK]

Since this has been asked a few times in relation to hardness, came across an old post that may answer some questions on some old ones that this user tested on his. Cant vouch for the accuracy, just thought it would be helpful. Again, old post, so newer ones aren't on there. Sorry for cluttering up the thread as this doesn't include K294.

I’ve measured all my Mules and this is the results:
MT02P CPM-M4 62HRC
MT03P CPM-S90V 60HRC
MT04P ZDP-189* 65HRC
MT05P 9Cr18Mo* 59HRC
MT06P CPM-S35VN 58HRC
MT07P DPS15/VG10 64HRC
MT08P Aogami Super Blue 63,5HRC
MT09P Cobalt Special SUS420J2* 56HRC
MT10P CTS-BD1* 57,5HRC
MT11P M390 60HRC
MT12P CruWear * 59,5HRC
MT12P CruWear (2)* 62HRC
MT13P Elmax 60HRC
MT15P CTS-B75P 63HRC
MT16P CTS-XHP* 60HRC
MT17P K390 64HRC
MT18P CPM-S110V 59HRC
MT19P PSF27* 61HRC
MT20P CTS-B70P* 60,5HRC
MT20P CTS-B70P(2)* 61HRC
MT21P CPM4V* 64HRC
MT21P CPM4V(2) 65HRC
MT22P RWL34* 59HRC
MT22P RWL34(2) 61HRC
MT23P CPM 20 CV 60HRC
MT23P CPM 20 CV(2) 61HRC
MT24P Maxamet 69HRC
MT25P L200N* 57HRC
MT25P L200N(2) 58HRC
MT26P PMA11* 63HRC
MT26P PMA11(2) 64HRC
MT27P Micro-Melt PD1 63,5 HRC
MT27P Micro-Melt PD1(2) 63HRC

[swknight]

Thanks for the CHART ABOVE!

YOU CLOK'ed that right...

Interestingly it showed ANKERSON's old figures are VERY similar...

Nice to have something right...

The S110V Mule around 60 RC -720 cuts
The S110V Manix around 62 RC - 1120 cuts

If one searches Darin's book on Knife steels - one finds out that 125V comes from 90V with % of adjustments...

[CLOK]

Thanks for the CHART ABOVE!

YOU CLOK'ed that right...

Interestingly it showed ANKERSON's old figures are VERY similar...

Nice to have something right...

The S110V Mule around 60 RC -720 cuts
The S110V Manix around 62 RC - 1120 cuts

If one searches Darin's book on Knife steels - one finds out that 125V comes from 90V with % of adjustments...

Looking back through old mule threads showed ANKERSON posted some great information.

[swknight]

Now all we need is a S125V MULE TEAM ARMY example...

[applejuice54]

So no one has poked the k294 mule for actual HRC.

I’ll check here again in future, would be nice to know.

[swknight]

Amen...

[Bolster]

Hardness has a major effect on edge stability, according to Larrin. The R-squared is .66, which means that the raw correlation is a whopping .81. Larrin also explores cases where "too hard" leads to edge instability, so it's not a straight line "harder is always better" sort of thing. But an overall correlation of .81 between hardness and edge stability in Larrin's research is a big, fat, important correlation that can't be ignored.

In this same document, Larrin politely and circumspectly states that his data doesn't well support Landes theories on the effect of carbide volume: "...the quantitative measure of edge stability presented here did not indicate clear trends with carbide volume...Hardness appears to be the strongest controlling factor for edge stability. There was not a clear trend with carbide volume..." though he follows up with caveats about why he and Landes' master's thesis don't arrive at the same conclusions.

Long story short: I don't expect to see Larrin putting obtuse 20-30 dps angles on his ATS-34 blades...but I really don't know how Larrin sharpens his large carbide steels...if he even owns any, he's clearly not a fan of them...so that's just speculation on my part.

[Deadboxhero]

In order to understand hardness, you have to think of hardness as body weight.

Generally speaking, heavier people are stronger, Mass moves mass. However, body weight alone does not determine a person's strength.

The body composition is an important clue to put the body weight into context.


Body composition is like the steels microstructure.

Yet, unlike steel, you can easily see the difference between two people at the same weight but different body composition at a glance.


So, if you have two athletes at the same body weight but one that has a higher amount of muscle composition at the same weight then technically that athlete has a higher potential of strength.


So yes, hardness is important and higher hardness with good microstructure will certainly be superior in edge stability than a softer hardness but not to the point that it's worth compromising the microstructure with increased RA to chase 0.5-1.0rc increase in hardness.

[swknight]

Thanks LOVE all four of your S15V Spyderco knives... Manix, PM2, Shaman, and the Mule Team...

[Bolster]

...So yes, hardness is important and higher hardness with good microstructure will certainly be superior in edge stability than a softer hardness but not to the point that it's worth compromising the microstructure with increased RA to chase 0.5-1.0rc increase in hardness.

So...you're saying...that if I want to get stronger I should lift weights instead of eating twinkies?

J/K! Good analogy, thanks!

I finally found Larrin's regression formula that includes hardness--just hardness, as a unique predictor--of TCC, below. That formula indicates that hardness, independent of all the other variables in the equation, predicts TCC. I wish Larrin had posted Betas as well as the coefficients, so we could compare the relative strength of contributions. But even without Betas, "the proof is in the pudding."

[kennbr34]

Anybody do a Rockwell test yet?

[swknight]

NOPE! NADA!

Love to have a list like that above on all the current and past knives...

Ole History becoming that...

Like Big Brown Bear mentioned above "it is kinda like weight"
The bigger the number the Bigger the Bear... ;>)