Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

test 2 numbers

Posts

test 2 numbers

#1

test 2 numbers

david weaver

another test, another migraine..

01 - 780 feet

tsunesaburo - 790 feet

ln a2 - 1000 feet

chinese hss - 1360 feet

3v - 1360 feet

v11 - 1570 feet

feet is until half of the length of the board omes out of the ut intermittently. first diffiulty is generally 10-20% prior

may upload some pictures later when migraine subsides.

Re: test 2 numbers

#2

Re: test 2 numbers

Charlie

If I'm interpreting what you're saying, I'm not sure Rob Lee himself would buy into those numbers. You should consider partnering with somebody as a sanity check. There's an air of implausibility creeping in I think. Don't take this as an insult, just as a caution not to let the process get away from you and have the whole exercise dismissed as farcical.

Re: test 2 numbers

#3

Re: test 2 numbers

David Weaver

I gather that to mean you don't like the results, probably because of your decade+ long issue with derek and LV.

There will be enough process description in my write up to allow anyone to duplicate this test, all the way down to modifying the smoother used so that it would work with thick and thin irons (which ruins the cap - I'm not wild about that).

Do I have any concern about the test becoming a farce? No. I think any honest individual willing to go to the same level of exactness as I did will probably get similar results (see comment about 3V below).

The only potential variant in this is the fact that 3V had to be tested in a record 5 1/2 while all of the other planes did this entire test in a lie nielsen smoother (the maple test was tending toward these results judging by which iron was cutting best before the irons found the inclusion. O1 was clapping out, the tsune was experiencing clearance issues and the edge of the A2 iron was undergoing fast degredation - the pictures in this progression will be similar, but the beech wasn't quite as hard on edges, so it lasted longer).

To be quite honest, I expected that I might catch the V11 iron not being materially better than A2 by more than 10-20%, which I expected to match the HSS chinese iron. I expected 3V would win because it's the most resistant to diamonds by far. I didn't state a prediction of sorts because what if that actually happens?

Not only do I have the distance planed, but I have actually weighed the shavings planed during the process and all failure points were checked by re-waxing the plane (that pushes the failure point out a little bit) and re-setting cut depth and checking by caliper (allowing variance in the thickness of the failed shaving would've spiked the test - all failed with the same shaving thickness).

If you don't like the way you feel after seeing the results, I can't help you. I spent enough time listening to you tell me that the photos in my cap iron article came from me despite the captions describing where they came from that I'm not interested in arguing with someone who argues about what they wish was correct than what really is correct.

This test doesn't mean that the same results would occur in a less than ideal test (this is a test of a clean environment with clean wood, a fairly stout final bevel angle, planing at hand speed, and with a wood type that is hard but not abrasive or abusive). After seeing the way the tsune took the bark inclusion, I'm fairly sure that impact into hard things or inclusions is more in its favor than the other irons. This test didn't include those. What about a dirty board that's been in the shop and hasn't been cleaned off? I don't know. We didn't test that.

Re: test 2 numbers

#4

Re: test 2 numbers

John in NM

In fairness, the air of implausibility may be in our minds rather than in David's testing.

While I am surprised that the v11 beats 01 by ~2x, that is more because of the dismissive things I've heard so many saying about how v11 is just a hair better but they prefer 01. The thing that strikes me about David's test is that he's optimizing the sharpening routines for each and taking photomicrographs of the blades - that could easily explain why he's seeing better results with the v11 than many people who bought blades and then dismissed them as not that much better than 01.

I have a couple v11 blades that I found impressive in my limited use of them. I fully expect to be less impressed after I try to hone them on my Arkansas stones. It may be time to buy a cast iron billet and face it down to flat and use up some of the diamond I have left over from gem cutting days.

I'm also reminded that while there is no free lunch, none was promised. A free lunch would be a vastly better blade at the same price. The other gripe about those super steels is that they cost too much for 20% improvement. Perhaps this is better viewed in the perspective of incremental improvements of technology with accompanying increment cost increases.

Re: test 2 numbers

#5

Weight planed by each iron

David Weaver

(i did not measure 1/2 grams. I have a postal scale that works in 5 gram increments. When it's right between two, it will blink back and forth between the two values and I will record the average of the two)

As irons have a tendency to start coming out of a cut more easily (generally the last 200 feet or so), it becomes more difficult to get a continuous shaving, so I would expect these weight samples to be slightly biased high vs. distance.

That is, while the O1 iron may have done 3/4ths of its shavings with ease, a higher percentage of them was either fighting loss of clearance or dullness of the edge (it's not really possible to tell which factor is really preventing starting or staying in a cut). V11 started (according to my notes) showing the first signs of clearance loss around shaving 300 but got through its next 50 after that without going downhill quickly. It did, however, become the same trouble that O1 was from shaving 150 to 195 right around shaving 350 (checking wax, more deliberately starting cuts - there is a reason for that, as issues coming out of the cut compound, and you can make one or two here or there by not being diligent at the start. A slow pass or two make correct an issue and prevent a downward feeling spiral. Once the iron is truly dull for a shaving thickness, it will start on a perfectly flat edge with no undulations left by a struggling edge and then struggle within a few shavings).

V11 - 377.5 grams

O1 - 175 grams

chinese hss - 305 grams

tsunesaburo - 190 grams

LN A2 - 275 grams

3V - 327.5 grams

The weight of the shavings is a redundancy from both subjective judgement of whether or not a shaving is getting away from spec during the test as well as whether or not something in the shaving quality (cap iron straightening, etc, lateral adjustment out of spec, etc) would cause a misleading check of shaving thickness.

Re: test 2 numbers

#6

Re: test 2 numbers

Wiley Horne

David,

I find your results perfectly credible, and I much appreciate yourmdiligence and honesty.

This question is for after your migraine subsides. I am interested in how many feet were planed (or ballpark percentage of the total run) before you would have sharpened. For example, if the plane is no longer positively pulling itself into the cut, I sharpen.

My reason for asking is this: If you use the metric “when the plane is no longer clearly pulling itself into the cut”, would the blades tend to score closer to each other, or would they be about proportional to the figures you show? I am trying to get at the question, do the high-durability alloy steels give significantly more top-quality feet planed, or just more acceptable feet planed?

Now, I am aware you have already said that only the clean steels give an unblemished surface, to your eye. (This is a claim I haven't been able to duplicate, but then my eyes have seen better days.). But I am asking this: If you go by ‘positive feel in the cut’, would the various irons score relatively the same as you have shown, or would the picture change?

Thanks, Wiley

Re: test 2 numbers

#7

Bravo

Bill Tindall, E.Tn.

Finally, after years of speculation we have some data to chew. Oh behalf of the shaving lifters of the world, thank you. These data need to be posted on the Main Forum where the makers of stuff and shaving lifters live.

From steel composition, and published properties of abrasion resistance, the blade ranking offers no surprises until one gets to PMV11.

M2 has a higher load of abrasion resistant carbides than all the others, twice 3V, so I might have though it would be better than 3V. However, the 3V is a particle steel and that is going to help. From my experience with jointer knives particle M2 is significantly longer wearing than "ordinary" M2. So 3V equal Chinese M2, we shouldn't be too surprised if abrasion resistance is the main driver of edge life. This data set does tell us that toughness is irrelevant for this 3V is way more tough than the presumed M2.

But what to make of PMV11? Let the speculation begin. For a shaving lifter, twice the life of O1. That's significant. The one obvious difference is this steel is highly corrosion resistant. Could corrosive wear be a factor? It is in high speed tooling for milling MDf but that is about a light year away from David's work.

Re: test 2 numbers

#8

question

Bill Tindall, E.Tn.

I realize the first experiment was prematurely terminated. That said, within the limits of the data obtained does the maple data at least qualitative agree with the beech data?

Re: test 2 numbers

#9

Re: test 2 numbers

Charlie

Peel yourself off the ceiling. Publish your results and we'll see.

Re: test 2 numbers

#10

The subsidized lunch...

David Weaver

..as far as cost goes, there is no comparison in performance vs. price to the chinese irons. I cannot buy O1 stock for less than the cost of two chinese irons, and I still have to make the O1 irons then.

It's fair to say that is some cases, preparing the backs of the chinese irons (which are HSS and often not flat) takes about the same amount of time as it took me to make the O1 iron used in this test ...and it was far less interesting, though I guess the chance of personal injury was greater handling glowing orange irons that make flames when you quench them).

I bought the LV iron used in this test for price plus shipping. It was something just south of $60, and I felt that given that it was not really intended for me to use in the long run, and just as a matter of having something to test that I bought (I wouldn't ask a manufacturer to supply me with an iron for something like this for two reasons).

Re: test 2 numbers

#11

I'm confused

Bill Tindall, E.Tn.

when the plane is no longer clearly pulling itself into the cut

I thought this was the end point David used. If not, will someone explain the difference for it is critical to understanding the test results.

Re: test 2 numbers

#12

Re: Bravo

David Weaver

The 3V may have benefited from being in the same plane. The LN planes a little bit more sweetly. I was concerned that it would have an advantage in my old record plane (which has been lapped and is not far off of LN's flatness spec, though it's outside it).

In both cases taking pictures, I cannot see what's causing V11 to last longer. On a 4 foot edge with me fencing the plane with my fingers, I can only keep the plane in the cut so much, so I can't test the idea that maybe I can make enough heat at the edge to favor the steels that temper at a higher temperature (given that Rob told us that we should treat this like carbon steel and not assume that it will tolerate heat).

I planed at a rhythm that someone would work an edge neatly, but not slowly.

I'm surprised by the results given that the steel feels like 440C to me, has a rubbery wire edge and will somewhat grade a washita but offers no resistance to the japanese natural stone that I showed.

I didn't know what to expect of the chinese iron, so I can't go further than saying if it's M2, it sharpens pretty easily for it, but I have confirmed with this iron that you can brown it and it will perform just as well after browning the edge as it does without it. On the muji plane that I have, I believe the steel is identical, I have ground that one until it glowed to see what would happen. Same thing, I can't tell any significant change in performance.

Pictures of the tsunesaburo, which is the other great surprise to me, will show the same pocking that occurred in maple. I don't think it does the steel any favors and I think that a more plain white steel would last just as long or longer and be sweeter to use.

I would probably never notice "sweetness" in use (in terms of smootness of the feel of the plane through most of the work, but the V11 plane was also the sweetest feeling at all points. My O1 iron had good sweetness and was at least as consistent. to the extent that the v11 felt nice, it could just be the fit to the plane, as the LN iron being made for the plane was apparent in feel at the outset. The cut was not as sweet as the V11 fairly quickly, though.

Uniformity and good clearance without bad behvior would signify, to me, sweetness - that's just a guess. Fitness to a plane would also make a difference. The back side of an iron doesn't need to be milled perfectly, but the chinese iron is painted and rough and it's a bit gritty in adjustment because of it. The v11 iron is good, so is the LN iron. The back of the tsunesaburo has what look like coarse blanchard grinding marks on it, and those have a lot of grip. I had to wax it.

Re: test 2 numbers

#13

another question

Bill Tindall, E.Tn.

Some don't understand experimental variance. Most experimenters have a feel for variance even if it is not measured. What is your feel for these data- +/- 10-20%?

The O1 maple data has been lost by me in the dozens of posts....what is estimated feet to failure for maple? My recollection was not a lot less than beech.?.

Re: test 2 numbers

#14

Re: Weight planed by each iron

Charlie

I know that when continuously planing some boards, they get nice and slick and start sucking the plane down in spots (at least that's what it feels like to me) then the plane releases and sort of jutters on. This happens with a freshly honed iron and it happens with one that's going off too. It seems to be a matter of the plane's sole and the surface of the board and has nothing to do with the condition of the iron. Ash seems to want to do this, and maple, two completely different species. I've never had it happen on mahogany or walnut but occasionally on cherry. Waxing the sole doesn't help.

I'm sure the amount of downward pressure I'm applying has something to do with it, and it's damned hard to keep pressure constant well into long shop session with lots of planing to get done or already done.

Food for thought maybe.

Re: test 2 numbers

#15

Well said David

Jonathan Elliott

It's data, period. Very good and defensible data. Everyone gets to decide for themselves how it will, or will not, affect them. David has already said he is not going to change anything, he likes his routine and results and cares more about surface quality than longevity.

Everyone has their unique set of priorities. LV provided data in MDF, people have long wanted to see what would happen in real wood with human power. There's really nothing to argue, if people like what they are using stick with it, if they want to experiment or try something else this perhaps gives them a more knowledge point from which to launch those trials.

So a "marketing claim" seems to be turning out more true than others to this point. Maybe it wasn't a "marketing claim" in the sense of "profit is all that matters" after all and was done to bring an option to the table that didn't previously exist. No, it isn't perfect, but maybe more perfect for certain people's list of priorities and they are happy to have that option.

The work here is very appreciated and interesting.

Re: test 2 numbers

#16

some abrasion resistance data

Bill Tindall, E.Tn.

Went poking in the literature....

M2 and 3V are reported to have about the same abrasion resistance. The 3V data is definitely for particle steel. Not stated what the M2 was.

What we believe PMV11 to be has nearly the same abrasion resistance as D2 in an ASTM abrasion test. Yet, 3V is reported to have significantly better abrasion resistance than D2.

Found some data from our old acquaintance Mr. Kees. He found the abrasion resistance of PMV11 to be twice better than O1(weight of metal removed with abrasive per unit of time). Now ain't that a coincidence?

These data are consistent with David's finding except PMV11 lasts longer in wood than expected. The plot thickens.

Re: test 2 numbers

#17

Re: I'm confused

Charlie

You have a gift for understatement.

Re: test 2 numbers

#18

Re: another question

David Weaver

I would be surprised if these results varied too much in beech. The progression was smooth and nothing surprising occurred during it. 10-20% is probably a good gauge, maybe less if you had an identical board with good smooth heart wood like this one.

The O1 (by feel at the time of stopping vs. this one) was probably set to go about 650 or so feet in maple before being pulled from the test. The LN iron took on more damage at the edge, and I'd be surprised if it lasted another 200 more (800). The same relative feel of wear at that point was occurring. The tsunesaburo iron at 600 feet was starting to give indications of lack of clearance or some sort of dullness and I would guess that it would've been done by 800 even without the inclusion.

V11 seemed to have a lot of visual wear in that test, but it was still planing very sweetly and I have no reason to believe it was close to stopping. The edge was very uniform, and you can't necessarily believe what you see in the pictures from the scope as one part of a picture can throw off the software's interpretation of another. The 3V was nowhere close to stopping on maple, either. I can't remember what the feel of the HSS was, but when you have one or two bad shavings testing an edge like this, sometimes they're become the plane before left a dud surface or you bob off balance. A good cutting plane removes the struggling plane's evidence in one or two strokes, so I did not refresh the surface between planes. If I had to guess overall, I think all of the planes would've lasted less long on maple, and maybe the relative differences would've varied a little, but I don't have any reason to believe that the order of finishing would've been any different.

Re: test 2 numbers

#19

It's a lot easier...

David Weaver

..to keep constant pressure on a plane when you're making an identical stroke over and over.

I went to bed sore last night. It's not like this is hard planing, but it is the same position and the same stroke over and over.

There is no significant error in my method on this test, and it's not like planing a board that just came from a jack plane. It's also not hard to be consistent doing it - you get locked in.

In testing when the plane came out of the cut, it made no difference if I had my hand on the rear handle only or thumb applied on the web in the casting in front of the mouth. I wasn't applying that much pressure there, using it mostly to make sure the plane was level at the start of the cut and so that I had a constant place to put my fingers to to fence the plane so that the wear happened in generally the same place on each stroke.

I expected my front hand would have more influence on keeping the plane in the cut, but the downward force from the back hand on the plane must be more.

As to variations in other wood, you're describing something that's more of a cutting problem. There was no significant variation in this wood. it was a 5/4 quartered beech piece of wood, heart only that I was saving to make planes. It has sap, too, but this test didn't reach it.

I've planed a lot of beech and never noticed the issues that you're talking about with it, it doesn't really do it. It is far easier to plane on the quartered side, but sometimes very tearout prone on that side, and on the flat face (which is what this edge becomes on a quartered board), it's very uniform and with more resistance than the quartered face probably by 50%.

IN all cases here, wax restored better function until the very end, so I did it often. Every 20 strokes. (start, 20 and then at 40 strokes). It lowered effort at the beginning of the test and then during the point when the iron was failing, improved function a little bit. There is no noticeable chatter or hopping either way, the plane just stays in the cut a little better with fresh wax.

Re: test 2 numbers

#20

Re: question

David Weaver

Yes, it does. There were no obvious differences other than how fast the A2 was going downhill at the 600 shaving point. I don't know if it would've lasted 20 feet or 200, but the edge was falling apart. The last time I tested in maple, it was with several A2 irons. All of them did that. I don't know what causes it. It seems to favor beech a little bit more.

Notice in the pictures when I post them, though, that the particles steels appear to release little grains and the LN and HSS seem to release bigger ones.

Re: test 2 numbers

#21

Re: test 2 numbers

David Weaver

I would've sharpened each of these probably 200 feet before they stopped in this test. I was initially going to make it so that any failure to stay in the cut was terminal, but as this was all on one board, it would've been in different parts of the board. I sharpen on a day to day basis at the first sign of a plane not entering a cut easily - not because you can't push it further, but because nothing good really happens then (intermittent cut means intermittent evidence on the surface or more work).

Strangely, the particle steels continued to enter the cut, but lost their touch staying in it. The carbon steels (O1 and tsunesaburo) failed to enter the cut first but then would stay in it once they got in part of the way down the board. The HSS and LN irons were between those in behavior, getting intermittent all over the board.

In reality, with the v11 if I was actually smoothing this much continuously, I'd stop to take a break and sharpen. I'd have attributed some fault of mine (like leaning on the plane, etc, to building resistance, though in comments i have made elsewhere Veritas plane vs. the wooden try plane sizing plane billets, I did stop in that case and the iron was actually dull - sizing a blank is split up by sawing the fat sides off, so you don't get stuck planing 400 strokes in a row.

So, long story short, if you figure somewhere around 75-85% of the strokes were what I'd call good performance, you'd be at it. The rest was progressively getting to the same place so that I didn't favor one type of failure over another.

As far as surface quality - the O1 iron remained the most consistent. V11 was actually slightly brighter to start, but brightness is a bit of a curse when it becomes absurd - you cannot have any mark anywhere without seeing it. The tsunesaburo iron, whatever the little pocks are, occurred right away and the finish was duller than I'd have expected but still shiny and very uniform end to end on the test. The V11 progressively got more dull looking but not until the end in beech did it show any faults. Chinese, 3V, V11 and A2 all showed very light lines by the end of their time, but unlike with the more punishing and variable maple, nothing showed up mid test other than that.

My experience in cherry with A2 is that I could not deal with the lines that it made after a little while. Why the beech didn't induce any significant failure, I don't know. The lines that were made were very faint in this case and none came early. The wood is also extremely smooth planing with little variation. Little grain variations in the maple would tear just a little as the irons were dulling. Does that mean that the carbon steels do better when something untoward is part of the wood or on the surface? I don't know. The tsunesaburo iron tolerated the bark inclusion on the maple far better, but that's one bark inclusion one time - not really enough to make a broad statement.

Re: test 2 numbers

#22

M2

David Weaver

I suspect if brent beach's work is still relevant, the difference between V11 and what is probably M2 or similar here with the chinese steel would probably be made up for by the ASW iron, but that's not something that can be said for sure without actually testing it.

I'm not sure I'd want to test those two irons against each other - if they end up identical, One is about $50 and the ASW blades were on the order of $120. I'm not doing this testing for people to take recommendations, and I'd be through planing 3000-3500 feet for two irons.

What I can say after the maple and beech with much confidence is that the figures in this test are pretty representative. Clean wood without harsh flaws in it is necessary to achieve them.

What would they be in pine? I don't know. I'd hate to find out. All kinds of curiosities abound, but I'm glad to hear that kees had similar experience on his planing machine. There just isn't much to this test once we got all of the nits worked out so that most of the tooling was working on the same plane with the same fit on the same board and alternating so that variances were minimized.

The mechanism that's causing it, though? I have no idea. The carbon steel (my O1 iron, which I tempered so that it would still be hard for O1 but not fail at all by chipping) did what it should in both tests. No chipping. I don't like chipping and I'm pleased with it.

When someone tells me they're new to woodworking and what should they get, I just don't even know what to say any longer. If they'd been working long enough to tell me the way they sharpen, what they use, and how long it takes (if it was very fine abrasives and their sharpening cycle took them a long time), I'd tell them any of the top three in this test. I think A2 is relatively obsolete at this point, but not in a way that someone should replace it. Plus, if you really like LN's planes, then that's what's in them. I had a milled chinese iron (not tested here) in this LN smoother, which allowed me to have an almost new iron to test these on a comparable basis.

Once one can sharpen fast and use a double iron, not much of this stuff is a big deal, but I get that there's lots of folks who don't want to learn either of those.

But I am glad to see corroborating data. The way carbon steel fails is it just looks uniform and then it appears to run out of clearance, as the edge remains crisp without showing light at various depths (suggesting the very edge isn't that round) and then it just runs out of clearance. I expected as the V11 was cutting that it felt sweet (I'll have to define what that means, not sweet like a surfer would say, but sweet like the cutting characteristics) like carbon steel (despite not being sweet like it when it's on the stones) - and that somewhere around or shortly after A2, it would just run out of clearance. That would still be nice characteristics. It had that sweetness until right around when the 3V and chinese irons failed, and I expected that it might beat me up for a couple of hundred more strokes. I'm glad it didn't.

I could come up with all kinds of theories about it and 3V - the edges when you look at them are fine, but they're not otherwordly. Why do they appear to release parts of their edges in clumps on beech instead of in rows like the O1 does? Is something being pulled off of it when clearance is threatened? The fact that kees weighed it and only half the mass got pulled off suggests that's not it.

It's just something that is at this point in the conditions that this test were done under. I didn't really believe the marketing claim and also wanted to see how good a cheap chinese iron was against something known (LN and tsunesaburo). I wish the 3V could've been done in the same plane.

I've got a gaggle of other tests to do now, but this one blew up my last two weekends so maybe I'll do those off and on. I see that beach said he tested irons over years. I'll do it chicken and feathers style - it's not my nature to do tiny bits at a time of the same test. All or none. I'll do more limited tests now because I don't want to plane away all of my beech, but I would like to see how V11 will do across end grain (that should be a lot more brief) vs. the O1 iron. I"ve been using stanley plane irons and O1 for a while and have planed the ends of a lot of panels off, as well as drawer overlays and It's never occurred to me that I should have something tougher.

But testing feet planed in end grain to check if ratios stay the same would be nice, and testing a lower level of edge refinement with the two (if people using diamonds can manage a wire edge, a single cast plate with one diamond grit in the shop would make sharpening a whole lot more productive) and a 30 degree angle would be nice. If there's something taking lots of steps and being very particular about final bevel angle like I was here (piggybacking off of what I learned along the way with A2), it'll be nice to see what's worth faffing with and what's not.

Day to day, I don't faff with much. I'd still put a beginner on a stanley iron, a bare leather strop and a washita if they said they wanted to spend two weekends with me. The washita and the double iron both made me forget about chasing edge durability.

Re: test 2 numbers

#23

Re: test 2 numbers

E Phillip Smith

Nice set of data!!!!

Re: test 2 numbers

#24

Thanks! thus far...

David Weaver

..I've planed about a half mile further than LV did in their testing. That strikes me as odd, but based on the results, I'll give them a pass.

11,420 feet of identically targeted strokes last weekend and this weekend.

Re: test 2 numbers

#25

Re: M2

Charlie

Chris Gochnour tested V11 against A2 on end grain white oak in a Fine Woodworking article. It did quite well - beat the A2 soundly. The results were more even between the two when he planed long grain teak.

https://www.finewoodworking.com/2012/10/25/pm-v11-tool-steel-review