Some CTS-XHP tools steel observations.

[Kees Heiden]

About one and a half year ago I did some testing on common handtool woodworking tool steels. I compared CTS-XHP (the steel formerly known as PMV-11) to O1 as it wears when planing beech and oak. I also compared CTS-XHP, A2 and O1 in how it wears away on a sharpening lap. After a while I found that tool wear is a huge complex subject and I kind of lost interest to spend all my time on this project.

So here is some of the rough data.

CTS-XHP is a powder metal with 1.6% carbon, 16% chromium and about 0.4% each of silicon, nickel, vanadium and manganese and some 0.8% molybdenum.
A2 has about 1% carbon, 5% chromium, 1% molybdenum, 0.3% nickel and some 0.3% vandium.
O1 has about 1% carbon, 0.5% chromium, 1% manganese, 0.5% silicon, 0.4% tungsten and 0.3% vanadium.

So, the big difference between the three steel types is the chromium content. CTS-XHP is a stainless steel. A2 has enough chromium to make it markedly more wear resistant, but also more prone to chipping (less tough) then O1. CTS-XHP avoids this chipping problem because of the powder metallurgy proces, wich helps to reduce the grain size of the chromium carbides, making the material tougher.

I tested the grinding action on the various steels with this setup:
Test steels sharpening small.jpg

The tested blade is set in a jig, weighted down with a carefully positioned weight, and pulled 10 times over a piece of 3M lapping film of 9 um (micrometer) AlOx grit. For each test a fresh piece of lapping paper. I started with a sharp edge for each test. Then I looked at the length of the grinded area under a microscope. Repeated the test a couple of times of course.

This is the measured length:
O1 89 um.
A2 74 um
CTS 64 um

But this is just one dimension. When trying to remove metal by grinding, you need to remove a volume of steel. When the length of the removed area increases, the volume of the removed material increases incrementally. We all know this when regrinding a bevel to a new angle, at first it seems to happen very quickly but when the length of the new bevel increases the work becomes slower and slower.

So, after calculating the removed volume of tool steel on my test lap, I found that:

CTS-XHP grinds about twice as slow as O1 and about 30% slower then A2.

[Kees Heiden]

The next test was planing on a narrow strip of wood, either oak or beech with the CTS-XHP steel and O1. Then I looked with a microscope at the length of the wearbevel at the underside of the edge. I did a bunch of tests and got rather confused, but in the end I could spot a certain trend.

The wearbevel on the CTS-XHP steel didn't grow as fast as on the O1 steel. So the claim from LV that this steel is more wear resistant sems to hold true. There are some caveats though. But here is a diagram of one of these tests:

Beech wearbevel length.jpg

Blue is O1 steel. Red is CTS-XHP. Planing length in meters is horizontally. Length of wearbevel is in micrometers vertically.

I don't think that this test is a very good value for the overal wear of a tool edge in a handplane. It only meassure one dimension in a rather complex shape.
Here is a picture of a wearing edge.
Worn edge.gif
The length Tb is what I was measuring during this test with the microscope. The wearbevel under the edge is the loss of clearance angle we know to be the root cause of handplane wear. When the edge doesn't want to "bite" into the wood anymore, we want to sharpen again. The more pronounced this wear bevel is, the more the upward force on the edge will be and the less the bite of the edge into the wood.

With this test I only meassure the horizontal length of this wearbevel, while the height of the concavity of this wearbevel would be more interesting. The higher, the more you loose clearance angle. But I can't meassure the height. Another thing is the edge retention Te. This determines how obtuse the edge becomes when looking at it from the front.

When I look at the diagram above of my measurement, I see that O1 has a steady increase in wear bevel length. But CTS-XHP seems to reach a plateau. I could imagine how at this point Te and Tb a re playing together. Tb increases, but Te increases at the same arte, so the result is a steady length of Tb.

So, for sure, CTS-XHP wears slower then O1 (no surprise with all those chrome carbides) but I wasn't quite sure how good this diagram represents real edge wear.

[Kees Heiden]

So, I fired up my plane force measuring jig.

foto 2.jpg

With this setup I can measure the vertical and the horziontal forces on a plane blade and see how advanced wear results in a change in these two forces.

The horizontal force in this case was not really interesting and the change too small to conclude anything. The vertical force though is very interesting. It is directly linked to the loss of clearance angle. When the clearance is lost, the edge is being pushed out of the wood, exactly what we want to know.

So, here is my diagram:

Oakbeech vertical force.jpg

This diagram needs some explanation. You see the measured force vertically and the planing distance in meters horizontally. CTS-XHP is still called PMV-11 in this diagram. There are measurements for oak and beech, so the diagram is a bit crowded.

When you compare the two lines for beech and the two lines for oak seperately, you will see that force increases about the same for both tools steels over a length of 100 meters. That's surprising, because it suggests that for these two rather bening kinds of wood the wear rate is the same for the two tool steels.

I should have meassured for a lot longer but I didn't. I also should have done a similar test with teak or so, a wood with high silica content. But I didn't do that either.

For a smoothing plane, 100 meter planing distance is quite a bit, you'd want to sharpen as soon as the edge looses much of its original sharpness. But for a jack plane it is just getting started! So it depends on the circumstances how valid this data really is. If the wear curve of CTS-XHP really flattens out after a certain distance, while the O1 curve keeps on steepening, then I think that this wondersteel performs best on abrassive kinds of wood and more in roughing operations rather then smoothing.

But these are just ideas and really no conclusions. The whole area of edge wear in handplanes is rather subjective anyway.

[Normand Leblanc]

Kees,

What kind of microscope do you use to measure the wear bevel?

I'm sure this post will draw a lot of attention.

Normand

[Kees Heiden]

Well I am not so sure anyone is interested in some independend information about PMV-11 overhere! ��

My usb microscope has 470 x magnification and an adjustable polaroid filter to remove the glare from the metal. It wasn't the cheapest one on the market but it ain't truelly professional either. It is made by Dino-lite.

[george wilson]

I am grateful to have found out about the composition of PM VII,as I have a natural curiosity about any kind of tool steel,and do use several of them. But my main thought is I LOVE MY PM VII BLADE!!!

I read the link to the patent data in the other,now closed(?) thread. There seems to be a good deal of possible variations in the PM's content.

[Patrick Chase]

I am grateful to have found out about the composition of PM VII,as I have a natural curiosity about any kind of tool steel,and do use several of them. But my main thought is I LOVE MY PM VII BLADE!!!

I read the link to the patent data in the other,now closed(?) thread. There seems to be a good deal of possible variations in the PM's content.

The patent "stakes out" a range of PM stainless steels, of which CTS-XHP is one example. That's pretty common practice, as it protects against the case where a competitor very slightly tweaks the composition to evade the patent. Here are the composition nominals per Carpenter: https://cartech.ides.com/datasheet.aspx?I=101&E=343.

Agree w.r.t. PM-V11 blades. They work very well for me.

[Chuck Nickerson]

Perhaps I missed it as the other threads blew up: how did you equate PM-V11 to CTS-XHP?
(This is a sincere question; I'm not trying to destroy the thread.)

[Patrick Chase]

Perhaps I missed it as the other threads blew up: how did you equate PM-V11 to CTS-XHP?
(This is a sincere question; I'm not trying to destroy the thread.)

XRF. Strictly speaking what we know is that the non-Carbon constituents of PM-V11 are a very close match to the nominals for CTS-XHP.

As I mentioned in the other thread, CTS-XHP was out of patent as of 2013, so it's possible that PM-V11 is a compositionally similar steel processed by somebody other than Carpenter, i.e. a "generic". If that's the case then it would be legally incorrect (due to trademark) to call it "CTS-XHP", just as it's legally incorrect to call your drug store's house-brand Acetaminophen "Tylenol".

Thinking about this yet more, another possibility is that Carpenter is processing PM-V11, but are selling it to LV under a contractual provision that prohibits LV from using the CTS-XHP/440XH brands. Again there is precedent in the pharma industry - makers routinely sell un-branded, lower-priced "generic" versions of their own off-patent drugs. A recent highly publicized example is Mylan's decision to offer a ~half-priced un-branded version of their Epi-Pen.

[Mike Henderson]

I sharpen by putting a primary bevel on by machine and doing a secondary bevel on water stones. Since the secondary bevel is quite small, I can put it on quickly and easily for any steel that I've encountered in chisels and plane blades. So "grinding time" is immaterial to me. What I'm interested in is how long the edge lasts in various operations.

My experience is that the PM-V11 edge lasts longer (usable) than other steels. The hard steel of Japanese chisels comes close.

Mike

[Mark AJ Allen]

Is there any way that these tests can be correlated to the testing that was done by Veritas? Has anyone ever attempted to repeat or reproduce Veritas results? From a scientific POV, that's the hallmark of valid experimental results.

http://www.pm-v11.com/TestingIntroduction.aspx

I would hypothesize that if any steel was compared to a Veritas PMV11 under the same controlled conditions in experiments, and resulted in statistically equivalent results, then you could say the materials behave the same under the factors being investigated. I'm not sure what conclusions to draw from the results presented in this thread ... are they in agreement with Veritas claims?

[Derek Cohen]

Mark, I was one of the testers leading up to the release of PM-V11. I do not have the actual data to hand, but the results were similar to those published by Lee Valley. My method was less sophisticated - using a variety of steels, I planed wood until the blade was dull, and counted strokes. It helps to use wood that has a high silica content!

As I mentioned in a previous thread, the buying public are little interested in the composition of the steel, and more focussed whether the manufacturers claims can be supported. I liked the analogy of asking the chef for the ingredients of the meal before ordering. There are some - my wife is one - who read cook books for fun, and claim they can taste the dish on the page. Most people are content to enjoy the meal in the restaurant.

Regards from Perth

Derek

[Larry Frank]

I may have missed it in one of the posts. But where did someone find the composition of the PM-V11? Is it published somewhere or did someone have it analyzed. Could someone provide a link to it. Just saying that it is the same as the CTS-XHP, really needs a bit of backing.

[Jim Koepke]

I may have missed it in one of the posts. But where did someone find the composition of the PM-V11? Is it published somewhere or did someone have it analyzed. Could someone provide a link to it. Just saying that it is the same as the CTS-XHP, really needs a bit of backing.

My understanding of the claim to them being the same is that they have similar chrome contents.

My understanding is also that there is no source that definitively claims they are the same material, it is speculation based on comparisons of published material on the two metals.

Maybe my ability to be pleased is too simple. It doesn't matter to me if there is some special secret ingredient. How the metal performs or in the example Derek offers, it is how a meal tastes that maters to me.

jtk

[Kees Heiden]

Several people have PMV-11 samples diagnosed. Patrick from this place, I have a friend who works on a technical university and he knows the right people. I know from another American guy who had a sample analysed and agreed with our results. This analysis is being done with an XRF machine like Patrick linked to above. This is a non destructive test. It's very simple and easy once you have acces to such a machine. It gives you the percentage rates of all the alloys in the sample except carbon. There is something with the size of the carbon atom that makes it hard to escape the steel matrix.

From my university friend I understand that it is really trivial thing to do, not any harder then putting the sample on your kitchen scales to determine the weight. You could also search for a company that does this as a service, these machines are everywhere where they are interested in the composition of things. It'll cost you a few bucks of course.

I was interested in this because of a weird contradiction in LV's claims. They claim that it is very easy to sharpen but manages to wear a lot slower then O1 and A2. Because those are essentially the same mechanisms, that's not very plausible. So. I looked at the composition of the steel and saw that there is a lot of chromium, but the rest of the makeup of the steel isn't that different from A2. Is the powder metallurgy really that magical that it can make the chromium carbides somehow softer on the grinding stone but plenty hard when planing abrassive wood? Or is there another wear mechanism going on, like corrosive wear? The chromium makes the steel stainless, so I suspect that it would help against corrosive wear, if that is an issue in handplaning.

Here is what LV claims about the sharpenability of their steel:

The blades were evaluated for ease of sharpening based on the effort required to obtain a specified level of smoothness (measured as roughness). For presentation purposes, the top-scoring material was assigned a score of 10 and the lowest-scoring material was assigned a score of 1. Intermediate-performing materials were assigned scores on a relative linear basis. (Note that this does not mean that the top metal scored 10 times better than the lowest metal.) While all blade configurations were assigned code letters for testing, we have revealed the scores of O1, A2, M4 and PM-V11 alloys. Based on the results of the previous tests, certain blades were excluded as we determined they were not compatible with our chosen sharpening medium. Material Ease of Sharpening Test Score O1 10 PM-V11 6.5 A2 6 Y-1 6 Y-2 6 Z-1 5.5 Z-2 5.5 X 5 R-1 3 R-2 3 W 3 S 2 P 2 M4 1 N1 1 N-2 1