saw plates old VS new

[Rob Streeper]

Okay, here is some more data. In a nutshell I took two pieces of 1095 0.035" thick and did some testing on them. The hardness of the first piece was measured in it's factory state - bluing and all. The other larger piece was divided into two roughly equal sections. One section was hammered on an anvil. Then both sections were ground just enough to remove the bluing. Then I did N45 testing on all three, fifty tests on each specimen for a total of 150 measurements. Here is the data and the first set of derived stat's. I've posted a full analysis of these results here at post #103: http://www.woodworkforums.com/showth...=182165&page=7

50 point measurements of 0.035 1095 steel_Page_1.jpg

50 point measurements of 0.035 1095 steel_Page_2.jpg

[David Weaver]

One would assume that the hardness difference is superficial, or the 1095 would lose its spring at 60 hardness.

Presuming you have coil at the thickness of carpenter saw plates, it would be instructive if you could find someone with a C tester to strike the same hammered areas you've tested with a superficial tester and converted to C values.

Hammering will likely have the same effect on a device created for superficial testing as would case hardening, and any rolling or hammering that was done to tension saw plates would've been done to increase tension, but not to increase through and through hardness, or the saw plates would've lost their spring and snapped.

[Rob Streeper]

C tester calibration data 011415a_Page_1.jpg

C tester calibration data 011415a_Page_2.jpg

C tester calibration data 011415a_Page_3.jpg

C tester calibration data 011415a_Page_4.jpg

C tester calibration data 011415a p 5.jpg

C tester calibration data 011415a_Page_6.jpg

C tester calibration data 011415a_Page_7.jpg

Same samples of 0.035" thick 1095 analyzed above on the N45 scale tester were analyzed here on the 150kgf Rockwell C scale tester. The results for each testing method are within the standard deviations.

[george wilson]

Since the available spring steel,I am told by a very reliable source,is now coming from India,and subject to varying quality and hardness,it might not be really meaningful to attempt to put down figures for the hardness of spring steel.

I am glad I'm as old as I am,since the World is suffering from a decline in many things I value. Things like good files,good spring steel,good mahogany to name a few.

[Patrick McCarthy]

"is not coming from India" . . . . .George, did you mean " is NOW coming . . " ??

[george wilson]

Sorry,typo. Wife was using a recipe on the computer and I was rushed. So the answer is Oh yes indee-de.

[Rob Streeper]

Since the available spring steel,I am told by a very reliable source,is now coming from India,and subject to varying quality and hardness,it might not be really meaningful to attempt to put down figures for the hardness of spring steel.

I am glad I'm as old as I am,since the World is suffering from a decline in many things I value. Things like good files,good spring steel,good mahogany to name a few.

George,

I wouldn't be surprised to learn that. The steel from different lots does seem remarkably consistent. I know ISO QC procedures are widely adhered to but it is surprising that there's not a little more spread. Sorry if I irritated you - that was never my intention. Controversial results always stir strong emotions, that's why I went to the efforts that I did so that I could make as good a case as my means allow.

Thanks,
Rob

[Joel Moskowitz]

While there might be some 1095 coming from india, i assume that what you mean by "spring steel" stuff comes from all over and in many qualities. there is even a1095 rolling mill in NJ. sandvick is a big supplier too. i dont thin i have even seen samples of 1095 that would lead me to suspect india. so don't give up hope.

[Rob Streeper]

While there might be some 1095 coming from india, i assume that what you mean by "spring steel" stuff comes from all over and in many qualities. there is even a1095 rolling mill in NJ. sandvick is a big supplier too. i dont thin i have even seen samples of 1095 that would lead me to suspect india. so don't give up hope.

Hi Jim,

I've been sourcing Lyon brand steel from Zoro tools. Relatively recently they started selling it in a very wide range of sizes in smaller quantities.

Rob

[Tony Zaffuto]

Some comments: ISO stipulates documentation developed by the company seeking certification, must follow a standardized order, but not the methodology used. Further, the quality of the underlying company is only as good as the registrar that audits the company to determine compliance to the standard. A company can be ISO registered and not have any procedure in place that describes use of hardness testing, let alone specifying hardness scales, or selection of proper equipment to the material to be tested.

As far as hardness testing, variability in readings can happen and again is determined by the consistency of the operator's methods as well as the equipment itself (not to mention proper use).

I am not a QC tech/QC manager/Director of Quality or any other title associated with the quality department in the manufacturing plant I own, though we do have a dedicated quality lab, with 4 employees administering our quality program and directives (in excess of 70% of our sales output goes into new cars).

[Kees Heiden]

I just did a simple test. I have a small piece of 1095 steel, 0.6 mm thick. One side I hammered with a 600 gram hammer on an anvil. The other side was left as it was. Then I took a file and filed small triangles in the sides of the piece, 20 file strokes per attempt.

First, the feel. The first strokes on the hammered side feel a little bit harder, but I don't know if that feeling would be reproducable in a blind test. Further down I am not able to feel any difference.
Then measuring the depth of the triangles. I can't really measure any difference. Some times the hammered side goes a little deeper, sometimes the other side.

Conclusion: I am not really able to feel much difference between hammered and unhammered springsteel. And the file dives into the steel just as fast.

[David Weaver]

Conclusion: I am not really able to feel much difference between hammered and unhammered springsteel. And the file dives into the steel just as fast.

That being because Rob is using a tester that tests surface hardness, apparently. If you were using 1095 and increased the hardness five points, you would be able to tell when filing.

Thanks for doing that test for us, it confirms my earlier comments.

[Kees Heiden]

In the last test Rob is using a real Rc tester.

My quick little test obviously doesn't really proof much. It proofs mostly that I don't have very sensitive fingers.

[David Weaver]

Controversial results always stir strong emotions, that's why I went to the efforts that I did so that I could make as good a case as my means allow.

Let's examine why that is. Here's a quote from your first post:

Further, modern custom saws made with raw 1095 steel are not as hard as are older Disston saws. If the plate hardness is important and if Disston optimized his products we've actually taken a step backward in our unquestioning use of 1095.

The entirety of the "controversy" is over that incorrect quote and the data that you've suggested that proves it. You have tested old saw plates under the handle, but not at the tooth line. Subsequently you admitted those saws can be filed, which proves (whether you like it or not), that they are softer than the 58 values you show in your table (substantially so unless you can only file a few teeth before ruining a file), and you stated that you wouldn't damage your saws by striking the tooth line.

The tooth line is where the hardness matters.

What you have done is akin to striking a chisel at a tang or a plane iron at the wrong end, then proving that there's little variability between your strikes at the wrong end and then declared something about plane irons based on those statistics.

I have offered to help you find some vintage saws that you can strike at the tooth line, and I have not heard anything back from you. Your conclusion will change because people have studied this before, including the old tools list individual who struck saws with a C tester under the supervision of an engineer who uses the same device.

You have not reconciled with the fact that your conclusions is either wrong (if the data strikes are not accurate due to a limitation with your tooling testing surface hardness) or misleading (suggesting that a saw that might be overhard under the handle but not at the tooth line provides relevant data for your conclusion if you choose to take the higher value under the handle).

I am challenging you to get it right, as you state that you have done the best you can, but that is not true. Doing the best you could would be guarding the quality of the data that you're sampling in the first place, and taking strikes under the handle of a saw is not your best effort.

Since it is already well known that vintage saws are not harder than 1095 that is shipped in the 50 hardness range, I will continue to challenge you until you either understand why the data isn't useful and your conclusion is misleading (and certainly casts aspersions on folks like ron bontz and others who make fine saws - none of whom I have ever bought a saw from nor do I have any relationship to) and admit that it is, or you do something to get a more relevant reading, and then incorporate that into drawing a conclusion.

[Rob Streeper]

Some comments: ISO stipulates documentation developed by the company seeking certification, must follow a standardized order, but not the methodology used. Further, the quality of the underlying company is only as good as the registrar that audits the company to determine compliance to the standard. A company can be ISO registered and not have any procedure in place that describes use of hardness testing, let alone specifying hardness scales, or selection of proper equipment to the material to be tested.

As far as hardness testing, variability in readings can happen and again is determined by the consistency of the operator's methods as well as the equipment itself (not to mention proper use).

I am not a QC tech/QC manager/Director of Quality or any other title associated with the quality department in the manufacturing plant I own, though we do have a dedicated quality lab, with 4 employees administering our quality program and directives (in excess of 70% of our sales output goes into new cars).

Hi Tony,
Yeah, I know about the issues with Quality Systems. I once heard someone say that, if a company made life preservers out of bricks, the only thing that the quality system would do would do is ensure that the bricks were all of uniform size and weight.
Hardness testing is fraught with complications, that's why the NIST best practices guidances suggest the testing procedures that they do. That variability and uncertainty is precisely the reason that I have used my efforts to produce what I have posted here and on the Woodwork Forums.
Disston's 52 has been discussed a lot but tested very little. There is a little information available on the web but I haven't found anything with the rigor of my study. I know my work here has limitations but I argue that some information is better than virtually none.
I've taken a lot of heat over my refusal to reveal the identity of Maker 3 but I think my reasoning is sound. I'm not out to damage anybody's reputation but I do feel that the saw making community can do better and I do feel that the saw consuming community needs to be warned to be careful. This problem is not by the way the only problem I have found in custom saws, a couple more have come to light.
My feeling is that a manufacturer has a duty to meet customer quality expectations no matter how small the manufacturer might be. Hidden defects are a big problem and nobody will find them without doing some testing.
One other poster said that he had observed inhomogeneities in a particular roll of saw blade steel he had. I think that information should come as a warning to everybody who makes saws for sale at the very least. If there's a problem with the raw material stream we should all be testing somehow be it with files alone or real hardness testers like I have used. Otherwise the buyers of these saws may decide that we don't really care about them.

Cheers,
Rob