Stiffening, the next chapter of my questions about Disstons' methods of sawsmithing

An accident in hammering a blade has propelled me to look into the next most mysterious step of the manufacturing process used at the Disston works for hand and panel saw blades. For those who've forgotten here are the steps:

Disston handsaw blade manufacturing steps (after blanking)

1. Toothing
2. Hardening
3. Tempering
4. Smithing
5. Taper grinding
6. Checking
7. Tensioning
8. Fine grinding
9. Blocking
10. Polishing
11. Stiffening
12. Etching
13. Setting
14. Sharpening
15. Fitting the handle

I've already undertaken and published here the hardening and tempering, taper grinding, tensioning and blocking steps of this process and the next is #11) Stiffening.

The Disstonian institute (Online Reference of Disston Saws -- Keystone Saw Works) gives this description of the process:

"..."Stiffening." As to this latter, the different processes and hammering under which the blades have passed, has altered the arrangement of the molecules in the metal and in order to restore the desired qualities and spring they are stiffened in a special bath, which was originated and is known only to Disston."

This part of the process hasn't been discussed very much on the Internet other than occasional mentions.

Back when I was building this saw I also made a couple of additional blades.

Attachment 378728


I over hammered one of the blades and created what the autobody men call a 'can condition'. This can condition corresponds nicely to the written descriptions of 'fast and loose' as the term is applied to hand saw blades. Essentially the blade acts as if it has a joint in the middle allowing it to flop back and forth between two un-sprung positions of near equilibrium. By judicious hammering and grinding I've managed to remove the majority of this defect from the blade but not eliminate it entirely. A similar condition was present in the blade of the saw that I tried to work out during our recent trip to Houston. Unfortunately the noise was intolerable and I had to stop before finishing the work (see https://www.woodworkforums.com/f278/s...mmers-203451/9 at post #122).

A conversation I had with a former QC lab manager for several major tool brands at the LN HTE in Dallas included a discussion of this very Stiffening step. We discussed the possibility that the stiffening was carried out in a molten cyanide salt bath, inducing a kind of case-hardening.

On further reflection I realized that I'd never noticed any evidence of a case on Disston saw blades and I'd never read anything suggesting that the blades were case hardened. This lead me to surmise that, if it was a salt bath, the purpose was a heat treatment, not case hardening or cyaniding.

While doing the heat treating on the sawsmithing hammers I thought to pull this canned blade out to test the effects of prolonged heat treatment at a temperature corresponding to the drawing temperature of 1095 that would give a hardness of ~HRC50. This number is about 695 ^oF so I set the oven up and loaded the blade in. Testing revealed that the blade had not changed in hardness.

Attachment 378733

The blade came out with a much reduced 'can condition', a minor bit was still evident. It seemed much stiffer though. I haven't done any before and after measurements but the blade definitely feels stiffer. It also has a nice blue color. I've taken several pictures against different backgrounds to give you an idea.

Attachment 378732

Attachment 378731


Attachment 378730

Attachment 378729

Fresh from the oven the color was a very light blue, it's darkened on sitting some but Paul's post here (https://www.woodworkforums.com/f152/s...story-188082/9 @#133) discussing blue Simonds blades made me think that perhaps both manufactures were doing a post-grinding/coldworking re-heat step to induce the spring.

Now to do a before and after study...

Rob

In trying to determine Disston's technique from over 100 years ago we also have to consider what was available to them in terms of technology.

A second issue is that whatever they did, it may have had no appreciable effect and was at best a clever marketing ploy. Disston were probably as good at marketing as they were at sawmaking.

I don't know how the blue was applied to the Simonds saws, but I would be surprised if it were a result of tempering. I am thinking in terms of a bonded finish. It had to be something that did not create friction.

This link may be useful :

Heat Treating Glossary of Terms | Solar Atmospheres

Your Alamo handsaws look very good by the way.

Regards
Paul

Thanks Paul,

I thought about the surface finish aspect too. At 100 years ago the only finishes I'm aware of that were commonly (and relatively inexpensively) applied to metals were electroplates, color case hardening, phosphating (Parkerizing and the like) and heat/fire, niter (hot) and rust bluing. Teflon finishes weren't widely available until after WW2, in fact I have a Teflon coated Disston saw. The German chemical industry was approaching its zenith but things like powdercoat, VPD & etc., nitriding and so on are much more recent techniques as far as I know.

If it was marketing faff it is impossible at this distance lacking documentation to tell.

Nonetheless, the reheated blade does feel different. I hammered it some today and got the can condition resolved but it now has a bow - later.

Cheers,
Rob

Mentions of tensioning patents ...

OldTools Archive

Attachment 412193

Looks like the alloy is work hardening....

I think what your experiment shows is that re-tempering the steel to stress relieve it a bit doesn't hurt the hardness at all...

It wouldnt surprise me - this work hardening effect... Perhaps by refining the grain a bit - the hardness increases slightly..

One caution on your statistics... Be careful about inferring your change in process yielded the results when you may be really measuring an effect of something else... 10 hardness measurements all over the same workpiece before vs after can tell you that maybe something happened and that it is consistent across the piece... But since its measurements of the same piece during a one time in a row treatment - you can't infer that the change you made is what necessarily Caused the difference... (Though in this situation it makes sense that your process probably did)

That analysis happens over multiple test runs on different pieces before vs after... I have run into situations where it turned out that all the knobs we turned ended up as "insignificant noise" and all the actual "process change" was occurring because of some other environmental variable that we had not accounted for... For example - the state and operation of automatic oilers on compressed air systems can have very surprising/significant effects on a downstream process. DAMHIKT....

The effect of hammering is pretty well clear in my work to date, it raises the measured hardness by about 4%. I've reproduced that effect on multiple blades. This thread https://www.woodworkforums.com/f278/h...lates-182165/7 beginning at post #103 describes my experiments with hammering.
The effect of re-heating has been measured on this single blade at this point but this line of inquiry will be extended...

For purposes of comparison, here are my results from analysis of No. 12 saws.
Attachment 412231
The standard deviation of my experimental blade above is higher at 0.24 than that found for the No. 12's at 0.17 but it should be kept in mind that the experimental blade is not yet polished as are the No. 12's and that the roughness of the experimental blade contributes to its' higher S.D.

The effect of the surface roughness on the S.D. of the hardness measurements can be seen in the difference between the Raw with (S.D. = 0.50) and Ground and Hammered blades (S.D. = 0.69). This effect is more thoroughly addressed in the above linked thread.