Thread: Saw File Hardness Measurements

I've got Marv's report but that's all.

Originally Posted by FenceFurniture

I've got Marv's report but that's all.

I'll get on the grinding of teeth today. I may have some data today or tomorrow, and I still owe you that spreadsheet. This year has been filled with a lot of other people's broken stuff and some of my own of course.

Here's an example of how I ground the files. The Tormek made short work of them. Now to testing.

Each was ground by running the file from tip to tang over the stone rotating parallel to the long axis of the file until all visible traces of the teeth were removed in an area large enough for five replicate tests.

file grinding example.jpg

Here's the calibration data - low end drift as before but for the relevant range everything looks good.
Saw file hardness calibration data 051815a_Page_1.jpgSaw file hardness calibration data 051815a_Page_2.jpgSaw file hardness calibration data 051815a_Page_3.jpg

Here's the analysis of hardness, the standard deviation of the hardness and user ratings. No correlations observed.Hardness and St dev hardness vs user rating data and plots 051815a_Page_1.jpgHardness and St dev hardness vs user rating data and plots 051815a_Page_2.jpgHardness and St dev hardness vs user rating data and plots 051815a_Page_3.jpgHardness and St dev hardness vs user rating data and plots 051815a_Page_4.jpg

No correlation between hardness and the standard deviation of the hardness either.Analysis of hardness vs stdev of hardness 051815a_Page_1.jpgAnalysis of hardness vs stdev of hardness 051815a_Page_2.jpg

Again, citric acid treatment appears to do nothing.

Effects of citric acid treatment on files 051815a.jpg

Here's the raw and calibration corrected data.Raw and calibrated data saw file hardness measurements 051815a_Page_1.jpgRaw and calibrated data saw file hardness measurements 051815a_Page_2.jpg

A grand job Rob - thank you very much for your efforts. Just to make it simple for the non-expert, could you put up a post of the "money" chart - the chart that is the most relevant?

The take home messages are, within the dimensions and limitations of this study:

1) The hardness of saw files, assuming they are 'hard enough' does not drive user ratings, i.e. users do not feel that harder files are better than soft files.

2) The variation in the hardness of saw files, again assuming that they are hard enough, does not drive user ratings. Users don't feel the variations in hardness or users don't care.

3) There is no correlation between the hardness of saw files and the variability in the hardness of those files, i.e. harder files are not more or less consistent in hardness than are softer files.

4) Citric acid treatment does nothing for saw files except etch them and users don't like (non significant dislike) etched saw files.

ALL of these statements carry the warning: "This is a study of a few users using a single file of each type and thus the results presented here are not definitive or generalizable."

Further, all of the caveats of post 24 apply here equally.

Originally Posted by FenceFurniture

A grand job Rob - thank you very much for your efforts. Just to make it simple for the non-expert, could you put up a post of the "money" chart - the chart that is the most relevant?

Thanks FF. I really wanted to move this project along. Those files have been sitting there looking at me making me feel guilty. I'll endeavor to drag the Foley filer out and run the new set of files to death as described above. Thanks for organizing and financing the effort.

It's been a lot of work but even a null result is still worth having.
However, it's limited to new files.
A very important question for users of files is also "What happens over time".
Do the harder ones last longer or do they loose teeth more quickly than they maintain their sharpness and VV for the
This is where metal toughness kicks in, maybe toughness is the real issue for long term users satisfaction?

Do you still have the original file with some teeth on them, a microscope on the teeth may reveal
% of teeth broken or % lengths of breaks on teeth.
Degree of sharpness glint on teeth etc?

Time isn't really a factor Bob. Of the current lousy standard, one file is lucky to sharpen 2, perhaps 3 saws - essentially using one corner for a saw, and may have to break into the second corner to finish it.More usually though, one file can't even sharpen one saw.....

Ain't no such thing as an old (still in use) saw file - not these days anyway.

Originally Posted by BobL

It's been a lot of work but even a null result is still worth having.
However, it's limited to new files.
A very important question for users of files is also "What happens over time".
Do the harder ones last longer or do they loose teeth more quickly than they maintain their sharpness and VV for the
This is where metal toughness kicks in, maybe toughness is the real issue for long term users satisfaction?

Do you still have the original file with some teeth on them, a microscope on the teeth may reveal
% of teeth broken or % lengths of breaks on teeth.
Degree of sharpness glint on teeth etc?

Hi Bob,

I think that the most important result here is the lack of association between hardness and user perception, another myth dispelled. This data tells us that modern files are at least hard enough to do the job, within its limitations of course.
If our prior testers reflect reality accurately the failure of modern files relative to old files is therefore not due to changes in the heat treating. Elimination of hardness as a suspect reduces the number of potential causes that our dissatisfaction can be ascribed to. Some other factor(s) must be responsible. Additional possibilities off the top of my head are:
1) Uniformity of toothing
2) Is a less appropriate alloy being used?
3) Is there a surface treatment that is being used or perhaps that was being used that has changed? Carburizing or nitriding perhaps?

I have the files and will look them over with a magnifier. I only removed about an inch of tooth on each. Toughness is also on my mind.

Cheers,
Rob