Standard Reference Material for Hanging Hair Test
A lot of the criticism about the use of the HHT to judge razor sharpness has been in the area of the incomparability between people, because the hairs used are different thicknesses, have more or less oil content, etc.
I've been thinking about how to instrument a sharpness test in order to get an objective, reproducible measure that will work at any stage of honing.
To this end, I've looked for replacements for human hair.
I focused on monofilament thread, as this would eliminate structure issues. I got some "invisible thread" in bobbins, with a thickness of 0.004" ( 0.1 mm). Unfortunately this is too light to let any razor I've sharpened cut it in a simple HHT. I will still try to use in an instrumented test or when I get a Lynn-honed razor back.
Another purchase was monofilament thread used in tying flies for fishing. This is Danville's Monofilament Fine, which is 0.006" (0.15 mm) diameter. It's available on ebay at
Danville Monofilament Thread Fine - Fly Tying - eBay (item 230507649553 end time Oct-03-10 14:34:03 PDT)
(auction #230507649553)
It's only $1.09 per 100 yd spool with $2.99 shipping. (I bought 6 spools to average out the shipping.) You might be able to buy it locally at a sports supply store.
This Danville's thread behaves very similar to the human hair I've been using for the HHT. It is slightly more resistant to cutting, but very close in properties.
The only issue is that it's clear, so somewhat hard to see in use.
The good news is that HHT results with this thread can be compared among members here with some idea that similar results should be expected.
I am continuing to look into colored versions of such thread, and into a possible test jig to standardize applied force or allow quantitative measurement.
First crude results using the FCT
I have two razors (both Fermaruds) that are clearly not shave ready yet, despite my first attempts at bevel setting and polishing. They both tug severely in shaving.
I have two more razors that shave acceptably well (a Fromm 72R and a Damisons).
I also have a new Supermax DE blade held in a hemostat to simulate a razor.
In my version of the HHT using my wife's hair, the DE blade cut hair easily 4 of 5 times, with one time failing. The two acceptable razors cut hairs in most spots along the blade. The two duller razors cut hair only occasionally only at a particular point on the blades. Thus the HHT correlates well with my idea of shaveability. (I strongly advise people who use the HHT to test their procedure on a new DE blade to calibrate what "sharp" means.)
I made my first crude test jig yesterday. All it was was a piece of 1x4" wood with a 2" equilateral "V" cut into the top and two 7/64" holes drilled 3" apart across the "V" on the top edge. One inch down on each edge I drilled a 7/64" hole on either side. The 4x4" chunk stands on its unmodified edge.
I put two small iron nails into the top holes as capstans, and used the side holes to pin in the thread with matchstick shafts.
The thread used was Danville's fly tying monofilament (0.006"), which acts close to what a coarse human hair does with razors. I chose this thread for several reasons: 1) monofilament, with simple surface structure; 2) diameter within the range of human hair, so relevant; 3) cutting force roughly similar to human hair; 4) commonly available and cheap. The only problem with it is that it's clear, so it's hard to see and therefore work with.
I pegged down one end of a piece of filament in one side, looped one full turn on the first capstan about 3/16" up from the wood, crossed the gap taut and did a corresponding loop on the same side of the other capstan, and then down to the other hole and pegged. The filament does not touch the wood even during the test.
I kept the filament taut rather than slack, mostly because it was easier to do.
I placed the block on my 400 x 0.01 g Ohaus digital scale. The Ohaus appears to update about once a second. I tared with the block and then pressed each blade down perpendicular against the filament as slow as I could.
The DE blade averaged about 12 g force. The Fromm about 14 g and the Damisons about 17. The first Fermarud as 25 g and the second 28. The results seemed more consistent at the higher end than the lower, perhaps because of the suddenness of the breakage.
From this simple pilot project I learned several things:
1. The usual HHT is qualitatively useful, particular with the comparison to the DE blade.
2. The FCT quantitatively supports the shaveability and HHT results.
3. My honing skills are not as bad as I thought: The Fromm is close to the DE sharpness.
4. The quantitative measurement of sharpness does not appear to be difficult to carry out.
5. I've got the right filament to use in testing.
The next step is to automate the test so that constant deformation rate occurs, as opposed to trying to maintain this by hand. And to use a force gauge that returns peak force, rather than trying to catch it on the scale.
3 Attachment(s)
Version 2 of the FCT and new test results
Revision 2 of the FCT ("Filament Cut Test").
This is a piece of wood 3.5" long, with two 4d nails at 0.5" and 3.0". I also bored two 3/32" holes on the ends for a piece of round toothpick to peg down the line. There is a center 3/16" hole to screw the 10x32 bolt end of my DFI2 force gauge into.
My DFI2 gauge remembers peak force in grams (unfortunately no decimal). It also has a computer interface. It's measurement rate is 400 sps, with the meter reading an average of 40 samples and the peak reading instantaneous. I will connect this up later and get some force-time curves. (If I connect to the motorized test stand, I can get force-deflection curves.)
The filament deflects a cm or so before breaking, so I don't think slack is a big issue.
I carry out the rev. 2 test as follows: Peg the filament down one side. Draw taut up to the first nail and wrap with two loops. Pull taut over to the 2nd nail and do two loops. Pull taut down to the other edge and peg down it down. Zero the meter. Then take the razor and press slowly down perpendicular to the filament at the center of the span. After fracture, record the peak value captured.
With this setup, I obtained the following values:
Black (Fermarud) razor:
After honing, before stropping: 36 g
After stropping: 29, 30 g [previous jig: 28 g]
Hanging hair test ("HHT"): Ok to Good (50-75% success)
Shave test: Poor (snags on chin)
(There is a known honing problem with this razor I haven't fixed yet: It doesn't lie flat at the heel.)
White (Fermarud) razor:
After honing, before stropping: 36 g
After stropping: 21, 23 g [previous jig: 25 g]
Hanging hair test ("HHT"): Very Good (100%)
Shave test: Fair (some snagging over chin)
SuperMax DE blade, new: 14, 16 g [previous jig: 12 g]
The first and second versions of the FCT were done with different jigs and different spools of filament.
Conclusions:
1. Good reproducibility, particularly given hand driven deflection, different spools and different test jigs.
2. Repeatability appears to be a standard deviation of about 1.6 g, or a relative standard deviation of 11% for the lowest value and 6-7% for the highest values. Appears to be a constant error rather than a constant relative error. This should clarify in more extensive testing.
3. Correlates well with my HHT.
4. Correlates well with shaveability.
5. I'm happy with the current jig. It appears to satisfy my needs with a minimum level of effort. It is a little time-consuming reloading filament. I don't have any good ideas on how to speed things up, other than having a better lockdown arrangement.
See pictures attached.