Official Luthiers Forum!

I have plenty of real world experience with different fret wire makes and alloys, but have long wanted some more controlled testing data to show solid comparisons in wear under the same conditions. This is just a beta version (threw it together in little over an hour this afternoon), but with a few refinements shows promise to yield good results. Stay tuned...

_________________
Eschew obfuscation, espouse elucidation.

Clever!
It will be interesting to see the results.

That is clever and looks like as good a simulation as you could get. I hope really hope you do Evos Gold David. I tend to use it quite a bit and from my what I see on my own guitars( always use it on my own) I will make a guess. Twice the life.

Thanks,
Danny

Danny: I'll up you on that and say at least 3 times and likely more.
Tom

_________________
A person who has never made a mistake has never made anything!!!

Oh great now I have to listen to that thing all day long...... JUST KIDDING!!!!

At least David won't be making me fret on and off all day long for weeks on end like he did when I was his apprentice....

Can't wait to see it in action in a couple hours, very cool Dave!!!

PS: Did David mention that we have a group of clients from one venue who are wearing out their stainless frets in 6 months time? It's amazing and I refuse to shake these guys hands as well telling them no offense but I need my hands to make my living.... Good time of year for fist bumps anyway for Felix Ungar here....

Hesh: That's amazing about the S.S. frets. I've heard reports of folks having stainless show little or no wear after a couple of years of heavy playing. It will certainly be interesting to see the final results. Thanks for doing this folks.
Tom

_________________
A person who has never made a mistake has never made anything!!!

Dave,

Hope you will be doing the StewMac products as part of your tests. I have been using them for some time on my flamencos, but have no real feedback on their wear characteristics.

Thanks,
Max

_________________
Max Bishop
Brighton, Michigan

Thanks folks. Yes I'll be doing EVO, StewMac, some Dunlop, probably all referenced against their equivalent size in Jescar nickel (can only do 2 wires at a time and keep the comparison accurate). I will have to do some test runs to tweak the tests, but I made it very simple to have a bunch of blocks to install wire in than connect to the tester with one screw. Nice thing about the sewing machine is that I can adjust the X-axis movement with the feed rate control independent of the Z-axis or depth.

I'll likely have to go through several tests like this along with wire gauge, tension, etc, but so far it looks promising to yield results representative of accelerated real world wear (current rate puts it under about 30,000 cycles per hour).

My guesses are in line with yours. Pulling numbers out of my rear, I'm expecting stainless to be in the 4-5x life as nickel, EVO at least 2-3, maybe more, and if the specified hardness ratings are accurate StewMac wire will probably be about the same as Jescar, or perhaps just a hair less life. We'll have to wait and see though.

_________________
Eschew obfuscation, espouse elucidation.

You'll get a lot of mileage out of 'standard' fret wire and Nylon strings. It's not something I would be too concerned about. Steel hitting an alloy is an entirely different matter.

looks neat! but what are your base line strings? diameters? materials? coated or not? what is the actual measured pressure exerted on the frets? questions, questions, questions....

This type of test is comparative rather than absolute, so these factors don't matter so long as they're consistent. I will test with a variety of string types, gauges, tensions, etc, and refine some of the dimensions and deflection amounts as I begin to experiment with it a bit more, but ultimately when running two wires simultaneously under identical conditions, what we are looking for are differences between the two.

_________________
Eschew obfuscation, espouse elucidation.

Looking forward to your results david.

Working out some bugs, but coming along. Hooked up a counter and ran a sample through 100,000 cycles today. Found an issue with flex causing an uneven load on the 2 frets, replaced the foot with a beefed up and leveled out support, and it seemed to correct the issue in a brief test run afterward.

I'm sure the next few days will be reserved for a few more refinements and calibrations, after which I hope to be able to run at least 10-15 comparison samples per week. Should be interesting.

_________________
Eschew obfuscation, espouse elucidation.

Wow David!

Thank you for this, I eagerly await the results

_________________
http://Michaudmade.com

It'll be a few weeks before we can gather enough data to be meaningful. Lot's of variables to work with, hopefully no big unforeseen bugs, but still in kind of a proof of concept testing phase.

_________________
Eschew obfuscation, espouse elucidation.

I'll be staying tuned. Cool little experiment and apparatus, only took an hour? Clever.

That's what I like to tell myself. Never mind the hour I spend tearing the machine down at home the night before (home hours don't count), then a few hours the day after of refinements, adding a counter, making a batch of test blocks - but yeah, the first version on the first day only took about an hour to throw together.

Then there's the other fairly critical glitch that occurred to me this morning, but I have a few ideas on how to straighten that out.

_________________
Eschew obfuscation, espouse elucidation.

I realized this morning that the nature of the design is somewhat self-regulating, in a bad way. As one side wears faster, the harder side will remain more proud, lift the string first, delaying contact and lessening the load on the side with early wear. This will in turn leave more pressure and accelerate wear on the harder side, automatically regulating the two samples to wear more evenly than they would if the pressures remained constant on each throughout the test. Early tests with stainless vs nickel showed a real difference in wear, but not nearly as much as I had expected, and this would explain why.

A few possible solutions to this fundamental flaw, but I'm leaning toward setting the base to pivot so that as wear develops the test block will automatically adjust so that the string always contacts and departs each fret at the same time, maintaining identical pressure on each side throughout each cycle. This of course means more precise test blocks with locators on the bottom to ensure frets are perfectly balanced off side of the hinge point, and probably a light clutch on the hinge to avoid wobbling and bouncing as it runs while leaving as little resistance possible to balance on contact.

These pet projects always seem so simple late at night with a beer and sewing machine at the kitchen table. Ah well...

_________________
Eschew obfuscation, espouse elucidation.

Made a few refinements today (some one-hour test apparatus, eh).

Built a cradle which self levels as the frets touch the string, with just enough clutch friction to self-adjust with minimal pressure as wear progresses. A locator pin is positioned to ensure each sample is placed so that frets are equidistant from the ends and should be subject to equal pressure.

Seems to be working well, and initial tests with identical wire in each position so far demonstrates equal wear, though of course a few more tests to verify reliability will follow before serious comparative testing begins. In this quick video I got a chance to play with the slow motion, and you can see why an old sewing machine seemed the ideal low-cost base for this testing, with easily adjustable range from direct up-down motion to over 1/8" sweep with a simple dial control (and a reverse setting to boot).

I will likely put out an article on the results once enough testing is complete (and hopefully not too many more bugs to be worked out). May be a while, but it should hopefully be interesting.

_________________
Eschew obfuscation, espouse elucidation.

David, I find, if something seems simple, I'm just not looking close enough.
I here by nominate you as the new host of Lutherie Myth-busters! We could just turn you loose with a big budget and a good junk pile.
Talk about niche entertainment.
Beautiful.

How about preparing different samples with one or two frets of the same wire on the fret board then expose them to a set amount of cycles, then compare the samples to each other, or run until a measured amount of fret wear, and then compare the numbers? ......And with like fret wire, it would be a good way to check if your set up is producing even wear.

Chuck

_________________
_________________

That's the general idea in this initial stage - running identical samples on each board through testing in different settings and orientations to ferret out any inconsistencies which may (or hopefully may not at this point) exist in the test apparatus. Then different conditions will be tested (string type/gauge, tensions, straight pressure vs sliding abrasion, etc) to determine an appropriate benchmark for wear or number of cycles to be standardized for comparison in these tests.

I'm sure the scope of conditions will expand to begin with until I can observe which conditions yield the most relevant results, at which point I hope to be able to narrow the conditions back down to the few most pertinent and revealing to use in more thorough testing.

_________________
Eschew obfuscation, espouse elucidation.

If only it could be rigged up to do Smoke On The Water............... all day long. Ya I know. Couldn't get much better then that right.

Good one Danny bro! As it is that machine chirping right next to me reminds me of my ex bitc..... er...... I mean talking at me.....

David,
A great idea to quantify the wear of the different fret materials, and
a creative way to generate the wear. Takes me back to Tabor wear
testing I did as a young engineer (waaaay back).

_________________
Gene

Politicians and diapers must be changed often, and for the same reason- Mark Twain