Official Luthiers Forum!

I've just made myself an aluminum template for fret slotting and I'm trying to evaluate whether it's good enough. I used an Optivisor, marked with a sharp knife, cut notches with an Xacto razor saw and generally did everything I could to preserve precision. But the bottom line is that any one fret position might be off by 1/64" (or if I was really successful being careful, maybe half that).

Having limited experience, I keep wondering if that's good enough.

What do you veterans think? Would a good player with a good ear likely be happy with this or should I part with the greenbacks to buy a template?

I suppose if it isn't good enough and I build with it, then I'll have some experience.

I wonder what the margin of error is on even pre-slotted boards from reputable places. Because I have compared them with ones that I hand cut, in a manor similar to what you describe, and mine seemed more accurate.

The important thing is to measure every fret from nut to the fret distance so as not to introduce a systematic error when measuring from slot to slot. My guess is that yes, someone with a fine ear will hear it but, they will hear everything else that is the problem with guitar intonation as well.

I have made a fret template by eye and a bunch of fretboards by eye over the years. Never had a problem with them. Yours should be fine.

Thanks guys. I'm sure it's true that there are many other source of funky intonation. (When I play, +/- 1 fret is pretty good.) I kept telling myself that before CNC everyone slotted fretboards by hand with a backsaw so how much worse could I be? I guess I wanted some confidence that 1/64" precision isn't wildly irresponsible.

I think you'll be fine too. There were millions of instruments made in the pre-CNC era, and still today there are tons of instruments made with various mechanical fret slotting jigs/machines, and most often the intonation is fine for most of us. Given that the slot is perfect, the way the fret goes into the slot or the way it is filed could vary the distance by your 1/64". Given that the slot layout is close, intonation is probably more often affected by the saddle placement, string height, or the strings themselves. You should be good to go.--Bob

_________________
“ The meaning of life is to find your gift and the purpose of life is to give it away” Pablo Picasso

The way I think about it is that at the peghead end of the fingerboard the frets are more than an inch apart so 1/64th inch is about 1 "cent". At the end of the fret board near the soundhole 1/64th inch becomes 3-4 cents, but string stiffness and string stretch has a great effect on the intonation also. In addition to those elements you have the whole even tempered tuning thing going on.
Staying within 1/64th of an inch I think is good enough.

Not a veteran, but just an observation.
Are you talking abut one fret, or all frets?
Just if you have adjacent frets +/- 1/64", that could potentially double any error between those 2 frets, making it 1/32

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

I mean each fret, as measured from the face of the nut, so at least the errors won’t accumulate.

I’m relieved to see several people agreeing that the errors are pretty small. Can surely see the value of CNC to drastically reduce one source of intonation error but I’m interested in playing around with scale lengths and I’m retired so I usually find myself with more time than money on my hands. If I can repeatably make a pretty darn good fretboard (without spending $40 or $50 for every 2 scales) I’ll continue to invest the time to make my own templates. Aluminum is pretty cheap at HD and it’s easy to work.

Thanks, everyone!

Good points about filing the frets and variability of strings. Even as they age on the guitar my strings seem to change quite a bit.

Well, certainly for me, 1 cent won’t matter. I haven’t tested myself so I don’t know if I could hear 4 cents out of tune but I wouldn’t be surprised to learn there are people who can. Those people probably won’t be playing my guitars any time soon. I just hope my guitar teacher isn’t one of them.

I once compared the fret spacing on two high end classicals of the same length. They were not in proportion to each other at all. I wondered if it had something to do with the country of origin in that maybe certain keys were more popular to play than others.

With magnifiers and exacto knife you should be able to get pretty close. So it's probably in the sawing operation. When I did it without a template, the rule seems to be to do each one exactly the same way so that if your Technique was off a little they would all be off in the same direction and amount.

Not sure how you are measuring, but I learned to measure in millimeters using a half millimeter ruler with magnifiers and exacto marking to the nearest 1/2 millimeter which is a little over 1/64".

New username, same Pat Mac

_________________
Pat

For a board, I'd fill the worst slots with sawdust and CA and recut them. For a template, scrap it.


At the very least, you should round to twice the precision of your ruler. So if you have 1/2 millimeter marks, half way between two marks should be considered a valid fret position, giving you 1/4 millimeter precision.

I use a cheap iGaging straightedge with 1/64" marks. I like that they're bit finer than 1/2 millimeter, although that does mean my fret position chart is in the rather ridiculous system of inches and decimal 64th's For example I might have 5 23.75/64, which means to go to 5 inches and then 3/4 of the way between the 23 and 24 64th marks. I can't actually achieve such precision, but at least I know that it should be closer to the 24 mark than the 23 mark, and should not be precisely on it.

Assuming that the error in layout is about 1/2 of the smallest rule division, and cutting error is 1/2 the kerf width (i.e., the cut is done with one shoulder of the kerf on the line), it seems like marking out with a 1/100" grad rule would give a 0.005" plus 0.012" total error, or just over a cent at the 12th fret for a 25" scale board. A rule with 64th inch grads would give 0.008" plus 0.012", or 0.020" error - about 1-1/3 cent at the 12th. Marking out with a 1/2mm grad rule would increase that mark-out error to 0.010" plus the 0.012" cutting error, for a 0.022" total error, or about 1-1/2 cent. These seem like pretty small errors, assuming all measurements are done from the nut, that alignment at the nut is without error, and the rule is adequately accurate. Assuming errors at both ends - a better assumption when I am doing the layout - gives 0.022", 0.028", and 0.032"...that's just under 2 cents to a bit over that at the 12th fret.

_________________
A constellation only takes shape when one maps the whole.
- Beth Brower

Had this same discussion over on the MIMF many years ago. Based on reasoning similar to Woodie's, an accuracy goal of +/- 0.010" was considered acceptable and attainable.

_________________
Steve Smith
"Music is what feelings sound like"

I read this with some issues, we are talking about measuring in fractions of inches or MM with steel rulers and this is not very accurate depending on the users skill and eye sight. I have found a 20" digital vernier is way more accurate and easy to use. These are available for a reasonable price from China and plenty durable and accurate for out use. I borrowed a very expensive unit fro a machinist buddy to check fret spacing due to an intonation problem and was so impressed I ordered a cheaper version from Ali Express and when it arrived we compared it to the other one and it was every bit as accurate but a bit lighter built and not suitable for constant industrial use.

See it at bottom of photo



Fred

_________________
Fred Tellier
http://www.fetellierguitars.com
Facebook page http://www.facebook.com/pages/FE-Tellier-Guitars/163451547003866

To lay out frets I have used a normal 6" digital caliper and an accurate 6" and 12" rulers to extend the caliper's reach to the upper frets. Just butt the ruler up against the nut and then register the caliper off of the end.

Pat, that's exactly how I measured, except using 1/64 in. increments instead of half-millimeters. I also tried to judge 1/2 increments (1/128th"), to the best of my ability.

There are so many sources of potential errors though, that it gives me pause. My visual judgement may be off, the marks on the rule seem kind of fat (at least through the magnifier), my mark may not have landed exactly where I thought it did (next time I try this I think I'll first grind the bevel off one side of the Xacto blade to improve my accuracy), my saw may wander off the mark when cutting slots in the template, etc.

Add to that the potential inaccuracies of sawing the board from the resulting template. Although, I think my jig and template are pretty good. My template has shallow notches cut with the razor saw @ .012" kerf, and I cannibalized an old feeler gauge for its .012" leaf to be the indexing pin in my jig. The fit is snug. As long as the template edge and the jig's fence are perfectly straight so the workpiece/template can't rock against the fence I think that part of the operation is pretty good. The Stewmac blade might have a teensy bit of wobble but it seems good -at least the slot widths are consistent across the fretboard (though I admit that's not very far).

It's all an approximation I guess but hopefully a pretty good one. I guess I'll find out when I finally string up this guitar.

Thank you for this analysis.

In my example, I used 1/64", my kerf is .012" so, optimistically, the tolerance in my template is around +/- .014" for each fret. If I've done my pitch calculation correctly (an iffy assumption) this would be about 2.2 cents at fret 12 on the high E string. That seems in the ballpark of your calculation so maybe it's even correct.

That's just the template error though. There's still the sawing error to account for but I don't have any data on that.

Man, that looks like a great tool for this job. I just took a quick look around ebay and Amazon and the couple of similar calipers I found were in the $hundreds. Maybe someday. Or maybe it'd be more cost-effective for an amateur like me to just spring for a few templates.

Fred, is this the one you have?

https://www.aliexpress.com/item/SHAHE-600-mm-Digital-Vernier-Calipers-Electronic-Caliper-Steel-paquimetro-digital-Measuring-Tools-Messschieber-Caliper-Gauge/32953312224.html?spm=2114.search0204.3.2.595329437AGqDL&ws_ab_test=searchweb0_0,searchweb201602_1_10065_10130_10068_10890_10547_319_10546_317_10548_10545_10696_453_10084_454_10083_10618_10307_537_536_10902_10059_10884_10887_321_322_10103,searchweb201603_45,ppcSwitch_0&algo_expid=9035e0ba-165b-449e-844c-a2dbb930f587-0&algo_pvid=9035e0ba-165b-449e-844c-a2dbb930f587&transAbTest=ae803_5

No I wanted to be able to do inside and outside measurements so got this one. It has gone up in price since I got mine. The inside feature is great for measuring nut to fret then add 1/2 fret width.

https://www.aliexpress.com/item/0-500mm-Heavy-duty-Digital-Caliper-500mm-20inch-Electronic-Vernier-caliper-measuring-tools-gauge-with-nib/1809114420.html?spm=a2g0s.9042311.0.0.39124c4dlf0Eme

_________________
Fred Tellier
http://www.fetellierguitars.com
Facebook page http://www.facebook.com/pages/FE-Tellier-Guitars/163451547003866

Thanks. That does seem a pretty reasonable price.

I use an ofcut edge from a pre slotted board with a home made jig as my cheap option.