Solved an Amateur Lutherie Trap

[Mark Gunter]

Luthiers are bad ass. They can make instruments that are beautiful works of art as well as wonderful, practical tools for making music. The good ones really know what they're doing.

Some of us can't readily afford the services of a good luthier, and being do-it-yourselfers, we sometimes engage in "amateur lutherie." I am one of those, but I've found that it is easy to screw things up when you don't really know what you're doing, and you have just enough knowledge to be dangerous.

I like a low action, and on one of my mandolins I cut the nut slots lower soon after getting it, because it needed the nut slots lowered. But eventually, I went a little too far, and had to fix it. I didn't want to "build up" the slot with dust or baking powder and CA glue (I've use that trick before on guitars) and I didn't want to shim under the nut with a playing card (likewise, done it before on guitars), I decided to replace the nut with a bone nut.

Since replacing the nut with a new one, I've never been happy with the performance of the A strings. I've found it impossible to keep them in unison, and impossible to keep the mandolin sounding right when playing the open A and E courses for the fifth interval and the unisons between A and E courses when playing 7th fret E on the A course.

The reason for my post today is that I finally determined for certain that at least one of the A strings is binding in its slot. I've checked for this before, and made adjustments to the nut before, to no avail, but as of yesterday and today I've determined without a doubt that this is my problem.

The way I know this is that when I tune the offending A string, I can stretch it by pushing on it between the nut and the tuner, then the string goes sharp. So what is happening when I tune up and play is that my playing (manipulation of A strings) causes slack to work a bit through the nut toward the tuning machine and makes the string go sharp by degrees as I play.

All this may or may not help other amateur luthiers out there, but I had to post about it because this small win is such a big win for me and my sanity playing this instrument! Yes, it should have been obvious what the problem was from the start ... but I swear I thought I'd covered that possibility long ago, and tried to make adjustments to the nut for it. Somehow, it is still binding in the slot. There can be no doubt now. I'll fix it.

[Tom Wright]

Somehow, it is still binding in the slot. There can be no doubt now. I'll fix it.

I find this same issue and tested it the same way by deflecting the string behind the nut to see if it returns to pitch. I addressed it by widening the slot on the peg side, so that it flares out a bit (very slight). This should mean it only contacts a short bit right at the fingerboard end. Luthiers sometimes round the backside or bevel it down but I like a more level nut surface to keep the string in the slot under heavy use. But this meant too much contact area.

I need to get a larger nut file, making do with a .013 and wiggling it to get a slot good for a .015 string. (Can’t bring myself to order just one file.) I also cheat and use a tiny bit of oil in the slot, which should probably be dry teflon powder, but the two tweaks seem suffcient to survive even heavy picking so that the A pair stays in tune.

[Br1ck]

What I've found is there really is no substitute for experience, especially when it comes to diagnosis of setup issues. By trial and error and a process of elimination you can finally track down your problem. I did this last week with what ended up being excess relief.

The other issue for me was tools. I resisted buying nut slot files for a long time. Now that I have them life is easier. I've lost track of what I've spent on tools. I started with a machinist's straightedge years ago. That has saved me a lot over the years.

I also think you have it in you to tinker or you don't. If you don't enjoy the journey, don't start, but if you do, nothing compares to building a mandolin. But I would caution against an f style as a first build, but then again, all's well that ends well.

Congratulations on your sucess.

[sblock]

I empathize with your story, and I've been down this same road before. I suspect that many of us have! Sooner or later, many (most?) mandolinists will want to take over the accessible minor aspects of setup, rather than having to seek out a luthier every time. There are many possible reasons for this, among them these:

1) You stand to save some money
2) There may be no qualified luthier in your immediate area who really understands mandolin setup
3) You may not wish to be without your instrument for whatever time it takes a luthier to work on it
4) You take personal pride in maintaining your own instrument
5) You may have particular demands that are not easily met, nor described to some other person
6) You may simply wish to experiment with different parameters until you find what you like best
7) You may have the occasional small emergency and need to deal with it, right on the spot

and so on...

Regardless of the reason, binding at the nut for one or more strings is a very common problem indeed. In my experience, too many folks tend to cut simple 'v-shaped' notches using files, and these are prone to sticking. (They also cut them too deep, or fail to angle the string properly towards the peghead -- so that the break angle is wrong, or the contact area is too small, or the break is not at the front edge of the nut).

What has worked for me was to break down and finally buy a set of proper nut files to shape and smooth the bottom surface of the nut notch, so that it's not 'v-shaped', but instead is well-matched to the half-round shape of the particular gauge of string in the slot. These specialty tools work far better than old-fashioned triangle files, torchtip files, or (most) needle files. Such files are available from StewMac, for example (NFI), and other places as well. They're not especially cheap, but they do a terrific job and pay for themselves quickly, compared with going to a luthier.



Of course, it also pays to lubricate the nut slots (graphite is my choice) regularly when you change strings, and to use a hard, uniform nut material. Bone quality can vary a lot, but a well-made bone nut works incredibly well. Mother-of-pearl nuts are much harder to cut, but may bind less.

Here is Paul Hostetter's (Santa Cruz luthier) famous drawing, from his earlier posts, that shows the proper way to cut a nut slot.



Note that he recommends that the depth of the nut slot be only ONE THOU higher than the height of the first fret. Note also than only half the string diameter lies in the nut groove. Many grooves are cut too deep, leading to binding by the sides of the string.

Paul H. has also pointed out that you can even make your own nut files by filing or sawing notches in the sides of a set of feeler gauges of the appropriate thicknesses. This presupposes that you get the kinds of feeler gauges that come with rounded edges, not square ones. I haven't tried that, but it sounds like it would only save at most $40 -- the price difference between 4 nut files ($60) and a feeler gauge set ($20).

[kurth83]

Great thread, I am about to try making my first nut in the next day or so, I bought 10 just in case.

I have filed many a pre-cut nut, and have the stewmac files and their wonderful string height gauge, so once I get the grooves in the right place I should be ok.

However, one thousandth of an inch over the first fret seems wrong to me.

I have never gotten below about .009 for A and E strings, and about .015. for the G string. Lower than that and the G will buzz when played open. I learned this the hard way, I have a G at about .012 and it buzzes when hit hard.

[sblock]

Great thread, I am about to try making my first nut in the next day or so, I bought 10 just in case.

I have filed many a pre-cut nut, and have the stewmac files and their wonderful string height gauge, so once I get the grooves in the right place I should be ok.

However, one thousandth of an inch over the first fret seems wrong to me.

I have never gotten below about .009 for A and E strings, and about .015. for the G string. Lower than that and the G will buzz when played open. I learned this the hard way, I have a G at about .012 and it buzzes when hit hard.

It is possible that Paul H really meant 10 thou, not 1. The nuts on my mandolins all have about 10 thou relief over the 1st fret height. But a zero fret is the same height as a first fret, and the string will have only come up a few thou between frets 0 and 1, at most.

[SincereCorgi]

I have never gotten below about .009 for A and E strings, and about .015. for the G string. Lower than that and the G will buzz when played open. I learned this the hard way, I have a G at about .012 and it buzzes when hit hard.

Yeah, that looks like a platonic ideal of a nut rather than a practical one. I can't imagine any of us have nuts where the plain strings sit half-way out of the nut slot, unless the slot is like a trench running down the middle of a hallway.

[sblock]

Yeah, that looks like a platonic ideal of a nut rather than a practical one. I can't imagine any of us have nuts where the plain strings sit half-way out of the nut slot, unless the slot is like a trench running down the middle of a hallway.

I understand your point. As long and the parts of the nut slot that actually make contact with the string only touch the bottom (round) half of it, it should not matter whether the string sits a bit deeper in the slot. If the slot widens slightly, above the height of the string radius, that would not be a problem. The problem happens when a slot is too deep and also narrow, and the sides of that same slot contact the string, causing excessive binding.

[Phil Goodson]

Yeah, I suspect Paul H really meant 10 thou, not 1. The nuts on my mandolins all have about 10 thou relief over the 1st fret. height.

If you built a zero fret instrument, would you make the height of the zero fret 10 thousandths higher that the top of the first fret?

[Tom Wright]

If you built a zero fret instrument, would you make the height of the zero fret 10 thousandths higher that the top of the first fret?

A fair point. The height above the first fret will be the result of the angle between the zero fret and the bridge. To have a higher clearance would indeed require a high zero fret. This is an advantage of a nut.

I have my plain strings at about .007” and quite high, like.015” for the C string.

As to slot depth, I see nothing wrong and hear no difference between deep or shallow. In principle, a proper nut slot will have no effect on the upper half of the string, only the contact zones on bottom and sides. I have fairly shallow slots on my electric, and although no string jumps out while playing, I have to store the pick carefully so I don’t accidentally lever a string out of its slot.

I have had the experience of a E string easily jumping out while playing (not my nut, but the luthier’s) from the combination of small break angle and shallow slot. I had to drill an extra hole in the peg, to wind the string more steeply downward.

I remember Fender using V-shaped slots for their electric bass, and the low string was prone to jumping out of the shallow V.

[sblock]

If you built a zero fret instrument, would you make the height of the zero fret 10 thousandths higher that the top of the first fret?

Actually, a zero fret has exactly the same height over the fingerboard as the first fret. That would make the effective nut height 0 thousandths higher than the first fret! A string leaving the zero fret will have come up several thou by the time it reaches the first fret, however, due to the action height at the bridge, of course (the open string is not parallel to the fretboard!), but the same thing is true for a regular nut. So Paul H's suggestion to use 1 thou might be perfectly OK.

[colorado_al]

It's all about having the right tools for the job and practice. After setting up too many mandolins without proper nut files I spent the $50 and ordered some. Should have done that in the first place and made things easier from the get go.

[Dave Sheets]

+1 for the nut files, easily worth the investment if you have multiple instruments around. Nut height makes such a difference on the feel of the instrument as well as the intonation.

[Mark Gunter]

I use the stewmac nut files that sblock pictured ... but I never have the exact gauges for the strings I'm slotting for, and obviously I need to invest in more files.

As to nut slot height: Paul's advice seems pretty sound to me. I prefer a low action. I want the nut slots to act similar to a zero fret. IOW, to check by eyeballing, what I do is depress string at first fret, hold it, and eyeball how far that string travels to contact the second fret. I want the travel from nut down to contact first fret to be about the same.

Once you get up the neck, fret-to-fret, this is what you play with. Why should nut to first fret be different?

[sblock]

I use the stewmac nut files that sblock pictured ... but I never have the exact gauges for the strings I'm slotting for, and obviously I need to invest in more files.

As to nut slot height: Paul's advice seems pretty sound to me. I prefer a low action. I want the nut slots to act similar to a zero fret. IOW, to check by eyeballing, what I do is depress string at first fret, hold it, and eyeball how far that string travels to contact the second fret. I want the travel from nut down to contact first fret to be about the same.

Once you get up the neck, fret-to-fret, this is what you play with. Why should nut to first fret be different?

I think you're right, Mark -- the height of a string leaving from the nut (or leaving from a zero fret, for that matter) need be no higher than the height of a regular fret. I think Paul H is spot on about this. You can add a thou or two to be on the safe side, but adding 10 thou or more should be unnecessary.

That said, if you are already using properly shaped nut files, and file the correct relief angles, I confess to be rather mystified about your report that the A string is still binding in the nut slot. I presume you tried a lubricant like graphite, of course. In that case, the most likely explanation to me seems to be the one that you just hinted at: you are probably using the wrong size of nut file to form the bottom of the slot for your A string, given the string gauge you actually use. Too narrow and you can develop some sticking. Too wide and you can develop some buzzing. The J-74 A-string is 15 thou. The J-75 A-string is 16 thou. The StewMac nut file for the mandolin set is 16 thou (part #0827) and works fine for 15-thou J-74 A-strings. However, the next available file size up from there is 20 thou (part #0828), which is too wide. If you use the heavier J-75 A-strings, maybe you should try tilting the file you have just a bit, and widen your A-string nut slot, ever-so-slightly?

May your action be low and your playing aspirations high.

[kurth83]

My stewmac nut files aren't exactly the same gauge as the strings either, and I don't think you would want that, when strings bend they will deform slightly, too tight a fit and they would bind up. A few .001's bigger is better IMHO. I am guessing that's why my deep nut slots work ok too. I have nut sauce but never had to use it.

EDIT: I am filing my new nut right now... pics to follow soon.

[Mark Gunter]

... if you are already using properly shaped nut files, and file the correct relief angles, I confess to be rather mystified about your report that the A string is still binding in the nut slot.

Exactly ... I was mystified as to why my A course was so troublesome, because I cut this nut with what I thought were good relief angles, and of course I assumed the nut was the problem with those strings, since I had replaced the nut myself. Nevertheless, over the past two years I have checked everything from one end of the string to the other, and even worked on the slots a few times. I do use graphite from pencil lead there as well. But obviously I kept missing the fact that those strings were still binding. I worked on it last night, and widening the slots a bit more was the solution.

My stewmac nut files aren't exactly the same gauge as the strings either, and I don't think you would want that, when strings bend they will deform slightly, too tight a fit and they would bind up. A few .001's bigger is better IMHO.

Funny thing is that when I went to work on this last night, I realized that the A strings were the only gauge that I had an exact file for: .016

My E strings are .0115 and since I have a .010 file and a .013 file, I think I must have used the latter back when I first cut this thing. So basically, I used various files to cut the other three slots as best I could, but I used only the .016 file to cut for the .016 strings -- and though I tried to widen the slots a bit, evidently I still hadn't widened them enough. Such a small thing has caused me some grief for a long time. The reason this got away from me for so long was that I had put a good bit of energy into solving the problem a long while back, never arriving at the solution, and this was not my first string instrument. Now, it is becoming my only mandolin.

I've read from luthiers everywhere that nut slots should be just a hair wider than the actual string gauge, and I thought I was accomplishing that. The moral for me: Do not put too much trust in having the "just right" file gauge for the string gauge. The only strings I had binding were the ones I had the correct gauge for!

[Tom Wright]

Why should nut to first fret be different?

I used your technique, observing deflection, for a long time but finally got a feeler gauge to be more definite. As to why it should be different, it feels a bit different when the string is tilting upward a bit, which may be why some players like relief, i.e. bowing

Also, it’s nice to have the more powerful high clearance open string tone available for expression. I found, after shortening my fingerboard for intonation reasons, that a very high C pair was better even in the low fret notes, and of course had a big sound when open.

Another question that interests me is what the string behind the fretting finger does. With some relief or a high nut the string is clear, but with a flat neck and zero fret the strng is in full contact. Does it buzz? Any other difference?

[sblock]

Exactly ... I was mystified as to why my A course was so troublesome, because I cut this nut with what I thought were good relief angles, and of course I assumed the nut was the problem with those strings, since I had replaced the nut myself. Nevertheless, over the past two years I have checked everything from one end of the string to the other, and even worked on the slots a few times. I do use graphite from pencil lead there as well. But obviously I kept missing the fact that those strings were still binding. I worked on it last night, and widening the slots a bit more was the solution.



Funny thing is that when I went to work on this last night, I realized that the A strings were the only gauge that I had an exact file for: .016

My E strings are .0115 and since I have a .010 file and a .013 file, I think I must have used the latter back when I first cut this thing. So basically, I used various files to cut the other three slots as best I could, but I used only the .016 file to cut for the .016 strings -- and though I tried to widen the slots a bit, evidently I still hadn't widened them enough. Such a small thing has caused me some grief for a long time. The reason this got away from me for so long was that I had put a good bit of energy into solving the problem a long while back, never arriving at the solution, and this was not my first string instrument. Now, it is becoming my only mandolin.

I've read from luthiers everywhere that nut slots should be just a hair wider than the actual string gauge, and I thought I was accomplishing that. The moral for me: Do not put too much trust in having the "just right" file gauge for the string gauge. The only strings I had binding were the ones I had the correct gauge for!

Glad you got it all worked out, and gratified that the cause (nut slot ever-so-slightly too narrow) was exactly what we'd suspected all along. Like so many things in life, the fix turned out to be very simple. But that's only in 20-20 hindsight, once you came to understand the vagaries of the problem, and explored many a blind alley! The next fix will come so much more quickly...

[SincereCorgi]

I've read from luthiers everywhere that nut slots should be just a hair wider than the actual string gauge, and I thought I was accomplishing that. The moral for me: Do not put too much trust in having the "just right" file gauge for the string gauge. The only strings I had binding were the ones I had the correct gauge for!

If you clean up the slot with some sandpaper, that might help too... that or the inevitable set of 80 graduated nut files from Stew-Mac.

[kurth83]

I would never put anything in a nut slot other than nut files and strings (lube maybe).

I once accidentally dragged a nut file horizontally (just one pass parallel to the strings) in a nut slot, it ruined the string's tone. Fortunately I had some room to work with, had to take it down a bit more the get the slot re-angled the right way.

I think it would be real easy to ruin a nut slot with sandpaper that way.

[SincereCorgi]

I would never put anything in a nut slot other than nut files and strings (lube maybe).

I once accidentally dragged a nut file horizontally (just one pass parallel to the strings) in a nut slot, it ruined the string's tone. Fortunately I had some room to work with, had to take it down a bit more the get the slot re-angled the right way.

I think it would be real easy to ruin a nut slot with sandpaper that way.

Interesting! I know a couple local guys that use sandpaper as their last step. Lots of way to skin a cat, though.

[Mark Gunter]

I used your technique, observing deflection, for a long time but finally got a feeler gauge to be more definite. As to why it should be different, it feels a bit different when the string is tilting upward a bit, which may be why some players like relief, i.e. bowing

Also, it’s nice to have the more powerful high clearance open string tone available for expression. I found, after shortening my fingerboard for intonation reasons, that a very high C pair was better even in the low fret notes, and of course had a big sound when open.

Another question that interests me is what the string behind the fretting finger does. With some relief or a high nut the string is clear, but with a flat neck and zero fret the strng is in full contact. Does it buzz? Any other difference?

Tom, my thoughts on your earlier posts was that there is probably a relief problem if fret slots cut per Paul H's diagram cause string buzzing ...

I checked my guitar and mandolin, and there is only the tiniest bit of clearance on the string-to-frets behind the fretted position. I experience no buzzing or other problems.

You're referencing your five course instruments, I presume? Could it be that the C course is a bit heavy gauge for the scale length? Honestly, I don't know much about all this, other than that I've always preferred a pretty low action on my stringed instruments. I have a history of blues playing, string bending, etc. I need to state here that I do not lower nut slots enough to be equivalent to a zero fret, only nearly so. I have feeler gauges, but generally do not use them to stress over exact clearances, because I know what action I like and can easily achieve it employing the eyeballing method I described. If I were setting up an instrument for a friend who requested a specific clearance, I'd use feeler gauges.

There's a lot to be said for precision to be sure, but as has been mentioned, there's more than one way to skin a cat. It's not anathema to use a bit of sandpaper, or diamond files, etc. to cut a slot if that's what you have to work with, but as I wrote in OP, amateur lutherie can lead to problems, like the one I've been dealing with here. Caution is advised; best practices recommended.

## Correction: It was post #5, kurth83's post, that made me think immediately of neck relief, but I'm not saying that was his problem, just something I would be sure to check.
## Also, I have used (and think I did on this mandolin, memory isn't great) feeler gauges to get close to depth before the re-stringing, test, tweaking stage which is always a PITA.

[JeffD]

Thing is, all that home luthiery cuts into my playing time. Just sayin...

[Mark Gunter]

Thing is, all that home luthiery cuts into my playing time. Just sayin...

Kinda like forum surfing, right?