Compensated Nut

[vfont]

There seems to be a lot of posts about intonation and not being able to get things quite right on a mandolin. In the banjo world, Geoff Stelling patented the compensated nut. Stelling banjos are known to have excellent intonation all the way up and down the neck. Has anyone ever tried compensating the nut on a mandolin or is this just too crazy a question to ask?

[craigtoo]

I've seen it done on an OM. Where the player switched from Wound to Steel strings on one course.. and rather than have a new saddle made.. he made a new nut with a little bit of compensation on that one course...

ch

[sunburst]

I haven't tried it on mandolin nuts, but I assume it would work. I've thought about it, and haven't ruled out trying it, but the "traditional" compensated bridge works pretty well on mandoilins.

BTW, Geoff didn't patent the compensated nut. It's an old trick that banjo players have done for many years.

[Jason Kessler]

Here in NYC, a compensated nut is an insane street person to whom you've given a buck so he'll go away.

[siren_20]

I've heard radio ads recently for this compensated nut manufacturer... I'd imagine that it would help with tuning stability overall, and it doesn't seem that hard to do.

[Stephen Perry]

Many articles, I have bunches somewhere. All directed at guitar. I've done it on a few. Very nice.

[woodwiz]

It would appear that it only makes a difference when playing open strings. When the notes are fretted, only the bridge compensation would appear to make a difference.

Am I seeing this correctly?

[sunburst]

In a word, no.

I've thought the whole thing through, and can understand how it works on stopped or capoed strings, but it isn't easy to explain.

First of all, the three things that affect the pitch of a string are length, tension, and string gauge. Let's assume we're sticking with the same string in this explanatory example, so the gauge can't change.

Imagine you have a string that notes incorrectly on a fret, even though it's tuned correctly. That string will be in tune open, but out of tune capoed. Now, suppose you put on a capo and retune the string so that it is in tune. What have you changed? Tension.
So, if you take the capo off, the only way to keep that tension (string gauge staying the same) and be in tune open is to change the string length. You can do that at the bridge or at the nut, but either way, the tension will be the same when you put the capo back on, and, in fact, the vibrating length will be the same as it was when you tuned the string (capoed) only if you compensate at the nut rather than the bridge, so that a compensated nut actually works better with a capo than a compensated bridge, and the string will now be in tune open and fretted.

[Bob DeVellis]

John -- nice explanation! I've always been puzzled. Banjo players swear that compensated nuts help with fretted and opened notes and I've never quite gotten how that works, until now.

[clarksavage]

Not much to add, I'm a novice, but my Eric Darnton Octave Mandolin has a compensating nut. The intonation is the best (balanced well?) my ears have ever heard. I recall surprise when I noticed it and spoke to Eric, the luthier, and he told me it helped a whole lot to get intonation right.

Clark

[Paul Hostetter]

One of the fundamental premises #of "correct" intonation on fretted instruments is that the frets are really correctly located in the first place, which is the case much less often than you would suppose. In other words we normally assume that they are really consistent with the equal temperament system that matches the piano, which is how most people learn to hear, and every one of them is exactly right. We question the placement of the ends of the string, but nothing about the middle. We should, however.

Let me rephrase John's post a bit.

Imagine you have a string that notes incorrectly on a fret, even though it's tuned correctly. That string will be in tune open, but out of tune at just that one fret. Do you assume it's something to correct with the nut? I wouldn't. I would look to see if the fret was in the right place. I contend that if they are, and if the gauges of the strings are realistic for the scale (which is really not the case with banjos), any anomalies in intonation can be adjusted at the bridge alone. Nut intonation is a micro-cure for something that can be solved in a macro way elsewhere. If there's an exception to this, I haven't found it yet.

Our ability to hear equal temperament as "correct" is quite ingrained. Our auditory flexibility when it comes to unconsciously accepting and mentally correcting pitch anomalies is well known to people who really study this sort of thing. Where it often goes haywire is when people shift into hyper-listening mode and suddenly hear things they never noticed before, even though they'd been there all along.

[HoGo]

I've made several compensated nuts for b@njos and couple days ago my friend visited me with his HD28 with intonation problem on low E. The first couple of frets were sharp (even with gentle pressure on the string). I added just a small piece of bone to the nut at that string and now the guitar notes much better (even with his gorilla grip ).
My experience is that compensation at the nut makes more change at lower frets gradually vanishing up the neck while compensation at saddle makes difference mostly at higher frets.
It is about relative length of string. The change in ratio of nut to 1st fret vs. 1st fret to bridge distances is larger than the ratio calculated for let's say 12th fret... In other words you introduce error to the placement of all frets (they seem all to move towards nut), and the errors is relatively smaller up the neck. That's how I understand it, and it seems to work for me.
On mandolin with properly cut nut I don't feel any need to compensate. If the bridge is OK then the mandolin should note OK....

Adrian

[sunburst]

Paul, it's true that frets are sometimes incorrect, and that would be the first thing to check given the situation I suggested. I was trying to keep the exlanation as simple as possible by leaving out as many variables a possible, so, one string, one fret.

Hogo, I agree that the compensated nut introduces more errors toward the nut end of the fingerboard, and that the errors are relatively smaller than those introduced by a compensated bridge.
I think of changing the nut or bridge position of the string as introducing an error (at least mathematically), and the re-tuning made necessary by the error as compensation for that error. It takes about the same amount of string length adjustment whether done at the bridge or nut, but when done at the nut, it is a smaller percentage of the distance between frets, because the frets are farther apart at that end of the fret board. In other words, the frets get closer together as you go up the neck. A 1/16" (for example) difference is much less noticable at the first fret than at the 22nd.

I once took two banjos, same fret scale, one with a compensated bridge and one with a compensated nut, to a strobe tuner and "mapped out" all the notes on both. The "experiment" showed that the one with the compensated nut was more accurate in general, and especially on the higher frets.
Interestingly, there were frets where one string would be sharp and the adjacent strings, or other strings, would be flat. At another fret, the situation could reverse. It seems to me we are only make things tolerable by compensation at the bridge or at the nut. To really get things in tune, each string of a different gauge would have to have it's own fret scale.

[John Flynn]

I'm surprised that no one has mentioned the Buzz Feiten tuning system that Don McRostie uses on his Red Diamond mandolins. Nut compensation is a key part of the system. I know a few luthiers who either do it or have seriously looked into it. The verdict I've gotten is that it definitely works, but at least on a mando it may be overkill and may not be worth the effort and expense.

http://www.reddiamondmandolins.com/feitensystem.html

Also, what's with an instrument set up system that has the word "buzz" in the title? #

[sunburst]

Buzz does most of the compensation at the bridge, with the nut-to-first-fret distance shorter than would be mathematically correct. There is only one nut position different in a Feiten guitar nut, if I remember correctly.
The shorter distance to the first fret "compensates" for the sharpening of the notes on the first 3 frets that normally comes from bending the string over the front edge of the nut to stop it at those frets. The intonation for the rest of the instrument is adjusted at the bridge, and with tuning.

[Gavin Baird]

I think that I have posted this before however... The calculation of fret placement has an error that occurs between the nut and the 12th fret. after that the calculation is accurate. If you measure the distance from the nut to the 5th fret you should record 1/2 the distance to the 12th fret. As calculated you will find this is not the case. Numerous builders remove material from the end of the fret board, at the nut to correct this difference. For a mandolin I remove .031" to compensate for the inaccuracy. It works. For longer scale lengths the error is larger...Gavin...The 5th fret is the harmonic of the 12th.