Nut slotting files for bone bridges

[Simon Barden]

I'm just trying to get across that there is very little hard science available on the subject and it's easy for what 'seems like a good idea' to become the perceived knowledge without any hard evidence.

There's also a lot of other factors like string break angle and the angle of the nut slot that can cause a buzzy nut, and these are probably the more likely cause of it than the shape of the bottom of the slot, but everything interacts, so nothing can really be taken in isolation.

I need to have a look at all my Hosco files, both the originals and the new black-bladed set, to see if their shapes vary at all, or they are all like the 0.050" file. I chose that one as it is the new addition to plug a gap and was unused, so there was no chance of wear modifying the shape of the file.

[fender3x]

A couple of things have occurred to me as I read back through these posts. First, I think Simon's point is well taken that we don't seem to have much empirical data to draw on here. Mostly we are just talking about theories. Those are only as good as the assumptions they are based on.

I thought about this when reading about "transfer" of sound by the nut. I have been working off the assumption that the nut's job is to keep as much of the vibration of the string in the string. That is, not to transfer vibrations, but to isolate them to the string. For this purpose you want the nut to be very hard, but it shouldn't matter much what shape the bottom of the nut is as long as all movement of the string is over the fretboard. By the same token you'd want the sides of the nut channel to keep the string from moving laterally. None of that argues one way or the other for a flat or curved bottom, so I am guessing the only way to know for sure would be with empirical testing.

The other assumption that I wonder about is if the file makes that much difference in the bottom of the slot, at least after a little while. I think it's reasonable to assume that tuning and playing will wear the string channel into a round bottom, regardless of the shape of the file. I wonder how much difference there would be in a channel cut with a radiused file vs a flat file after a little playing?

[Simon Barden]

The wounds strings are more likely to wear the bottom of the nut slot into a shape of their own choosing. The plain strings, less so, being much smoother. They probably have more effect when old and corroded.

Exact movement of the string over the nut is a bit of an unknown. The physics-style descriptions of open and fretted strings and the increase in string tension when fretted (and hence part of the need for saddle intonation) all seem to only mention the length between the nut and the saddle, not the lengths on the headstock or after the saddles. Yet when you turn a tuning peg, the whole string's tension increases, so surely you need to consider those extra lengths of string?

Basic physics and Young's modulus shows it takes less force to stretch a long string than a short string (given the strings have the same diameter and composition), so surely the tension increase calculations should take into account those extra lengths of string?

Or should they? This is where friction seems to come into it. Under my new digital microscope, I simply can't see the string moving on the headstock side of the nut when I fret a string. I know I have a very low action on all my guitars so it's not a big change in string tension, but there is some change. But there is something going on, for if I pick a headstock string length and measure the note the string makes on my Petersen strobe app on my phone, then fretting a string and increasing the string tension on the string actually causes the headstock note to fall by a few cents, not rise as you'd expect if the string tension increased!

So what is going on? My hypothesis is that two things are happening. One is that the friction of the string in the nut is enough to hold it in place so that small variations in string tension are not passed to the headstock (and probably the post-bridge length) and the other is that the increase in tension is lifting the headstock side of the string slightly off the rear face of the nut, so the string length increases slightly, and as it's now effectively slightly longer, its pitch drops. The first part I am now quite sure about, the second is just guesswork, using the only mechanism I can think of that might explain the drop in pitch.

For the first part, I think we need to look at good old friction at work, and the difference between static and dynamic friction. It always takes more force to get an object resting on a surface moving, than it does to keep it moving. Sometimes people use the term 'stiction' to describe this effect.

With the string held in the nut slot by downward string tension, it will take a certain amount of lengthways force to start the string moving. Until that point is reached, the tension increase stays within the nut to saddle length. But if I then bend the string, rather than just fret it, then the string tension increases enough to overcomes the static friction, the string is able to move in the nut and the tension distributes itself all along the string length (or at least at the headstock end) and the pitch of the headstock string length then rises.

However, a very small movement of a tuning key, with a well cut nut, can cause what appear to be instant changes in string pitch, (though with a sticky/badly cut nut this can be very jumpy). So what's going on here?

Given that the nut has some friction and can in theory separate the headstock length from the neck length of string until the static friction is overcome, means that you could view the headstock length as being stretched first by the tuner and so increasing its tension, which once there's enough tension so that the static friction is overcome, allows the tension change to distribute itself along the string and the pitch goes up (or down).

Say the headstock length on a top E string is 30mm, then 1mm of tuner rotation gives a 1 in 30 change in headstock string length*, whilst a 1mm change in a 648mm scale length through fretting is a 1 in 648 change in length. As the amount of force required is proportional to the string length, a 1mm change at the headstock length generates roughly 22 times the force that fretting the string does (this will reduce with higher string actions as this generates a bigger change in string tension), so you should be able to appreciate that it's easy, using the tuner, to build up enough tension to unstick the string from the nut, get it moving and so tune the string. Whilst a 1mm increase in string length on the neck from fretting (and 1mm is probably quite a large amount and would need a fairly high action to achieve), on its own doesn't produce enough force to unstick the string.

*I know the actual tuner to nut length length actually stays the same and the cross sectional area of the string gets smaller or expands as well as the tension increasing or decreasing, but the 1mm still either gets wound on or off the tuner post and I'm trying to keep the explanation simple.

So when tuning you get a build up of tension in the headstock length until there's enough force to get the string moving over the nut. Once moving, the string friction is less and so the string keeps moving until the tension along the string is even (or very close to) at which point the string stops moving and 'sticks' again to the nut.

When a string goes 'ping' when you're tuning it due to a dodgy nut slot that's probably too tight or very rough, you are hearing this principle in action. The headstock length is building up a tension difference across the nut (be it positive or negative depending on tuning up or down), but the nut is gripping the string too tightly for it to move as it should do. So you need to turn the tuner more, and at some point the force is sufficient to unstick the string, which is released with enough force and speed to set the headstock length of string vibrating, and hence the 'ping' noise.

With a well cut nut, the same thing is happening, but on a much smaller scale and a much faster time frame. As the force required to unstick the string is far, far less and can be generated by a very small movement of the tuner (and not say 1/4 turn), not enough energy is released to get the headstock length vibrating at an audible amplitude and all seems smooth and quiet.

Rather like a swan serenely gliding along the top of the water, when under the water its little legs are paddling furiously.

[JimC]

The physics of friction and nut slots are probably even more complicated. I don't know about you folks, but I kinda assume that the nut and the string are completely rigid, but on the scale we are talking about that's probably not true. I had my mind boggled recently when it was pointed out to me that the size contact area between the steel tyre on a railway wheel and the steel rail is proportional only to the weight, and not significantly affected by the diameter of the wheel, because tyre and rail deform elastically until equilibrium. It seems feasible, then, that a similar effect could be happening between string and nut, especially if the contact area is very small, as it would be with a V or flat bottomed slot that makes contact only at one end of the nut. I shouldn't be surprised if the amount of elastic deformation affects friction/stiction too, but who knows. Moral - lubricate the damn thing!

[ross.pearson]

Wow, this turned into a good conversation! I'm eyeballing some Uo-Chikyu Hiroshima file sets now. I found some for AU$125 inc. delivery. Better yet they are a super light set which I've not been able to find elsewhere (yes, I play super lights, I gots weak web developer fingers).
https://www.plazajapan.com/upc-4/?setCurrencyId=1

For that price I would even consider getting a couple different sets to get some variety.

In regards to rigidity, I'll probably make a file holder a'la the stewmac one.

https://www.stewmac.com/luthier-tool...t-file-backer/

[McCreed]

Even if you play 9's, you'll want to cut the slot with a .010" file.
I'd get the 10-56 set (which is what I have). You can easily do 9-42,10-46, 10-52, 12-56 whatever...
You cut the slot a couple of thousandths above the string gauge.

At least that's the way I was taught, but I'm sure opinions will differ.

As for the rigidity and file backers, I've not had any issues with even the two finer gauges of my files (.010", .013"). And that's using them in bone. They might have some benefit in the comfort of holding the file, but again, not an issue that I've encountered.

[Simon Barden]

Wow, this turned into a good conversation! I'm eyeballing some Uo-Chikyu Hiroshima file sets now. I found some for AU$125 inc. delivery. Better yet they are a super light set which I've not been able to find elsewhere (yes, I play super lights, I gots weak web developer fingers).
https://www.plazajapan.com/upc-4/?setCurrencyId=1

For that price I would even consider getting a couple different sets to get some variety.

In regards to rigidity, I'll probably make a file holder a'la the stewmac one.

https://www.stewmac.com/luthier-tool...t-file-backer/

Hosco do a set of files for 9s: https://www.hosco.co.jp/en/luthier-t...-nf-eg009.html

In the UK I can get them for £75 delivered and that includes a holder, though I have also seen them listed for over £100, which makes them more expensive than the regular Hosco range.

Though I would still use a set for 10s to cut the slots for 9s. Not that I think it's wrong to use the same sized files as the string gauge, just that it allows you to fit 10s without worry if you ever change your mind or want to sell it on and you don't want the new owner moaning they can't fit their set of 10s on. I use the same size nut file as the string gauge if there's one in the set, or the nearest size up and don't have any issues. I haven't measured them but I think they may be slightly wider than the stated size to prevent sticking.

Note that the StewMac file holder (127mm long overall) is shorter than those Uo-Chikyu files (150mm), so I don't think they will fit, as the holder appears to have closed ends to the slot.

My Hosco 10 file is quite bendy, but is fine for existing slots. If I ever need to cut a blank from scratch, then I'll start with a 13 until there's enough of a groove to keep the thinner file centred and then carry on with that. The nut will invariably be too high and need filling down, so the wider slot section will vanish.

[Simon Barden]

Moral - lubricate the damn thing!

Graphite is probably the best and safest slot lubricant. With bone, you need to be careful if using a grease-based lubricant as the grease will gradually travel along the hollow tube structure within the bone and discolour it.

[ross.pearson]

Note that the StewMac file holder (127mm long overall) is shorter than those Uo-Chikyu files (150mm), so I don't think they will fit, as the holder appears to have closed ends to the slot.

Not an issue if I'm making my own though =). Just a scrap of wood, shim stock, and some grub screws.

I'll likely hold off until the new house is built and I'm in my own shed however, unless the bug to build another guitar hits pretty strong between now and then.

[fender3x]

Moot point now for Ross, but in case anyone is still interested in this topic, I bought the smallest, cheapest jeweler's saw frame that I found on eBay. The frame is 2.34" and set me back $11.

I tried it using the smallest tip cleaner that I have. It flexed like crazy no matter how tight I got the frame. I have to conclude that the tip cleaners flex too much to do a guitar nut. I don't see how you could do this without making a curved slot, either with or without the saw frame.

I will say that I think I will be able to use it for the next bass nut that I cut. The bass-sized tip cleaners are much stiffer, and seem to work well in the frame. So both tip cleaner and saw frame seem viable for bass nuts. Still slow. But viable.

A bit of a surprise is that my feeler gauge saw blade worked pretty well in the jeweler's frame too.



I was able to cut a slot in bone pretty well with the .009 "blade" in the saw frame. Not sure if this is any faster, and the bottom of the blade is still flat, but it seemed to work reasonably well.

I am not sure if cutting is any faster with the frame, but it is certainly easier on the hands.

I have no idea what the protuberance on the bottom left of the saw is for. It seems to have no function. Eye-balling my guitars it does not look like it would be in the way of cutting the nut, but I didn't try this with a nut that was glued in place. It might need to be trimmed if it gets in the way.

Offered FWIW. When people would ask if the beer we made in college was any good, our stock answer was "It'll get you drunk." I feel similarly about this tool. It'll get the job done, but I'll still envy the guys with decent quality nut files.