Shaping First Saddle - Question

[ChrisN]

I'm shaping a new bone saddle to go with what I hope is a successful neck reset on my old Conn. The fretboard's radius is 15" so I shaped the blank's top to 15" so it matched the board. My goal is a 12th-fret action of 5/64 and 3/64, so a little closer than typical. My playing is not big strumming - just mild strumming, pick and finger.

I then checked my Erlewine book (always check after doing the thing you're working on, never before) and see that he says to leave the bass side higher/flatter to avoid string buzz. That seems to preclude following the fretboard radius.

1. Which is it? Do I want the saddle to match the board's radius? I like the strings to be the same distance from the fret tops, and that seems to be the way to do it. Erlewine's way would seem to leave the bass-er strings a bit higher from the fret than would a strict radius.

2. Having already radiused this blank, do I need to start over? Or is there a save? The blank is still its original height. I've just trimmed the ends and done the radius.

Thanks for any input.

[charles Tauber]

As with all things luthiery, different folks approach it differently.

My approach is that "saddle radius" is irrelevant. What matters is that each string is the desired height, say as measured at the 12th fret. For each string, if you measure the height you have and subtract the height you want, double the result and that is how much needs to be removed from the saddle where each string breaks over the saddle.

What results can be a stepped saddle with 6 different, discontinuous heights. Alternatively, one can join the 6 independent points in a smooth curve. That curve will NOT follow the curvature of the fingerboard: it will be taller at the bass side to accommodate a progressively higher string height for the bass strings.

Others do it differently and use a radius. Whatever gives you the string heights you desire is fine.

[ChrisN]

Originally Posted by charles Tauber

As with all things luthiery, different folks approach it differently.

My approach is that "saddle radius" is irrelevant. What matters is that each string is the desired height, say as measured at the 12th fret. For each string, if you measure the height you have and subtract the height you want, double the result and that is how much needs to be removed from the saddle where each string breaks over the saddle.

What results can be a stepped saddle with 6 different, discontinuous heights. Alternatively, one can join the 6 independent points in a smooth curve. That curve will NOT follow the curvature of the fingerboard: it will be taller at the bass side to accommodate a progressively higher string height for the bass strings.

Others do it differently and use a radius. Whatever gives you the string heights you desire is fine.

Thanks Charles. While I understand what you're saying, at this point, I don't understand how your method could arrive at a different string height at the 12th fret than would result from having the saddle radius follow the fretboard radius. It would seem to me the strings' bottoms would be the same heights from the fret tops, in either case.

I found a fellow who seems to say it's OK I've started mine the way I did:



While he doesn't go as far as your individual-string-height approach, he seems to get me where I need to be on this first effort without worrying about the taller/flatter bass string section Dan referenced. I understand I can't set saddle/string height and intonation until the neck angle is re-done first. Baby steps, but so far no stumbles that cause me to crack my head open on the corner of the coffee table!

[Howard Klepper]

Originally Posted by ChrisN

My goal is a 12th-fret action of 5/64 and 3/64 . . .

I like the strings to be the same distance from the fret tops . . .

How do you picture a setup that achieves both of these goals?

[charles Tauber]

Originally Posted by ChrisN

Thanks Charles. While I understand what you're saying, at this point, I don't understand how your method could arrive at a different string height at the 12th fret than would result from having the saddle radius follow the fretboard radius. It would seem to me the strings' bottoms would be the same heights from the fret tops, in either case.

Then I respectfully suggest you haven't understood what I described.

If you "follow" the radius, you will have uniform string height from 1st to last string. It can't be otherwise. That isn't what most people want: they want the treble strings lower than the bass strings. That means the saddle needs to be lower at the treble strings than bass strings. That means it is a different curvature than the surface of the fingerboard/frets. In Erlewine's language, that's a "flatter" radius on the bass side. It's simple geometry. It isn't complicated to achieve.

John uses a method where he butts the saddle against the end of the fingerboard and traces that contour onto the saddle, eliminating the need for radius gauges and/or measurement of that radius. Seems like an efficient approach.

[murrmac123]

IMO , the sooner we drop the whole "radius" thing in regard to saddle profile, the better. Btw the sooner we drop the ridiculous "64th"s measurement, even more better.

Conventionally, for most players, what they want is a drop of .020" between the E string action and the e string action. So, for example, if the E string action is .100" , then the e string action will be .080". If the E string action is stipulated to be .080", then the e string action would normally be .060".

The optimum method for the intermediate strings is to have an incremental drop in action of .004" per string ... so for example with a .100" E string action the A string action will be .096", the D string action will be .092", the G string action will be .088", the B string action will be .084", which takes us back to the e string which is of course .080".

Now, geometrically , this doesn't work with any kind of predetermined radius, and the unfortunate reality is that although you can set the saddle correctly for the E and the e strings right from the git-go, the optimum height for the intermediate strings can only be achieved by filing from the top of the saddle.

This can be done in situ, or it can also be done with the saddle off the guitar and held in a vise, if you can measure with sufficient accuracy.

Not to belabor the point, but the resultant optimum saddle top does not have any kind of definable "radius".

[ChrisN]

Originally Posted by Howard Klepper

How do you picture a setup that achieves both of these goals?

Good point, I left off part of my process. Once the strings were the same height, then I'd adjust the saddle bottom to achieve the differential. As mentioned below, I like the E and A a little higher because I do get a little buzz below 5/64.

[ChrisN]

Originally Posted by charles Tauber

Then I respectfully suggest you haven't understood what I described.

If you "follow" the radius, you will have uniform string height from 1st to last string. It can't be otherwise. That isn't what most people want: they want the treble strings lower than the bass strings. That means the saddle needs to be lower at the treble strings than bass strings. That means it is a different curvature than the surface of the fingerboard/frets. In Erlewine's language, that's a "flatter" radius on the bass side. It's simple geometry. It isn't complicated to achieve.

John uses a method where he butts the saddle against the end of the fingerboard and traces that contour onto the saddle, eliminating the need for radius gauges and/or measurement of that radius. Seems like an efficient approach.

My plan was to get uniform string height from 1st to last string, but then to adjust the saddle's bottom to get the tilt that gives me the 5/64 and 3/64. The top curve/radius would remain intact. This is my first attempt at making a saddle, but adjusting the saddle bottom is what I've done when lowering/adjusting saddles made by others. I assumed I'd do it the same way. I'm not familiar with the method of leaving the bottom square and adjusting string height at the top. That's a fine idea, I just hadn't heard of it before now.

I've got under-string radius gauges I used to radius the saddle's top.

[ChrisN]

Originally Posted by murrmac123

IMO , the sooner we drop the whole "radius" thing in regard to saddle profile, the better. Btw the sooner we drop the ridiculous "64th"s measurement, even more better.

Conventionally, for most players, what they want is a drop of .020" between the E string action and the e string action. So, for example, if the E string action is .100" , then the e string action will be .080". If the E string action is stipulated to be .080", then the e string action would normally be .060".

The optimum method for the intermediate strings is to have an incremental drop in action of .004" per string ... so for example with a .100" E string action the A string action will be .096", the D string action will be .092", the G string action will be .088", the B string action will be .084", which takes us back to the e string which is of course .080".

Now, geometrically , this doesn't work with any kind of predetermined radius, and the unfortunate reality is that although you can set the saddle correctly for the E and the e strings right from the git-go, the optimum height for the intermediate strings can only be achieved by filing from the top of the saddle.

This can be done in situ, or it can also be done with the saddle off the guitar and held in a vise, if you can measure with sufficient accuracy.

Not to belabor the point, but the resultant optimum saddle top does not have any kind of definable "radius".

It'll be more work, and I suspect the saddle top wouldn't be pretty, but I could try it that way. Thanks for the info.

[TNO]

Not an expert, but this is what I do- first adjust the trussrod and set the string height at the nut. Then, with the original saddle I check the string height at the 12th fret, looking for around 3/32 on the low E and 2/32 on the high E. What ever change you need to make just double that number at the saddle. Without getting to precious about it I make the saddle radius just a bit flatter than the neck radius. A cheap Harbor Freight belt sander can rough in a intonated saddle in about 2 minutes.

[brianhejh]

Hi Guys.

I learned a lot from this topic, thanks to all.

A related question: Does the same graduated or similar process IE- 100 TO 80 THOU across the strings need to apply at the nut? To this point I set my strings at 18 thou equally.

Thanks
Brian

[murrmac123]

Originally Posted by brianhejh

Hi Guys.

A related question: Does the same graduated or similar process IE- 100 TO 80 THOU across the strings need to apply at the nut?

No, it doesn't.

The correct height for the base of the nut slots is the same level as the tops of the first two frets.

[ChrisN]

I finished the saddle blank and put it into the slot. I radiused the top edges, but didn't otherwise reduce the height.

When I look at the blank installed in the slot, it appears lower on the bass side (less saddle is visible).
When I measure the slot, it's deeper on the bass side (relative to the top of the bridge).
When I measure the bridge height from the soundboard, it's thicker on the bass side.

The slot and bridge do not look like they've been modified, to me, but who knows.

In contrast, the bridge on my modern Taylor is the same height from the soundboard on both bass and treble.

I'm sure the saddle will work in the bridge, as is, but something doesn't seem right.

Any thoughts as to what might be going on? Or is this a normal bridge from the period/region?

EDIT: Huh - apparently it's not unusual for the bass side of the bridge to be thicker - http://www.acousticguitarforum.com/f...d.php?t=323319

Said Ned Milburn: "And it is very common (more common than not) that acoustic guitar bridges are thicker on the bass side. This is to allow for a consistent saddle protrusion so that the bass and treble strings have a similar break angle over the saddle, since the bass strings are set up with higher action than the treble."

Also, Kimsey said he goes for a .140-.150 mid-saddle exposure and suggested he could go to .180, but my new blank in this bridge provides only .138 max at the mid-saddle point. What's up with that? I know it's "close enough," but why is so little saddle exposed when I haven't adjusted the height yet? Should I have gotten a taller saddle? I bought the 6033v (shaped and 25/64 tall) from http://www.stewmac.com/Materials_and...e_Saddles.html Should I have gotten the unshaped 6030v with the taller 15/32 height? I suspect the answer is yes, but I think .138 exposure will work. Thanks for any input.

[John Arnold]

Many guitars (including Martins) have bridges that are thicker on the bass side. That facilitates a more uniform saddle protrusion when the action is adjusted higher on the bass side. If the action differential is 2/64, then the bridge thickness differential should be 4/64 (1/16").
My procedure for making a new saddle is to fit the saddle to the bottom of the slot first, then adjust the height from the top. I string the guitar up with a saddle that is slightly taller than the projected final height, then measure the action at the first and sixth strings. I then calculate how much I need to remove from the saddle to get the desired action. I remove the saddle, measure the total height at the first and sixth strings, then cut the saddle down to the calculated height. The fingerboard radius is used to connect the two points, but that is not a hard and fast rule. In general, if my saddle radius is different from the fingerboard, it is a bit less (more curvature), giving the middle strings a bit more height than would be consistent with a uniform gradation from bass to treble. That counteracts a tendency for the middle strings to buzz more than the outside strings.

[Truckjohn]

I set up the height of each string individually using precision feeler gages or pin gages. I find this allows me to really understand where I am at and hit a much more consistent setup.

Starting with the same radius as the fretboard generally saves a lot of time.. But the precision of a pencil mark is what it is...

In general - I typically end up with low E will be slightly higher than the high e string at both the nut and saddle. The rest of the strings are progressively closer to the frets. At the nut - it may only be 0.006" higher - but it's still a bit.