Break angle article in SUS August newsletter

[RogerHaggstrom]

Originally Posted by zmf

This point -- "The saddle actually acts like a filter - it only allows some of the vibrations through." -- seems not to be brought up that often (I probably missed the references), but seems obvious in any discussion of saddle material.

Different saddle material (Tusq, ivory, bone) will have different spectral filter properties.

I have been deep into that with my segmented saddles with soft wood between posts of bone under the strings. Different woods, in my case spruce or cedar, between the post gives different tone. The filtering of the full frequency recipe coming from the strings in the saddle is a big player when it comes to the shaping of the tone of a guitar.

[Robin, Wales]

Originally Posted by RogerHaggstrom

I have been deep into that with my segmented saddles with soft wood between posts of bone under the strings. Different woods, in my case spruce or cedar, between the post gives different tone. The filtering of the full frequency recipe coming from the strings in the saddle is a big player when it comes to the shaping of the tone of a guitar.

I have done a lot of playing around with resonator guitars and mountain dulcimers, changing the saddle material to greatly effect the timbre. My latest has been to swap the bridge on my Godin archtop from Tusq (lots of top) to ebony (still lots of top) to hard Indian rosewood (better) to the soft "rosewood" of an Ebay cheapy (the best by far!!!).

Also, I have played around with saddle shapes on guitar and (from a small sample of two guitars) a bone saddle with a rounded non-compensated top seemed to mellow out and strengthen the plain strings - taking away the edgy twang. I think that many compensated saddles leave just a small edge for the b and e strings to run across, and this is often at the outer edges of the saddle - I think that this may filter out the bass compared with running the two plain strings off the wide rounded top of the centre of the saddle. So, if your saddle slot is well positioned so you can get away without compensation for the b string and e string, then I would certainly try a round topped "old style" saddle to see what that does to the timbre.

I'm going to sit down and make one for my D-18 when I have a bit of time. It would have been the saddle shape that they would have come fitted with back in the day. And I think it is the saddle shape that they are putting on the Authentic Series. It wouldn't be the first time that I had found an instrument that folks have said sounds wonderful only to realise that the wonderful sound was actually because of the saddle, rather than because of the more expensive bits of the build - and a few changes to a cheaper version makes that sound wonderful too!

[zmf]

Originally Posted by Robin, Wales

I think that many compensated saddles leave just a small edge for the b and e strings to run across, and this is often at the outer edges of the saddle - I think that this may filter out the bass compared with running the two plain strings off the wide rounded top of the centre of the saddle.

Interesting. I was under the impression (which means I read it on the internet) that a rounded saddle could cause that nasty sitar-like effect, and was to be avoided.

I'm guessing it's a matter of the extent of the rounding?

[Robin, Wales]

Originally Posted by zmf

Interesting. I was under the impression (which means I read it on the internet) that a rounded saddle could cause that nasty sitar-like effect, and was to be avoided.

I'm guessing it's a matter of the extent of the rounding?

Yes, it is the extent of the rounding. A sitar saddle is an inch or more wide, and it is actually quite difficult to get the radius shallow enough and with the correct profile on purpose to get a good sitar sound (I have tried!!!).

Frets are round topped and sing quite sweetly - you can do the same with a saddle - it is when you leave an area flat or nearly flat that you are more likely to get "sitar".

[Alan Carruth]

I skimmed most of that Graph-Tech article: it's good, but they're trying to sell you a fancy saddle...

Impedance is a measure of how hard it is to move something : it's the ratio of Force/velocity at a given frequency. Strings are simple structures and the impedance of the string can be calculated fairly easily: it's low at the resonant pitch and partials ('harmonics') and higher everywhere else. Guitar tops are more complicated; they can be modeled on computers pretty well, but for most of us it's just easier to build one and measure what you want. Any guitar top will have a number of resonances, which are not simply related in terms of pitch. Also, most of them have 'nodes', inactive lines that separate the active vibrating areas, that can run through the bridge. Impedance is high at nodes, so it can be easier or harder for a string to drive a top resonance even if it's tuned to that pitch.

The first diagram in that Graph-Tech post showed a string tied to a support that had zero impedance: not matched to the string, but lower. The reflected wave was in the same phase as the incident wave: the incoming wave was 'up', and so was the reflection. When the impedance match is the other way, low-to-high, the reflected wave is in the opposite phase; up-to-down. When the impedance match is perfect the wave just keeps going, and nothing is reflected at the boundary.

Since the impedance of both the strings and the top varies in more or less complicated ways with frequency and location along the bridge the amount of sound that gets reflected back into the string vs passed through to the top varies a lot. In general, though, we want to see a bridge impedance that's much higher than the top impedance, so that most of the energy stays in the string. This is what defines the pitch of the string, and also the sustain of the note. Most of this mis-match comes from the mass and stiffness of the top at the bridge.

I once tried to measure the difference in sound of a particular guitar when I swapped the bone saddle for one made of high density polyethylene. I could hear a difference, but the setup I had at the time couldn't see it. It's probably mostly in the higher frequency range, between 2000-4000 Hz, where normal hearing acuity is highest.

It's pretty common for the 'main air' resonance (actually the lower half of a 'bass reflex couple') to be close in pitch to the G on the low E string, third fret. Air pressure changes in the box cause the top and bridge to move a lot, and the impedance of the top system is very low at this pitch. It's close enough to the string impedance so that the energy gets out of the string and into the top, and turns into sound, very quickly. You get a note that's twice as powerful for half as long. We're not very good at hearing differences in loudness, but the lack of sustain really shows up. Also, the lack of feedback between the string and the bridge means that the pitch of the note is not as well defined, so it sounds 'thuddy'.

[AcousticDreams]

Originally Posted by RogerHaggstrom

I have been deep into that with my segmented saddles with soft wood between posts of bone under the strings. Different woods, in my case spruce or cedar, between the post gives different tone. The filtering of the full frequency recipe coming from the strings in the saddle is a big player when it comes to the shaping of the tone of a guitar.

I have always enjoyed your experimentation's Roger. Please post some more pictures of your latest Segmented saddles. I have on occasion used two different bone materials...one for the High E and B string...and a different material for the 3rd, 4th, 5th & 6th string. Mostly now I use only Unbleached bone. But do hope to experiment more with your segmented saddles and or Laminated saddles combining different elements.

Originally Posted by Robin, Wales

I have done a lot of playing around with resonator guitars and mountain dulcimers, changing the saddle material to greatly effect the timbre. My latest has been to swap the bridge on my Godin archtop from Tusq (lots of top) to ebony (still lots of top) to hard Indian rosewood (better) to the soft "rosewood" of an Ebay cheapy (the best by far!!!).

Also, I have played around with saddle shapes on guitar and (from a small sample of two guitars) a bone saddle with a rounded non-compensated top seemed to mellow out and strengthen the plain strings - taking away the edgy twang. I think that many compensated saddles leave just a small edge for the b and e strings to run across, and this is often at the outer edges of the saddle - I think that this may filter out the bass compared with running the two plain strings off the wide rounded top of the centre of the saddle. So, if your saddle slot is well positioned so you can get away without compensation for the b string and e string, then I would certainly try a round topped "old style" saddle to see what that does to the timbre.

Very excited to hear about your experimentation with rounded tops for E & B. Would you be able to post some pictures? While I do need compensation...I have some thoughts about a possible way to still round over.

[Robin, Wales]

Originally Posted by Knives&Guitars

Very excited to hear about your experimentation with rounded tops for E & B. Would you be able to post some pictures? While I do need compensation...I have some thoughts about a possible way to still round over.

Here's one I made from an oversized blank.



Another advantage is that it is easy to work on the top of the saddle as well as the bottom. We are very used to taking material off the bottom of compensated saddles to get the action to the right height. This is because the tops are too complex to adjust. However, with a rounded top it is really easy to take a little off and re-round and polish the top. This is very useful when the saddle radius profile would benefit from being just a little tighter than the fretboard radius, so you can set the action of the 6th, 5th, 2nd and 1st strings a little lower compared to the 3rd and 4th strings. Small adjustments to the saddle top are easy to make.

I play a lot with a capo, so any saddle compensation set for open strings is thrown off anyway by the capo. Therefore it is actually more straightforward for me to tune and play without specific string compensation beyond the generalised slant of the saddle slot in the bridge.

[zmf]

Originally Posted by RogerHaggstrom

I have been deep into that with my segmented saddles with soft wood between posts of bone under the strings. Different woods, in my case spruce or cedar, between the post gives different tone.

Roger -- just wanted to make sure I understand what you're up to with the saddle.

You have "posts" of (low impedence) bone under the strings, with each post separated by soft (high impedence) wood.

Is the purpose to localize/keep separate the energy transfer from each string? Not sure of the purpose of the soft wood between the bone posts -- why not just empty spaces between the bone posts? Is the soft wood coupled to the bone posts and contribute to energy transfer?

Thanks.

[RogerHaggstrom]

Originally Posted by zmf

Roger -- just wanted to make sure I understand what you're up to with the saddle.

You have "posts" of (low impedence) bone under the strings, with each post separated by soft (high impedence) wood.

Is the purpose to localize/keep separate the energy transfer from each string? Not sure of the purpose of the soft wood between the bone posts -- why not just empty spaces between the bone posts? Is the soft wood coupled to the bone posts and contribute to energy transfer?

Thanks.

I did write about the segmented saddle in the American Lutherie #144 in some detail. Here is a slide of a recent one with spruce in between bone posts in a tight-fitting saddle ditch.



What I get from it is better string separation, lower weight than a solid thick bone saddle (a thick saddle is needed for my nut intonation) and a change in timbre in a good way from the wood in between the bone posts.

One obvious thing is that each string will be less coupled to the other strings, making each string in a chord stand out for itself with better separation. With a thick bone saddle the saddle also acts a little bit like a limiter, the headroom with a segmented saddle is greater. The segmented saddle with spruce as a filter will pass through some of the frequencies that works well with the spruce top, making for a better timbre.

The attack and volume will decrease a little and the guitar will be less aggressive - not to everyone's taste but very sweet sounding. I found a way to counteract that and get all the good stuff from the segmented saddle and still have a guitar with high volume and aggressive attack, I talk about that in the article too.

I did try having nothing between the posts. I got the separation and better headroom, but no change (for the better as with spruce/cedar) in timbre.

[Alan Carruth]

zmf wrote:
"You have "posts" of (low impedence) bone under the strings, with each post separated by soft (high impedence) wood."

Bone, as a material, is both fairly dense and quite stiff, so at a given size it has higher impedance than the same size piece of wood, or plastic. The soft wood pieces in between have much lower impedance. More to the point, the soft wood also is more compliant, particularly at low frequencies, so it could allow the bone pieces to move sideways a bit without transmitting much force to the next piece of bone; hence the greater isolation of the strings. Individually the small pieces of bone are fairly easy to move, but they're in slots which restrict the motion, and sitting on the bridge, so they probably have similar impedance to a normal bone saddle in the 'vertical' direction, but not so much 'horizontally' across the top.

It took me a long time to get the concept of impedance clear in my mind...

[zmf]

Originally Posted by Alan Carruth

It took me a long time to get the concept of impedance clear in my mind...

Thanks for the correction, Alan. Looks like I've got more homework to do.

I was attempting to describe the relative ability to transmit energy, which I assumed would be greater for bone than soft wood.

[RogerHaggstrom]

Originally Posted by Alan Carruth

More to the point, the soft wood also is more compliant, particularly at low frequencies, so it could allow the bone pieces to move sideways a bit without transmitting much force to the next piece of bone; hence the greater isolation of the strings. Individually the small pieces of bone are fairly easy to move, but they're in slots which restrict the motion, and sitting on the bridge, so they probably have similar impedance to a normal bone saddle in the 'vertical' direction, but not so much 'horizontally' across the top.

Yes, that's the way I think it works. The spruce will filter the string's frequency recipe differently than the bone post and change the timbre. When I did my experiments, I could hear a not so subtle difference in timbre between spruce and cedar. Other harder wood did not sound that great, the wood needs to be soft to get the nice separation of the strings, allowing the bone post to move sideways.

Comparing cedar and spruce, I could hear "more of everything" with spruce, cedar made the sound "darker" and not as loud. I guess the filtering of the spruce in the saddle was a better match for the spruce top.

[KevinH]

Originally Posted by Robin, Wales

...So, bringing this back to the pragmatic, how do you think that this "pulse" string movement is going to effect guitar set-up?...

I just posted what is at least a partial answer to your question here:
https://www.acousticguitarforum.com/...d.php?t=651350

The summary is that it results in a rule of thumb for setting up a guitar to get the lowest possible string height w/o buzz, namely that the relief should be 1/6 the action at the 12th fret.

There are assumptions buried in there: the frets are level, the neck angle has been set correctly, strings behave ideally (e.g. bending stiffness is ignored), the neck relief profile is a circular arc. But I think its a good starting point, and will affect how I approach my setups in the future.

Originally Posted by Robin, Wales

...I flatpick from open chord shapes - usually with a capo at frets 2, 3, or 4. I use medium gauge strings. My "problem child" has always been the D string on frets 2 to 5. My set-up on my two 24.75" scale guitars is around 0.008 relief at fret 7 and an action of 2.5mm bass to 2mm treble at the 12th fret. I have 14" radius saddles on 16" radius fretboards. Those two guitars play cleanly.

I have just got a second hand D-18 (frets seem to be level when checked with a rocker) and I'm struggling to get that D string clean on frets 2-5. I have about 0.010 or so relief and an action of 2.7mm bass to 2.1 treble. But it is a longer scale so I'm thinking of upping the relief and lowering the saddle - perhaps to 0.013 relief (Frank Ford's recommended average) and bringing the 12th fret action down to 2.5mm bass to 2mm treble. I already have a 14" radius saddle fitted (16" radius fretboard).

I wonder if, because of the longer scale, I'm getting more "pulse" displacement as well as less space above the adjacent fret between the 2nd and 3rd, or 3rd and 4th, or 4th and 5th frets than on my shorter scale guitars? Therefore, I may need more relief for my playing style on that particular guitar. I'm pretty heavy handed with a thick rounded pick and Frank Ford suggests up to 0.026 relief for some bluegrass players - I don't think that I need to go anywhere near that amount.

Not quite sure how to answer that. I made a model that shows how close a plucked string gets to each fret given all the "inputs": pluck force and location, scale length, string tension, action at the 12th, neck relief, and probably some others I'm forgetting at the moment. If I use the values for your 24.75" scale guitars and set the plucking force to just below what would cause buzz, then apply that same force to your D-18 (with all of its parameters), the string clearances actually get a little better.

Even though the scale length is greater for the D-18, the string tension goes up and the initial displacement of the string goes down a small amount compared to the shorter scale guitars. Add to that the fact that you have greater relief and action on the D-18 and I would have thought the D-18 less likely to buzz. Maybe there is something else going on, like the neck profile has a different shape on the D-18? Or is it possible you're playing more aggressively on the D-18?

But, back to your thought about lowering the saddle a bit and increasing relief, that should make things better on the D-18, and it puts you closer to the "ideal" relief (action@12th/6) which would be 0.015".

I'd be interested to hear if that works out.

[printer2]

Not much to add, others have done a great deal of research. I did find when I did not have a saddle to use I made a temporary one out of aluminum. Strangely it gave the guitar a metallic tone.