Resonate Frequency Problem ?

[Gary Gallier]

I've got a problem on a finished ready to ship instrument and need some help. No, it's not a guitar, but the mechanics are exactly the same, so posting here. I have been building high end fusion instruments for 40 years, Mountain Dulcimer with a guitar pin bridge, uncoupled from the fretboard staff. Please refer to my webpage for a look, www.DulcimerCentral.com In that time I have had this happen maybe 3 or 4 times, with no clue on correcting or what the cause is other than the resonate frequency of the box cancelling the note. This is a fully chromatic instrument. I've got a note, b4 493.88Hz, that barks a bit louder and dies out sooner than the adjacent chromatic frets. This is occurring on the 2 unwound strings, different frets same note, only 3 strings tuned dAD. Not much tension, so bracing is not as heavy as guitar, but advanced X braced pattern as a guitar. Frets are well seated, joinery for bridge attachment is very good, not flat spots on the saddle. Notes are clear, no buzzes. Due to the size of the instrument, and corresponding sound hole, I cannot reach with my fingertips into the instrument further than an inch in front of the leading edge of the bridge. The only way I can work on this on the inside is to remove the back. I do have wood from the same board, so not really a problem but... Need some brainstorming ideas on how to solve this. Admittedly I am quite obsessive about balance up down and across all strings. I am also a high level player...I would not be satisfied with this. The problem barely crosses the line for me, so who knows of customer would notice...I notice. I just don't see what I might be able to do from the outside.
SO, has this happened to you on a finished build, and what can I do assuming I remove the back ? If you know someone well versed on this type problem, feel free to tell me who if you think they might not see this post. Or should I cross post somewhere else on this forum ? Thanks so much in advance. Except for this problem, the instrument is great.
Gary Gallier
www.DulcimerCentral.com

[Brian Evans]

I hope that Allan Carruth chimes in. He has done more research on resonant frequency issues than most. Have you made an identical design that did not have this problem? Can you determine the air resonant frequency of the box? Often the resonant frequency can be influence by the size and placement of the sound holes. With acoustic guitars it's kind of understood that all instruments have a resonant frequency in the range of play, and that a dead note may occur as a result, I have influenced this by masking or closing off part of a sound hole. Intuitively B4 is quite high in pitch, and the resonance may be quite peaky, maybe if you tune up or down a quarter tone you might avoid it?

[Alain Lambert]

Here is my 2 cents: Seems to me that the B4 is also one of the frequency of the instrument or the Helmholtz and therefore cancelling the note. You need to shift this frequency of the instrument a little bit. Try blocking the sound hole. Does is also cancell? If not may be changing the sound hole size (adding an inside ring, etc.) could solve it. If the problem is still there with the sound hole blocked, then it is coming from the box. As you say there is not much you can do from outside. Try changing the bridge weight, which may change the box frequency just enough. Put some weight on the bridge with "fun tack" and see if this make any difference. If it does, then adding or removing weight from the bridge might do the trick. Even replacing the pins with heavier ones like bone or brass.

[Mark Wybierala]

I've had resonant issues and they are evil. I wouldn't assume anything and simply try to rule things out one at a time. I like the idea of trying to add weight to the bridge. Its just difficult to maintain objectivity. Maybe feed some foam rubber strip through the sound hole to change the volume of air inside the instrument in order to identify the problem as internal volume resonance rather than top resonance. Light clamping of the top with a cam clamp to see if you can change or modify the response. Replace the saddle with a different material, corian, or tusq... I doubt string gauge will address the problem as you've clearly identified it as being a frequency problem by duplicating it on two different strings and fret positions. You might get away with sneaking a short top brace through a sound hole to tame the top. It sounds "out there" but since its a non-standard instrument anyway -- a sound post? --- something like a bridge doctor??

[Brian Evans]

I was looking around and reading a bit about this, and found this post, that I thought had value - particularly as the frequency was close to yours:

"about 10 years ago i had a conde flamenco that had a woofy note around 440 with the 4th and 5th string a notes being the culprits. i took the guitar to al carruth who did his blu-tac diagnosis and then shaved a little off a strut and all was well with the conde. i had the guitar for about a year thereafter and the treatment remained effective. al explained to me that too many components were resonating around 440 so he lowered the body resonance to about 436 or thereabouts. al, i hope i did you treatment justice."

[Alan Carruth]

It does sound to me like the classic guitar-type 'wolf' note. There is a strong resonance (on the guitar the Helmholtz-like 'main air' resonance) at that pitch. It pushes on the top and vice versa, so it's strongly coupled to the string. All of the energy 'leaks' out of the string in a hurry, and is turned into sound, so you get a note that's twice as loud for half as long. The instrument is just too darn efficient at that pitch.

The remedies are to either alter the resonant pitch or do something that makes the bridge a bit harder to move. Altering the resonance works better in terms of preserving the power of the sound, but it can be hard to do. Adding either mass or stiffness at the bridge will slow down the energy transfer from the string to the top, and can also alter the resonance in a helpful way. Ideally, you'd try to get the resonance to be between played notes. If you're not tracking that as part of your process it's sort of the luck of the draw: the ones that don't have this problem may have just happened to end up with the resonant frequencies better placed.

There are a number of ways to keep track of resonant frequencies, many based on Fourier transform computer programs. A friend on mine recently built an Android app called 'Luthier Lab' that includes a spectrum analysis module, and runs nicely on a tablet. You can probably find it with a search. Best of all, he figured out that there's not enough of a market to warrant trying to sell it, so he's just giving it away. Check it out.

[Alain Lambert]

I made one such software also and if you own a PC with a microphone you can download a working copy from here:
https://alainlambert.wixsite.com/luther ... m-analyser
This will help you decide if you should try to go up or down with your frequency change.

[Gary Gallier]

Thank you Mr. Carruth, and all who have responded. I can’t figure out how to reply individually, so summerizing here. To Brian Evans, yes, I have built approx 150 instruments of this design and do try to be exacting in repeatability. This is Very rare to happen. I do not have the math or tech skills to measure resonate frequency. My own efforts...I completely covered the soundhole with no improvement. Great idea on detune, I will do that incrementally as an analysis tool today to get a better idea if and where the problem goes away. Yes, by it’s nature a dulcimer is very midrange peaky, this note is in the middle third of the first octave on the treble string. Alain...My bridge design does not lend itself to removing mass. It’s as light as I feel comfortable with to be structually sound. Done for better response as part of the design. My instruments are significally more full and louder than a trad design in part because of that. I will order some brass bridge pins today ! Might be a simple fix. Mark, I did stuff a wadded paper towel “wall” at the waist between the upper and lower bout. The fretboard staff over the upper bout does transmit vibration into the upper bout however, and I didn’t get a change, or almost imperceptible if I did. The bridge end is much more of the business end than the upper bout however. Alan Carruth , thanks so much for your expertise in this type of thing. I will experiment with the tacky putty approach. Aside from what I have mentioned above, I am pretty resigned that I will need to work from the inside, which includes removing the back...I’ll need to replace it, I’m not good enough to remove it unscathed. But, not hard or dangerous, just Father Time, and I have the same board for a new back. Here is a tidbit on dulcimers. My inhanced volume is important, but not as important as sustain. This is where it is unacceptable...sweetness of the note rather than harsh banjo like. I’m afraid I am a bit more intuitive than technical in assumptions, and the responses generated some assumptions. Please feel free to say if I am right or wrong. It seems to me the resonate frequency of the box needs to be treated as a unit, which would include internal air volume, and stiffness of the back as well as the top and bridge movement you mentioned. In removing the back, it is no problem for me to reduce the side depth as well...maybe 1/8” or so all along the length, thereby reducing internal air volume. Would this have a likely effect on resonate frequency ? I assume it would raise it some? Sitka Spruce for the top, I used African Mahogany, or so the wood supplier said, for back/sides but it does seem a bit less dense than some mahogany I have used. Would the less dense wood with stiffer bracing on the new back have a positive effect on moving resonate frequency? My top bracing is light but plenty for structural support. Over time I have had zero problems with belly behind or cave in front of the bridge. 3/16 “ thick x 3/10” high, not tapered except within 1” on the ends next to the liner. Based on your comments I am assuming that beefing up this bracing around the footprint of the bridge is probably a good idea, to make the bridge harder to move as you stated. Again, sutain at that note is my biggest problem. Ok...I suppose I could beat this to death with maybe this and maybe that, so enough for now. I can’t tell you, and everyone else, how much I appreciate your continued advice and insights on this. So any additions you would be knd enough to add are greatly appreciated. I will update as appropriate.

[Mark Wybierala]

I can certainly relate to the negative thought of having to remove the back. It is so much easier to build an instrument than it is to repair one. Consider just hanging it up an letting the wood age for a year or two. Maybe humidity changes and settling of the joints under tension will improve the situation enough to move the instrument back over the line. I often come up with new ideas if I just walk away for a while.

[Gary Gallier]

New Information - the problem is very specific to the note, I'd say narrow. From open D tuning, dAD, if I tune 20 Cents sharp or 20 cents flat, the problem disappears. So...now I know exactly what to do ! NOT. Perhaps the brass bridge pins will do the trick. Wouldn't that be nice.

[Alan Carruth]

That would be characteristic of a resonance issue: it's linked to pitch rather than, say, buzzing on a particular fret.

Any musical instrument will have several different resonances. Often, particularly in the lowest pitch range, these will be most strongly affected by one or another par of the instrument. The lowest resonance is often a 'main air' resonance, where air is moving in and out of the sound hole at a particular pitch. This works more or less the same way as blowing across the mouth of a wine bottle, and the pitch is mostly determined by the size and location of the hole and the volume of the box. It's almost never 'pure' though; resonances of the wood, and even other 'air' resonances, can alter the pitch or the way it works.

This only gets to be more the rule as you go up in pitch. There are multiple 'air' and 'wood' resonances, and they interact (couple) with each other depending on a bunch of different factors. It makes it hard to attribute a particular peak in the output to any one part of the instrument with any confidence. It also makes it impossible to build the instrument to have, or avoid, a peak at any particular pitch: you can more or less control how many peaks it will have over a certain interval (say, per octave), but you can't say in advance what exact pitch any of them will be at.

I've built a few 'matched' pairs of guitars, the latest just a couple of years ago. I'd start with tops cut side by side from the same piece,and backs and sides the same way. All of the bracing was matched as well as possible, as were the necks and so on. The weights and thickness of the parts were carefully controlled, and the resonant pitches (tap tones) of the tops and backs were matched as well as possible before they were glued to the rims. In every pair people were easily able to tell them apart in 'blind' listening and playing tests. On the latest pair the spectrum analyses I ran on them showed them to be virtually identical in response below 1000 Hz, which is the range where you have some hope of keeping things under control in advance. above that they started to diverge more and more as the frequency went higher. This is undoubtedly the range that made them sound different. One problem there is that the wave lengths of the sounds get so short that it's hard to measure them with any assurance. It might be that the instruments produced sound at different pitches, but it could also be that the sound was going out so directionally that it was just missing the mic in one case where it was hitting it in the other. Not only are they likely to be doing different things, but you can't even be sure of what they're doing!

For two resonances to couple strongly they have to be fairly close in pitch. Just how close depends on what the frequency peaks they produce look like. Sounds that are not well defined in pitch and die out fast tend to produce broad frequency peaks, which can overlap with others over a wide range of pitches. Things like strings produce very narrow peaks, and won't necessarily couple strongly unless the pitches of he peaks line up closely. Working as carefully as you say you are you're probably making instruments with tightly defined resonant peaks that are very similar in pitch from one to another. A very small change, such as wood that's a little more or less dense than usual, can shift one into a range that will couple strongly with a string and cause a problem.

A light weight bridge is often a contributor to this sort of thing. Adding some mass there could well solve the issue without impacting the sound much.

[John Clifford]

Another thing you could try is attaching small lead weights with putty to the top, back or even the sides of the instrument in different places. Try it on the outside first, and if you find a spot that works, then you can move it to the inside. I wouldn't take the back off unless and until you know exactly what you need to do to correct the problem. Otherwise, how will you know if you've fixed it before you put the new back on?

[Gary Gallier]

Agreed, but if I don't get a definitive answer, I will have to take my chances. I think the odds of stiffening, or adding mass to the bridge area are with me considering the number of close to identical instruments I've built that did not have this problem...at least that's the optimist in me. I think a key will be not over doing it. It's really is a small amount of freq movement that I need. Hoping the brass bridge pins will be enough. Thanks.

[Clay Schaeffer]

Hi Gary,
I used to see your name on the Sweet Music Digest. Is it still around?
What John Clifford mentioned about adding weight to various places on the instrument to move a resonance is more fully explained in Arthur Benade's book " Fundamentals of Musical Acoustics" ($10 Amazon) He has good non technical explanations followed by technical information and also practical solutions to problems.
If you can find places on the top that the weight will kill the resonance, which you can also access on the inside (through the soundhole) you may be able to add weight without removing the back. Removing wood in some areas will have a similar effect. Benade suggested using weight to find the spot, and then thinning the area to change the resonance (gluing the weight to the inside also works).

[Gary Gallier]

Yes the Sweet Music digest is still around but has reduced participation due to various Facebook groups. Thanks for the book reference. The problem for placing bracing, or shaving bracing on the finished instrument is there is limited access to where the area might be through the sound hole due to its size. For example, if the effective adjustments need to occur from the bridge or behind, I am out of luck without removing the back. It's no big whoop to do that other than time. I do think I can emulate gluing the new back on with some ultra thin very strong two sided tape I have. Similar to what some sound board transducers are mounted with but comes in a long roll. Virtually like gluing, but the back can be removed with a box cutter. If I have to remove the back, that will hopefully give me the ability to judge the effects of internal adjustments before making it permanent.