Page 27 of 32

Re: An Interesting Dulcimer Experiment

PostPosted: Mon Jan 09, 2017 10:44 am
by Strum-Numb
Wow, that's something. That last image looks as if there are braces... so clearly your bracing patterns do match the natural flow of... let's say negative energy ditches or bridges. Modal pattern edges.

This thread has been ongoing far longer than I have been aware of this site/resource, but believe it or not, I have indeed read this entire thing, 8+ years worth. Some of it was less meaningful to me at the time, than it may be now. Perhaps it is time for me to revisit from the beginning.

I want to thank you sir for sharing your amazing experiments and findings with the community. Certainly you are the foremost researcher to ever live. I have to wonder... are you the most prolific builder on your continent? And is this your primary business? If you don't mind me asking. Maybe that is better addressed in PM.

Here is something that I have been tinkering with... the little carved slot around the perimeter of the top plate. Very traditional, many builders do this. The current most prolific perhaps, Warren A. May.

Pardon me if it has been covered, but have you any test results regarding that "accent line" groove around the tops?
Mr. May seems to use very little bracing, and I think none on the tops, other than beefing up around the sound holes if they are delicate shapes. But, he seems to also use pretty thick plates and sides. Perhaps because of that thickness, the grooves provide a lot more flexibility in the top.

- Tony

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Jan 11, 2017 7:57 am
by rtroughear
I've probably forgotten more of all of this than I've remembered, and of course some things were just wrong, or less right let's say, than I I'd claim now. This thread started with the proposition that the top plate hardly affected the sound of a dulcimer at all, other than keeping the air inside and allowing it to resonate. I'd moderate that now to something like "top plate material and physical parameters are not nearly as important in a mountain dulcimer as in a guitar or violin".

Australia has no prolific mountain dulcimer makers as far as I know. In the 1970'sfolk boom there were a number of makers who turned out dulcimers in the hundreds, but no longer. There's no Australian equivalent of McSpadden or Folkcraft. That's one of the reasons I started making them again after the first few 20 years earlier (in Vietnam). A co-worker expressed interest in one of my old dulcimers and I set out to find one for her. Hardly any music shop in Sydney had heard of them (in 1998), and none had one or knew where to get one. So I got some floor boards and made a couple myself. I continued making them for my own satsifaction, and followed the normal route of giving them to friends and relatives until I ran out of people and they got good enough to sell. But I'm retired, it's only a hobby and I only make about ten a year. I know of a couple of other makers who sell them, but I don't know what their throughput is. Guitar makers, of which Australia has a number of world class, might make a dulcimer for someone if asked, but a hand-made guitar at $2,000 upwards is much more attractive commercially than a $500 dulcimer.

I make no claims that my dulcimers are particularly good, or that these experiments provide some sort of formula for success. I've come to the conclusion that there is no "good" or "bad", only "different" as far as the instrument itself goes. And you can't take the person playing the dulcimer out of the equation. Much as we agonise over and analyse the sound of the instruments we make, the set up (string weight, tension, action etc) has a large effect on the tone, and a player's particular style adds most of the colour to the music rather than the tone of the instrument itself. But somewhere out there is the perfect dulcimer - incredible tonal qualities, superb playability, and looks amazing too. I think that's why amateurs like me make the next one, and the next one, and the next one... it's out there somewhere.

I haven't seen a dulcimer with the accent line around the top plate, but I know what you mean and have asked the same question. Did Uncle Ed Thomas know something that we've lost when he put that groove around the edge of the top plate? Ken Longfield tells me that the groove is so slight that it's just decorative, and I think he might be right - a token replacement for the inlay around the edge of a violin. But the idea that a groove along the edge of the top plate might increase the flexibility of the top clearly has some believers because Tayloer Guitars have a patent on the idea (US Patent 6,759,581). Taylor Guitars have that groove, so I'm told.
Taylor Guitars_Groove Patent.jpg

I've also cut a groove on the inside of the top on several of my dulcimers. The first one was such a good instrument, to my ear, that Ikept it as my own dulcimer.

I reported on this some way back in this thread. I did it on a couple of subsequent dulcimers, and also attempted to achieve the same result by thinning the edge of a dulcimer from the outside (also reported some way back). None of those subsequent attempts made for superior sounding dulcimers, or a discernable difference in the sound of the thinned edge dulcimer. As others have pointed out - a guitar top is a lot wider than a dulcimer top; there might be an effect in mountain dulcimers but it might be proportionately smaller than in a guitar, and not really audibible. I don't do it any more, but it still seems like an OK idea.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Jan 11, 2017 8:47 am
by KenH
That line around the top edge of a dulcimer is really nothing much more than a decorative nail scratch. That's how it's usually made -- a couple pieces of wood to act as spacers and a sharp brad or similar point to scratch around the top edge. We're talking probably less than half a millimeter deep. Uncle Ed was, IIRC, not the only olde tyme builder to use that scratch line as a decoration.

I think (but can't prove) that the scratch line was put on in imitation of the similar line found around the top edge of most fiddles -- purfling.

I find the Taylor patent of an internal "channel" (to distinguish it from a simple scratch) around the edge very interesting. Certainly a fiddly bit of construction. But I just can't see how that tiny groove could have any significance to the flexibility/vibration of the top or bottom plate.

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Jan 11, 2017 10:47 am
by Strum-Numb
"...make the next one, and the next one, and the next one... it's out there somewhere."

Or, rather... It's IN there somewhere. Just waiting to get out.

The journey is the adventure; more so than the destination.

Back in the 80's and 90's, in Florida, I built quite a few small boats, dories, skiffs, etc. Or bought interesting hulls and refurbed them into something really cool. Many of the craft that I built were furniture quality millwork and joinery. Lots of teak, tidewater cypress, mahogany, etc. The marinas and yacht owners that I worked with and for, called me "Tony Teak".

I would dream and design, and craft these boats... launch and run each of them... once, twice, maybe three times... and then sell it. It wasn't about having the coolest boat, it was about building the coolest boat. They sold very well though, I never advertised, just parked it out front and within a week or so, it was gone. And by then I was already into the next one.

So I get it man. I get that it's more about the creative process and crafting, than it is about keeping something or high production.

"...followed the normal route of giving them to friends and relatives until I ran out of people and they got good enough to sell..."

This is where I am at now. And I am retired/semi-retired. Or rather, we are, my partner and I. And she is very supportive of this dulcimer thing. And patient with my playing ability (or lack thereof). I just turned 54 a few days ago, and I think I have enough time left in life to start one more "thing". I think that "thing" is going to be dulcimers. I don't need to make a million bucks at it, or grow some big business or shop. I just want something creative and fun to do for a while, until I get too old and decrepit to do it anymore. One last project to add to my little legacy.

So that is why I keep asking all of these questions and digging around in the research.

The woodworking and using the tools/equipment... I got. I have a lot of shop tools already and I'm not afraid to spend a few more grand getting whatever I need or want. I already have my wishlist made up. Right now, we're looking at properties... I want a better location for doing this. We'll probably keep the farm, for its resources and standing lumber. But we're looking at homes with a detached shop, or space for a shop, in a more conducive area for a retail location than where we are now. Augusta KY and Morehead KY are prime suspects. I have an appointment on Saturday to look at a place in Augusta.

I hate giving up living out here in the country, as I look out the window to see a hawk flying by and two deer standing in my field about 80 feet away... but we hope to find something just outside of a little town that will be better for business. Maybe semi-rural as opposed to this... very rural.

So anyway... I'm not sure why I shared all of that... but I felt it only fair for some reason.

So now you know why I ask so many dumb questions and try to pick all of your brains for little morsels about building. I truly do want to produce the very best instruments at an accessible price point. I think as a "folk" instrument, these should be available to Folks... even the less fiscally able folks. That's sort of my goal, is to make very playable and good quality instruments at an affordable price for folks. I may do some higher-end models occasionally, but I think the staple stable will be in the $100 to $400 range. This is my concept... we'll see how well it actually flies.

- Tony

P.S. - To keep this post as an on-topic entry: I just shared with you all, my own, "interesting dulcimer experiment".
Wish me luck 8)

Re: An Interesting Dulcimer Experiment

PostPosted: Thu Jan 12, 2017 8:46 am
by rtroughear
Ken, I agree that the scratxch line is not really a groove - it's just a bit of decoration. To increase the flexibility of the top/side joint to give it some chance of influencing the sound, I think a groove would need to be at least about half the thickness of the top plate deep. I t shouldn't matter much where the groove is, on the inside, or the outside, but rather than a groove on the outside it would be easier just to sand the edges thinner for the same effective result. As you may recall I've tried both of these with no real benefit. But that's not to say it might not be wotrthwhile in some instances, and it's almost trivially easy to do. It took me about two minutes to put the internal groove in the top pictured above, using a round burr in a hand-held Dremel.

Tony, I would have thought that relying on mountain dulcimers for an income is almost a guarantee of poverty, but as a supplemental income they certainly serve the purpose of keeping me occupied, and paying their own way. And if they are well made they could still be around in a hundred years time, and still making music, which is a satisfying legacy.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Thu Jan 12, 2017 1:18 pm
by Strum-Numb
Faux purfling. Very traditional, but that's about it. Can be added as an afterthought, doesn't need to be in a master plan.

And yeah man, I have no illusion (or disillusion) of making a living, making dulcimers. Although it probably could be done, if done on an assembly line, factory shop setting... not really my cup of tea. I'm considering more of a scale that can be done at my leisure. When and if I feel like it, or have the time. I don't really need or want a full-time job. I already have a small farm to run; that keeps me pretty busy as it is, especially in the spring and summer growing seasons.

Thank you very much for your input, I truly appreciate all that you share.
- Tony

Re: An Interesting Dulcimer Experiment

PostPosted: Mon Jun 26, 2017 9:14 am
by rtroughear
Mountain Dulcimer Acoustics – Effect of a Free Top Plate Edge

I report this experiment for those who may be interested.

The Kantele is a traditional Finnish instrument, an unfretted zither, sometimes with an open back and sometimes an enclosed box, with a sound hole. Over the past decade studies have been undertaken, by Henri Penttinen and his students, into modifications to traditional construction that might lead to improved sound output and tone. This led them to a Kantele that was an enclosed box type, but instead of a sound hole, there was a narrow gap between the edge of the top plate and the sides. They claim some sound enhancements result from this.

An interesting masters thesis of one of Penttinen’s students, Henna Tahvanainen, is available on-line at This has a lot of information regarding the acoustics of stringed instruments, much of which is applicable to mountain dulcimers.

In the past I have done vibration mode tests on completed mountain dulcimer tops in the hope it would allow “tuning” of a top before it was glued to the sides – as some makers do for guitars and violins. No such luck. However I did notice that the vibration modes of free tops were, in general, more complex than the vibration modes of the tops of completed dulcimers.
Fixed Edge_Free Edge_Vibration Modes.jpg
This led me to wonder if a more “interesting” sound might result if the top edge of a dulcimer was not glued to the sides but was separated by a small gap. So I just cut the top plate of my test dulcimer around its periphery, allowing the top plate edges to vibrate freely.
Test Dulc_Free Top Edge_Half No Top.jpg
The resulting saw cut was only about 1mm wide and adds about 1500mm2 (about 2.3 to the sound hole area – about 50% more than was there already. Not an excessive increase (total about 6, up from 4 but it would raise the frequency of the first air resonance by about a semitone. The reason the saw cut is about 1cm away from the dulcimer edge is that the internal side linings are that wide (from a previous experiment).

Fairly under-whelming.

I recorded three tunes with the intact top, then cut a slot in one side, made the same recordings, then cut the other side and made more recordings. My overall impression was that there was no dramatic change in tone for fixed top edge (normal situation), half free/half fixed, or fully free-edge top. In blind listening tests (repeated twice) I preferred either the fixed (normal) top or the half free/half fixed-edge top. I consistently didn’t prefer the fully free edge top. But there was not a lot in any of it – nothing obviously changed for the better or for the worse.

This was clearly not an exhaustive experiment, just a rough attempt to see if there might be a large effect. The Finnish group found tonal differences with different air gaps, and even an optimal air gap for maximum radiation efficiency (loudness). So some variation of this idea might still be profitable for mountain dulcimers, but it won’t be a simple as just leaving the top edge free – some tuning would probably be involved. And it won’t be a game changer – improvements would be subtle, as with most other constructional factors.

When I’d done with the free edge, I thought I might as well cut half the rest of the top off, eliminating most of the internal air resonance sound contribution. As expected that didn’t improve the sound either (but it did seem just as loud)

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Thu Jul 06, 2017 8:30 am
by rtroughear
Mountain Dulcimer Loudness - Six Strings vs Four Strings

I was recently asked if a six string mountain dulcimer would be louder than a four string one, and I automatically replied that it should be. But on reflection it’s not that clear cut.

If we assume that each string contributes equal loudness to the sound if struck equally, then adding two extra strings to a four string dulcimer should increase loudness by 50% That translates to 3.52dB increase in sound pressure level (SPL). (20Log (1.5)) (I’m not adhering to proper convention here – loudness is not the same as sound pressure; one is a perceived level of sound strength, and the other is a measured quantity – the output of a microphone).

It’s generally accepted that most people will perceive a 10dB increase in SPL as a doubling in loudness, but a just noticeable change in loudness is harder to pin down. All sorts of things influence it, but somewhere between an SPL increase of 1dB and 3dB most people will start to hear a sound as louder than it was.

This means that the measured SPL of a six string dulcimer needs to be at least 1dB greater than the same dulcimer with four strings, under the same conditions, if it is to be perceived as a louder instrument.

The Experiment

Five six-string dulcimers were within arms reach so I measured the sound pressure level of each of them with six strings, and then four strings, and compared the results. It was expected that the six string recordings would produce higher SPL readings, but would they approach the simple expectation of 50% higher (3.52dB), or if not, would any SPL increase be likely to be perceived as louder?

Each six-string dulcimer was mounted on a stand and six open-string strikes were made of all three strings by a plectrum mounted on a rigid pendulum. Recordings were made.
Six String_Four String_Pendulum Striker.jpg
One bass string and one middle string was carefully removed without disturbing the recording set-up and six more string strikes were recorded.

The first four seconds (to 10ms accuracy) of each strike was analysed with the PRAAT software package for average sound pressure level and the results for 4 and 6 strings averaged for each dulcimer and the two levels compared.


The data is summarised in the table below:
Six Strings vs Four Strings Loudness.jpg
The first dulcimer (#20) actually measured as louder when set up as a four string. However, I think I reversed the direction of the pendulum swing between the 6 and 4 string arrangements – i.e. the plectrum struck the treble string first in one set of recordings, and the bass string first in the other. That alone is interesting, implying that direction of strumming may have an effect on loudness. When I repeated the same dulcimer at the end 6 strings were recorded as higher SPL than 4 strings.

Only one of the dulcimers approached or exceeded the 3.52dB necessary for the simple expected 50% increase, That dulcimer was the Ebony instrument. The remainder had a six string SPL increase over four of about 1dB or less.


Most of the dulcimers tested as six string instruments were only about 1dB higher in SPL than when set up with four strings. This is just about the point where most people will start to perceive an increase in loudness. So it seems that, in general, six string dulcimers are not likely to offer a noticeable loudness increase over a four string dulcimer, and certainly not a substantial one.

But as with all “rules”, there are clearly exceptions. The Ebony dulcimer was 4.4dB higher SPL when strung as a six string – well within the range where most people would hear it as louder than four strings. In this case, it is quite a heavy instrument and the bass strings seem to carry more of the total power, relative to the treble strings, than in the lighter instruments tested, so taking off a bass string removed more of the original sound power, and made it relatively quieter as a four string.

The two #20 tests demonstrated that the same instrument can produce opposite SPL changes if the strings are stuck differently (but with the same force). (However, neither SPL change would likely be heard as a loudness change).

Why didn't the six string set up generally approach the notionally expected 50% increase in SPL readings? Can't say. Perhaps close string pairs interfere with each other and lose energy doing it; perhaps there are more complex couplings at the bridge and nut for double strings; perhaps a plectrum doesn't strike each string equally when they are close together. It doesn't really matter much in a practical sense.

So why make six string dulcimers at all, or have a double first string for that matter? It must have more to do with appealing changes in the timbre of the sound than with producing a louder instrument … with some possible exceptions.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Thu Jul 06, 2017 8:57 am
by KenH
Great experiment, Richard! I've always said that doubled courses gave a richer sound, but not a louder sound; now you've more or less proved the "not necessarily louder" part.

Re: An Interesting Dulcimer Experiment

PostPosted: Fri Feb 09, 2018 6:48 am
by rtroughear
Mountain Dulcimer Acoustics - The Sequence of the First Four Resonances

This note is mainly for the record – not of immediate practical use.

There are twenty or so resonances below about 2000Hz in a standard sized mountain dulcimer. These are a combination of internal air cavity resonances, vibratory resonances of the wood plates of the instrument, and the interactions between the two. eg. the variations in internal air pressure caused by an air resonance will cause the wood itself to vibrate, which in turn will modify the internal air pressure.

The first few resonances probably influence the general tone and loudness of the dulcimer more than all the others, and may also be under the control of the maker to some extent, so their origin may be of interest to some.

Using a combination of sound spectral analysis, loud-speaker driven dulcimer vibration mode analysis (Chladni patterns) and sound hole size variation, I think I have a handle on the first four dulcimer resonances.

Tapping the bridge of a dulcimer and obtaining a frequency spectrum of the sound thus made will indicate what resonances are present in the instrument, and their frequencies.

Chladni modal analysis shows where the actual vibration is occurring on the instrument, in patterns of sawdust. It can also reveal whether an internal air resonance is involved – just holding a finger above the sound holes shows whether air is flowing in and out or not. Upper and lower sound holes may individually have air flow, or both might. If there is no observable airflow from the holes, then the resonance probably originates in wood vibration only.

Blocking off the sound holes, and observing what happens in the tap spectra can reveal the frequencies of the Helmholz resonance(s) - strictly the first air resonance because of the box flexibility. It seems that mountain dulcimers, of the traditional full fretboard/four hole type, might have two Helmholz resonances – the tone obtained when blowing across the top of a bottle. I don’t have enough acoustics theory to explain why this might be.

So what are the lower resonances of a mountain dulcimer?

The lowest resonance is the 1st Air Resonance (“Helmholz”) and it falls in frequency between about 150Hz and 250Hz for a standard sized dulcimer – a fairly wide range dependent on box size and stiffness, and sound hole size. Smaller holes will lower it, larger holes will raise it. A larger box will lower it as will a more flexible top, back and sides. It is unlikely that it will get low enough in frequency to strongly support the fundamental of the low string. Blowing across the lower sound holes will produce it (If they are not of complex shape. If the holes are complex it may not be possible to produce it by blowing, but it will still be there.).

The next resonance is usually the 1st Bar Resonance which depends on the length cross section and mass of the dulcimer. The first (and only?) bar resonance might fall in the range of about 220Hz to 350Hz.. The headstock and the weight of the machine tuners can modify this resonance by a couple of semi-tones. Heavier end weight and a longer dulcimer will lower the frequency of this resonance.

The third resonance is the 2nd Air Resonance and is the tone produced by blowing across the upper sound holes. It seems to act like a Helmholz resonance in that if the sound holes are blocked off, both the first and second air resonances are extinguished – they don’t appear in the tap spectra.

The fourth dulcimer resonance is the 1st Wood Resonance – the lower bout of both the top and the back vibrate strongly with a simple circular mode in the region of 400Hz and are probably in phase with each other because there is no airflow from the sound holes. The sides may be going up and down out of phase with the top/back.

Here’s a couple of dulcimers to demonstrate what happens when the sound holes ore systematically blocked. This sequence, that has been repeated on seven dulcimers, seems to show that:

With all holes open, the resonance sequence is 1st Air; 1st Bar; 2nd Air; 1st Wood.

With upper holes blocked, the 1st Air resonances falls by up to 5 or 6 semitones and the 2nd Air by 2 or 3 semitones. The 1st bar and 1st wood don’t change much.

With lower holes blocked the 1st Air resonance (“Helmholz”) disappears and the 2nd Air resonance remains basically unchanged. 1st Bar and Wood remain unchanged.

With all holes blocked the two air resonances disappear leaving the Bar and Wood resonances unchanged.

Changing the size of the sound holes has merged the 2nd Air resonance with the 1st Bar resonance in the dulcimer above. Sometimes it is difficult to know where the resonances are if they are stacked one on top of the other as in the dulcimer below – the 1st Air, 1st Bar and 2nd Air all fall in the same frequency region. This is not generally a desirable situation – it is thought better to have a good spread of resonances for good tone and to prevent boomy areas of the spectrum.

Not all dulcimers will have this sequence of resonances – some may be missing, or occur in a different order, but this generally seems to reflect the low frequency resonant behaviour of a standard mountain dulcimer.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Fri Feb 09, 2018 4:35 pm
by Jon
Richard, I just happened to read your 4 vs 6 string post. In my experience, they are two different instruments in that in addition to the number of strings, there are several construction differences, at least in the ones I make. My 6 string dulcimers have a wider (9"") and deeper body(3"), as well as a wider fretboard, than my 4 string instruments. They are also tuned as baritone instruments. I've tried octave tuning on the middle and bass pairs, but didn't like the sound. The 6 string instrument sound louder to me (I've not measured them), but the number of strings may have little to do with the loudness.

Re: An Interesting Dulcimer Experiment

PostPosted: Sat Feb 10, 2018 10:25 am
by rtroughear
I don't make any distinction in the construction of 4-string, 6-string or baritone (6 or 4 string) dulcimers. Any string tension differences between them is well within the stuctural strength of even quite lightly made dulcimers. I strongly suspect that if you made two similar dulcimers, one a 4-string and the other a 6-string, you would hear a clear tonal difference, but the loudness of the two (the subjective exerience of sound strength) would be barely noticable, if at all, and the actual sound pressure produced would be within a deciBel or so. If you hear loudness differences between one of your 4-strings, and a 6-string, I think it's much more likely to be because of the construction differences than the number of strings.

Also it's not a fair comparison between a standard 4-string and a baritone 6-string - the harmonic series of the two are quite differently spread. I haven't done any tests of loudness differences between a standard 4-string and the same dulcimer strung as a 4-string baritone, but there may well be a loudness difference there, I can't say at the moment. The test is if you take two strings off one of your 6-string baritones and compare it with the loudness as a 6-string (by measuring the SPL under the same conditions).

For me, this is one of the small number of experiments that seems to have a clear-cut outcome - for the same dulcimer six strings are not generally louder than four strings, so it's not the number of strings that determines the loudness.

I'd ask the question - why do you construct 6-string dulcimers differently to 4-strings? or baritones for that matter. String tensions for heavier strings are the same as for lighter strings when tuned to the correct note (give or take). And the moderately larger box size you use for a 6-string/baritone (A-E-A) will still be nowhere near large enough for the lowest resonances to support the fundamentals of the bass and middle strings, so you don't gain much acoustically by making the box a bit bigger. Maybe it's just what the customers expect?

Richard T