Page 26 of 32

Re: An Interesting Dulcimer Experiment

PostPosted: Sun Sep 25, 2016 2:10 am
by harpmaker
Making the top as light as possible has always worked out well for me, and I agree that while the thickness of the back and sides does have an effect on the sound, it is not anywhere near as important as on the top.

In regards to the slot under the fret board, I glue my fret boards to the top and then slot it.

I usually do not install any cross braces on the top.

Re: An Interesting Dulcimer Experiment

PostPosted: Tue Sep 27, 2016 8:17 am
by Robin the Busker
I tend to need a lot more acoustic volume than other players so that's high on my requirements list for a dulcimer. What I don't want though is a lot of bass, so contemporary designs like the Probus wouldn't work for me. The Glen dulcimers work for me as does the Orthey or Heritage model, and a number of other trad designs.

I've just spent a week on Lundy Island and took a Heritage model. I played a number of sessions with a DADGAD guitarist, and we punched out a lot of fiddle tunes. I reckon that DAd chord melody playing would have got lost against fast open tuned rhythm guitar. I played noter drone with 022w, 013, 013 strings up at EBB tuning (the guitar was capod fret 2). To get more punch I used the bone from a seagull as a strummer. The instruments balanced perfectly and we certainly had the Marisco Tavern jumping! Although my dulcimer itself has good volume, it was the set up, tuning and playing style that allowed me to match the guitar.

BTW - noter drone dulcimer and DADGAD guitar is a phenomenal mix !!!!!


Re: An Interesting Dulcimer Experiment

PostPosted: Tue Sep 27, 2016 11:10 am
by rtroughear

I do recall Lois Hornbostel once saying that when playing acoustically with other instruments the principal requirment of a mountain dulcimer is LOUDNESS. Other tonal subtlties have to be subbordinated to that. So you have a point as a player that non-performing makers such as myself sometimes don't appreciate.

And the comment about not too much bass is also interesting. Many players these days say they want a "warm", "mellow"; "resonant" sound in a dulcimer, and the bass string probably contibutes most of that component of the sound. In earlier posts in this thread I've drawn attention to the mis-match between the lowest note the strings are tuned to, and the lowest note the box is capable of producing - the bass string asking the box to produce a note it's not capable of doing. But like everything to do with musical instruments, it's not that simple. The box does reproduce the bass note, it just gets less and less efficient as the string note gets lower, and the lowest box resonance (the Helmholz!) gets higher. So that situation might well suit your style of playing, Robin, and noter-drone in general, especially with small bodied dulcimers. But many clearly want improved bass contribution to the sound (or say they do), so there's a move towards larger bodies with a view to improving the bass sound.

I've been wondering if this is a valid argument or not i.e. bigger box = better bass It hasn't been my experience that this is so. And when you look at the frequencies involved, the size of the box to efficiently reproduce the fundamental harmonic of the bass note would be so big as to be impractically large. Which is why I built myself a Tennessee Music Box - to see if it was big enough to reproduce the bass fundamental better than a standard sized instrument. I'll write this up in due course.

But a question arises. Just how important is the fundamental harmonic of the bass string to the sound - does it really matter if it can't be reproduced efficiently. Here's two sound clips of single note plucks, edited together. Firstly, bass string with dulcimer played on the bench, then same string with dulcimer on the knee.

D 54_01.mp3
[ 263.31 KiB | Viewed 3847 times ]
The bench vibrations, driven by the dulcimer, are probably contributing a lot to the strong fundamental compared to the knee playing.
The second clip is the same dulcimer, bench mounted, with the bass fundamental intact, then with it filtered out.

D 54_02.mp3
[ 324.53 KiB | Viewed 3847 times ]

So if you want warm and mellow, it seems the presence of the fundamental harmonic of the bass string might matter.
(Note: You'll need more than 1" computer speakers to hear any difference.)

How to get a strong bass string fundamental harmonic? Unfortunately I haven't got a clue.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Tue Sep 27, 2016 6:35 pm
by Robin the Busker
Hi Richard,

I have to say that many of my customers ask for a dulcimer with a mellow tone; I don't think I've ever been asked for a dulcimer with a bright tone. Although a few more recently have asked for something with a 'traditional' tone. In some respects building for a noter drone player is simple: just copy an old design! Building for contemporary playing styles is far more complex as you have demonstrated throughout this thread.

Set up and playing style has a BIG impact on timbre. The little Mawhee dulcimer has a massive voice simply because it is set up with 3 x 0.020 plain wire strings wound up to Gdd that have the hell thrashed out of them with a turkey quill. The McSpadden that I staple fretted by ear and set up (high action and heavy strings) for noter drone playing is the loudest McSpadden I've played.

I suppose what I'm saying is that most old dulcimers when set up and played in trad styles are perfectly good folk instruments for playing Appalachian old time tunes and songs either solo or with other old time instruments at sessions - the instrument sits in a unique space in the sound mix and punches above its weight. I have said many times that I don't play in noter drone style because I'm a traditionalist but because I find it the pragmatic way to play the dulcimer in the settings I frequent.

Obviously, it is no surprise that if you build, set up and play the dulcimer as it was traditionally built, set up and played then it works just fine :D

Re: An Interesting Dulcimer Experiment

PostPosted: Mon Dec 05, 2016 12:08 pm
by Bob Gerard
That's just amazing!

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Dec 28, 2016 11:13 am
by rtroughear
Mountain Dulcimers - The Bass String Fundamental Harmonic

Following on from the previous posting, I made myself two Tennessee Music Boxes.
to see if a larger bodied instrument might produce the bass fundamental better than a normal sized dulcimer. These were made from Radiata pine (same density as Poplar) bought in the local hardware store, with dimensions obtained from David Schnauffer’s description of TMBs on a web page. They only took one day each to make.

Dimensions are:
TMB 01 - 31*9.8*71.5cm (12*4*28 inches); plate thickness 8mm/0.3”; internal air volume of 16,561cc / 1010cu in; lowest air resonance 80Hz

TMB 02 - 23*9.5*71.5cm (9*3.7*28 inches); plate thickness 3mm/0.12”; internal air volume of 13,954 cc/ 851 cu in (84% of TMB01 capacity); lowest air resonance 70Hz.

(Compare this with a typical mountain dulcimer - air volume of about 4000cc/250 and a lowest air resonance of about 200Hz)

For these two TMB’s one had thick plates typical of old TMB’s, and one had thin plates typical of mountain dulcimers. The larger, thick plate TMB01 had a higher air resonance than the smaller TMB02 because TMB02 had thinner plates and was therefore more flexible, which in this case more than counteracted the larger capacity of TMB01.
(Increased size – air resonance goes down Increased flexibility - air resonance also goes down)

Both TMB 1st air resonances were lower than the lowest tuned note - C3 (131Hz) which might indicate that the fundamental harmonic could be produced well.
TMB_1 and 2.jpg

But why bother with this at all? I’m sure there has been a lot of research on the importance, or not, of the fundamental to the perceived tone of a musical note. And it’s probably quite complex and theoretical in terms of the psycho-acoustics of it all, but I wanted to see in a simple practical sense whether the strength of the fundamental harmonic matters much to the sound of a mountain dulcimer, and particularly to the perception of the “mellowness” of the instrument.

To that end I :-
-Recorded single note strikes for each open string of eleven dulcimers (and an epinette), both with the instrument mounted on a wooden bench, and with the instrument on the knee. Two dulcimers were baritones (AEA) and one was a bass (DAD). The rest were tuned CGc. Conditions were standardised.

-Took the sound spectrogram for each pair (bench/knee) of notes for each string of all dulcimers (11 dulcimers; 3 note pairs/dulcimer; 66 sound spectrograms).

-Noted the loudness and duration of the fundamental harmonic of the 66 spectrograms.

This is the end result for two of the instruments – TMB01 and D17, a smallish, but typical mountain dulcimer:
Spectrograms_TMB 01 and D17.jpg

Left panel is TMB01, right panel is D17. Horizontal scale is time (0 to 10sec), vertical scale is frequency (0 to 4000Hz). The “flags” represent the strength (loudness) and duration of the harmonics of the notes. In each of the six-sub panels the lowest harmonic is the note fundamental.

Each note pair (bench/knee) for each string for each instruments was rated for sound preference – which of the two did I like better?

These preferences were then correlated with the strength/duration of the fundamental harmonic of the spectrogram for that note/instrument.


For the 33 note pairs:
- 18 were equally preferred i.e. both bench and knee note sounded equally good to me.
- The bench note sounded better in 8 cases, and
- The knee note sounded better in 7 cases.

In ALL cases where there was a preferred note (15 cases) the fundamental harmonic was stronger and/or longer than the non-preferred note.

In some of the 66 notes analysed the fundamental was completely absent.(6/66); in most cases it was weak and/or of short duration.(33/66). In some cases it was long, but weak (4/66). In only 5/66 cases was the fundamental both strong (loud) and of long duration.

It appears that the presence of the fundamental harmonic in a note improves its tone – at least to this listener. And in particular it seemed that a noted was more “mellow” or “full” or “rounded” when the fundamental was present.

So the answer to the question “Does the fundamental harmonic of a note matter in a mountain dulcimer” seems to be a fairly clear “Yes, it does”.

The next question is how might we increase the strength of the fundamental? I can’t say for sure. However a clue might be in the construction of the instruments that demonstrated the strongest/longest fundamentals in their notes.

The Orthey replica dulcimer had fairly strong fundamentals on five of the six notes recorded, but of only medium duration (about 5 sec). It is a very lightly constructed instrument with an overall tone that I like a lot.

TMB02 had long fundamentals on five of six notes (about 10 sec), but of medium loudness. I would characterise this instrument as having a “smooth” sound (whatever that means). It is much heavier than the Orthey, and about twice as heavy as a standard dulcimer, but the plates are also larger and may be relatively as flexible as the Orthey.

The other dulcimers, including the stiff TMB01, were more variable in the strength/duration of the fundamental harmonics.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Dec 28, 2016 11:15 am
by rtroughear
Mountain Dulcimers – The Contribution of Box Volume to Mellow Tone

The previous post is related to the frequent claim that a bigger dulcimer body leads to more mellow sounding mountain dulcimer. This makes sense in that a larger capacity body should have a lower Helmholz air resonance (more strictly, 1st Air resonance which is lower than the Helmholz because of box flexibility).

But it hasn’t been my experience that this is so. My dulcimers almost all have the same plan outline, but I don’t have a standard side height, so there is a wide range of box volumes, determined by the height of the sides. At one extreme the sides were 62mm/2.4” high on two dulcimers that were noted at the time as having a “sharp” sound, and on the other extreme, two dulcimers at 40mm/1.6”, and 38mm/1.5” that were noted as “mellow”. The opposite to general expectation. So the contribution of the box volume to mellowness-perception must be able to be counteracted by some other factor.

I have also noticed that many of the more mellow dulcimers I have seen have wider bodies, particularly in the lower bout. In addition to that, one of the factors that Dave Harpmaker Lynch proposes as influencing dulcimer tone is the distance of the bridge saddle from the side of the instrument. This is equivalent to saying that the lower bout is wider at the level of the bridge.

To gain some modest idea whether box volume is a major contributor to dulcimer tone I’m making three dulcimers at the moment, each with the same internal air volume, but with different plan outlines – the height of the sides varying to keep the capacities the same.

What might this show? My suggestion is that the width of the lower bout may be more important in determining tone than is the air volume of the instrument. If this is the case then the widest of these three should be the most mellow, and the narrowest, the least mellow. And it would point to the possibility that when makers have made progressively larger bodied dulcimers, and have attributed increased mellowness to the box volume, it may have been the increase in top width that was the cause instead.

We’ll see in due course.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Wed Dec 28, 2016 12:53 pm
by Strum-Numb
You sir... are freakin' nuts.
But, WOW!


I can hardly wait to see the results of your newest exploration.
All of us who enjoy whittling and creating saw dust owe you a huge debt of gratitude.
You should be nominated for some kind of research award or medal.
Maybe someday when there is a Dulcimer Hall of Fame, they'll name a building after you.

Re: An Interesting Dulcimer Experiment

PostPosted: Sun Jan 08, 2017 4:41 am
by rtroughear
Mountain Dulcimers – The Contribution of Box Volume to Mellow Tone –(Continued)
The three dulcimers referred to above are finished and are all very nice instruments in my estimation. As usual the outcome of the experiment was not quite as expected.

Body and top woods were from the same billets and were as similar as is reasonable. The bodies and fretboards are New Guinea Rosewood (pterocarpus indicus, a Padauk) and the tops are Kauri Pine. I spent considerable time matching the mass and stiffness of the three fretboards to within a couple of percent of each other.

The side heights were set so that the three had the same internal air volume of 5525cm3/ 337 . This is a little larger than my usual air volume of about 4500cm3/; so if the orthodoxy is correct they all might be a little more mellow than normal for my dulcimers. The side heights were 60mm for the standard outline; 50.5mm for the wide outline and 44mm for the Extra Wide outline.

Soundholes were the same for all three so the lowest air resonance (the Helmholz, set by internal capacity and sound hole size)) might be expected to be the same for all three, however the boxes of the thee are different in flexibility because of the physical size differences and this modifies the theoretical Helmholz frequency. I expected the lowest air resonances to be different between the dulcimers and they were: 211Hz for standard, 186Hz for wide, and 172Hz for extra wide – a 3 semitone spread.

The dulcimers have no side linings, however there are top and back braces, the same for all three, but clearly he shorter bracing of the standard outline might make the cross grain stiffness of the top and back higher than the longer braces of the two wider instruments.

I didn’t attempt to match the cross-grain stiffness of the plates because I don’t think it matters nearly as much as, for instance, the treatment of the fretboard, although perhaps the back bracing could have been a little lighter.

Re: An Interesting Dulcimer Experiment

PostPosted: Sun Jan 08, 2017 4:49 am
by rtroughear
[u]Mountain Dulcimers – The Contribution of Box Volume to Mellow Tone –(Continued)
Measurements etc.[/u]
II made the usual measurements that I do – spectral recordings of the tap tones of the completed top and the completed instruments, and also recorded under standard conditions, notes from each string of each instrument for subsequent display as sound spectrograms.

The main purpose of all this was to see if a larger top area, particularly in the lower bout, translated into a more mellow sounding dulcimer, and especially to see if the fundamental harmonic of the bass string correlated with the perception of mellowness in these three dulcimers.

I also recorded test tunes, and played live and listened critically to the three dulcimers.


The frequency spectra of the bridge tap tones were different as expected (because of the different flexibilities of the bodies)
BridgeTap_104and105and106_No Hardware_a.jpg

However, the lowest resonance of all three is still well above the lowest tuned note (C3, 131Hz) so none of them seem to offer support to the fundamental harmonic of the bass string. But in the general trend of frequency regions of comparable strength, the three are broadly similar.

The sound spectrograms for individual string plucks looks like:

The horizontal scale for each of the nine panels is 0 to 10seconds and the vertical scale is 0 to 4000Hz. The lowest of the horizontal flags in each panel is the fundamental harmonic for that string. The wider and blacker it is, the longer and louder is the sound.

The bridge tap spectra of indicates how the dulcimer box “likes” to vibrate given its physical structure. The sound spectrograms indicate how it is actually vibrating under the forcing influence of the strings.

In this case the spectrograms don’t show a lot of difference between the three dulcimers. Particularly the fundamental and the first few higher harmonics of the bass strings are very similar in strength and duration for the three. Same goes for the trebl string – fairly similar. The middle strings seem to show the wide profile instrument to be a little stronger in the lower harmonics.

And basically that was how I heard the three dulcimers – there were no clear, unequivocal differences between the three. There are tonal differences, but to the casual listener they all sound pretty much the same. Playing them live, and playing the recorded tunes, the standard outline instrument might be judged slightly less mellow than the other two, but there’s not a lot in it. For playing on individual strings the standard outline dulcimer sounds the most mellow of the three on the middle string, while the extra wide outline sounds the least mellow. The bass string tone is almost identical for the three.


The end result is that these three dulcimers, quite different in the size and shape of the top plate, are very similar sounding instruments. And more than that, the sound is not noticeably more (or less) mellow than my dulcimers generally are.

So the proposition that a larger top plate should lead to a more mellow sounding mountain dulcimer does not appear to be true.

Then what has happened here? What are the factors that are common to these three instruments, and what are the difference?

What is the same for all three:
- The internal air capacity is the same
- The stiffness and mass of the fretboards is the same
- Finished weight is close (3.5% spread)
- Plate thicknesses are the same
- Sound hole size is the same

What is different between the three:
- side height is different (35% spread)
- lower box/air resonances are different (3 semitone spread)
- surface area of the top/back is different (35% spread)
- major/minor bout width (40% spread)

There are probably lots of other variables that affect the tone, but the ones listed are largely under the control of the maker.

If we say that these three dulcimers sound very similar (and you’ll have to take my word for it), then increasing the area, width or shape of the top plate is not a sure-fire pathway to a mellower instrument. The fact that these three are about 25% larger in air capacity than my average instruments, and yet are not appreciably mellower, might imply that a larger body is also not a guarantee of mellow tone.

So what does produce the goods might still be one or more of the factors listed in an earlier post:
Construction Factor -------------------------------Bright Dulcimer-------------------------------Mellow Dulcimer
Top Thickness.............................................Thicker; >3mmm...........................Thinner; < 2.7mm
Side Heigher...............................................Higher; >50mm.............................Lower; < 45mm
Top/Fretboard Stiffness...........................Stiffer; < 100/1000” deflection...............Flexible; >130/1000” deflection
Stiffness of Back Braces.................................More stiff....................................Less stiff
Height of fretboard at strum hollow...................Higher; >10mm..............................Lower; < 7mm
Bridge position from internal edge of end block......< 50mm......................................> 80mm
Total Area of Sound Holes...........................Larger; > 30cm2 (4.5 sq in)..................Smaller; < 10cm2 (1.5 sq in)

How to square this with the experience of makers over the years who have produced larger bodied dulcimers and noted tonal changes, I don’t know. I’d be interested in comment by makers, and any counter arguments.

Finally, this experiment, and the results, are for full length fretboard dulcimers – it may not be relevant to truncated fretboard instruments.

Richard T

Re: An Interesting Dulcimer Experiment

PostPosted: Sun Jan 08, 2017 11:02 am
by Strum-Numb
These seem to have a lot of bracing. Probably more than most builders will use.
On the very wide example, this amount of bracing may be necessary. But on narrower, or more "standard" width instruments, this looks like more than most builders would incorporate into a design.
I like that you tried to keep the bracing uniform across the examples, but I'm not sure it is proportionally equal in regards to stiffness and mass. The narrow example would enjoy less deflection and be burdened to move proportionally more mass (in bracing material) in doing so, relative to the wider example.

Perhaps this is irrelevant to this experiment, or so minor that it shouldn't have a discernible effect.
I'm not sure how it could be accurately calculated for a perfect ratio, but it seems like the bracing for the narrow example could have less mass. In the profile or cross section view of a brace, it would be narrower and not as tall. Somehow proportionate to the ratio of mass and stiffness compared to the wider example.

I think that this consideration could make a difference in your results. But probably not enough difference to matter.

I feel that your examples are fair enough comparison for the purpose of this experiment. But, for discussion's sake, I submit this post.

As a side note: Very nice instruments; those all look great. Very nice job.

Re: An Interesting Dulcimer Experiment

PostPosted: Mon Jan 09, 2017 8:51 am
by rtroughear
The bracing is typical for my dulcimers. Sometimes I leave off the brace nearest the headstock, it probably doesn't contribute much to overall structural integrity being so close to the solid end block. The braces are in those positions because long ago I calculated that's where nodes of some of the internal air resonances should be and it seemed like a good idea at the time to have stiffer bits where the wood was less likely to be asked to bend. Probably complete rubbish, but I still do it anyway. In sticking to the same pattern I used last week and the week before, I claim solidarity with most of the old time builders who found a successful pattern and stuck to it.

The bracing looks a bit bigger than it actually is because of the sun angle, but your comments about relative stiffness between the three tops are perfectly true. However, notice I said that I didn't attempt to make the cross-grain stiffness the same for the three tops because results of other experiments indicated that top bracing doesn't seem to have much affect on the tone of the instrument (for full-length fretboard dulcimers). This experiment further supports that idea because even though the three have relatively different cross-grain top stiffness, they still all have a very similar tone, loudness, sustain and dynamic range. I think this is primarily because I closely matched the mass and stiffness of the three fretboards, so the long-grain stiffness of the three tops is the same. In another experiment I've shown, at least to my satisfaction, that ladder bracing like this doesn't much change the long grain stiffness of the top.

It all keeps coming back to the idea that the nature of the fretboard is the prime determinant of what a dulcimer sounds like. In a typical dulcimer the hollow fretboard is about 150 times stiffer than the top plate itself. So when it comes to bending the wood to move some air around, it's the fretboard that has to be overcome more than the top plate itself. Most of the lower vibration modes of the top seem to involve the fretboard, as well as involving the top plate. Here's a typical vibration mode:

The middle part of the lower bout top/frteboard goes up and down while the outer edges go in the opposite direction. The top plate does vibrate in local areas that don't include the fretboard, and where you might think that bracing could modify the vibration, but those patterns are at frequencies above 1000Hz, and are mostly small enough to fit well within any bracing pattern. Those higher vibration modes are usually too complex to analyse, let alone understand, but possibly add a finishing colour to the sound. Looking at my files just now I can't actually find any example of a top mode that doesn't also include the fretboard, but that may be because I just haven't recorded any, being too complex to make any sense of.

For the back plate however, I'm coming around to idea that bracing might noticably affect the sound. The back vibration modes do sometimes follow the brace lines, but mostly the modal patterns seem to ignore the bracing. Sometimes they appear to follow the back braces, but coincidentally. Here's a picture of two dulcimer backs - the top panel has one of its modal patterns overlayed on the bracing pattern, and the mode matches the bracing closely. But the bottom pictue is essentially the same vibration pattern in another dulcimer, and that dulcimer has no back braces. So it's all a bit of a mystery.
Braces_No Braces_Vibration Mode.jpg

Richard T