An Interesting Dulcimer Experiment

Just share stories or offer advice

Re: An Interesting Dulcimer Experiment

Postby rtroughear » Mon Jul 27, 2015 4:15 am

Longitudinal Waves in Mountain Dulcimer Strings (continued from previous posts)
Here is a picture of a string being plucked whilst being tuned. The harmonics enhanced by the Longitudinal Wave can be seen as the string rises in pitch and the harmonic matches the frequency of the Longitudinal Wave.
D85_L-Wave vs String Pitch 189Hz to 231Hz_LoRes.jpg
The blue area highlights the harmonic region affected.
This is what this series of notes sounds like:

D85_L_Wave_vs_String_Pitch_189Hz_to_231Hz.mp3
[ 693.19 KiB | Viewed 4284 times ]


If you can hear this tonal addition, and if it annoys you, then what can be done about it? The short answer is – not much.

The longitudinal wave is inherent in the string and for the string materials and lengths we use, its frequency is right in the middle of the harmonic spectrum the strings produce. So the potential to excite, and hear it is always there. But how can this excitation be prevented?

I tried a number of things:

1. Changed strings – different thicknesses, different metals
2. Damped the strings behind the bridge and in headstock
3. Changed bridge saddle shape – square, triangle, broad, narrow
4. Changed bridge material – wood, bone, shell
5. Weighted the fretboard (with clamps)
6. Changed sound hole size
7. Added weights to top plate
8. Knee damping of back plate
9. Make taller bridge by lowering the bridge pedestal
10. Changed bridge break angle

None of these moderated the longitudinal wave twang.

There were two things that DID reduce the twang effect.

Firstly, the pitch of the string can be fine tuned to eliminate the L_wave, as seen in the picture of the note series above. i.e. tune it so a harmonic doesn’t fall at the L-wave frequency. This is fine if you are playing on your own, but no good if you must tune to concert pitch.

Secondly, adjust the position of the saddle. On one dulcimer, over a range of saddle positions from 655mm to 670mm, there was a sweet spot at 660mm where the twang was eliminated. Unfortunately the frets were then in completely wrong positions, being a 667mm VSL instrument, and the dulcimer was unplayable.

Another solution is to have a dulcimer that does not produce harmonics above, say, 2000Hz. A heavy and flexible one. But that’s not the sound I want.

Playing style can moderate it, but that’s another player workaround that has to be managed and reduces playability that little bit more.

This is not a problem peculiar to mountain dulcimers, but is relevant to all stringed instruments, and there doesn’t seem a clear solution. But now you might know what it is when you hear it and won’t waste time trying to eliminate it.

Richard T
Last edited by rtroughear on Thu Sep 03, 2015 10:53 am, edited 1 time in total.
rtroughear
Super Mbr (501-2000 posts)
 
Posts: 612
Joined: Tue Jun 05, 2007 8:18 am
Location: Brogo near Bega, NSW, Australia

Re: An Interesting Dulcimer Experiment

Postby KenH » Tue Jul 28, 2015 8:56 pm

Ohh! I hate that squeak! Many thanks!

I like the unfiltered sound more -- more of that 'high silvery' sound that Robin Clark and I like so much.

Since I play mostly solo, I usually just off-tune enough to kill the harmonic 'wolf tone'.

Once again you've enlightened us mere mortals, Richard. I for one am so happy that someone (you, Mate) is taking the time to investigate the dulcimer scientifically rather than subjectively.
User avatar
KenH
Dulcified! (>2000 posts)
 
Posts: 10213
Joined: Fri Mar 15, 2002 7:16 pm
Location: Afloat in Fort Myers, FL

Re: An Interesting Dulcimer Experiment

Postby rtroughear » Wed Jul 29, 2015 5:25 am

Yes Ken - if I wasn't doing this I'd have to be out with the chain saw cutting firewood - it's winter here Richard T
rtroughear
Super Mbr (501-2000 posts)
 
Posts: 612
Joined: Tue Jun 05, 2007 8:18 am
Location: Brogo near Bega, NSW, Australia

Re: An Interesting Dulcimer Experiment

Postby harpmaker » Wed Sep 02, 2015 1:07 pm

Did you check the string slots to make sure it is properly seated? I have found that sometimes the twang comes from a string that is vibrating in a slightly loose slot or in a slot that is not properly angled back away from the contact point. Many times a couple of quick swipes with the proper solt file kills that twang.

However, if the bridge is really close to the end of the instrument you may be stuck with it. In my experience the closer it gets the more twangy (which some call call silverly) the voice is likely to be.

If the bridge is out on the sound board have you tried dampening the sound board on both sides of the bridge?
harpmaker
Dulcified! (>2000 posts)
 
Posts: 4685
Joined: Mon Feb 10, 2003 12:33 am
Location: Pershing, IN

Re: An Interesting Dulcimer Experiment

Postby rtroughear » Thu Sep 03, 2015 10:49 am

Yes Dave, I know the sound you mean with a poorly seated string at the bridge, or nut, and that's annoying and has to be fixed when it happens, but this is not that. It's the string vribrating along it's length at a frequency determined by the string length and the properties of the string material, and that longitudinal vibration (about 4000Hz for a steel string) then annoyingly reinforcing one of the strings normal harmonics to the point where that harmonic can be clearly audible as a separate sound. This mechanism is fairly well known, I just happened to notice it on several dulcimers recently, and now I hear it all the time. That's ther price you pay for trying to find out something - sometime you'd rather not find it. Like the audiophile with the ear to the speaker listening intently for the distortion he hopes isn't there.
Richard T
rtroughear
Super Mbr (501-2000 posts)
 
Posts: 612
Joined: Tue Jun 05, 2007 8:18 am
Location: Brogo near Bega, NSW, Australia

Re: An Interesting Dulcimer Experiment

Postby Robin the Busker » Mon Sep 07, 2015 5:17 am

Hi Richard,

I've just found your experiments on longitudinal waves.

I have noticed that 17th harmonic (string squeak) but not, until now, known what it was. But I do have a few 'cures' (you know me - as pragmatic as ever!). The first you have mentioned already; and that's to shift the tuning a bit. The second is to shift the string gauge. But what I do 95% of the time if the high overtone is not fitting what I want to play is to simple change my pick and angle of attack. This can be something as quick as spinning the quill I'm using around in my fingers to switch the cut angle on its end - and I can pretty much eliminate the squeak. Or I switch to a boot leather pick of wood pick or whatever other material is to hand. I'm VERY fussy about the tone I want from my instruments and the shape of the timbre - it has to suit the tune I'm playing and the situation. So I know that every pick and strumming style effects how much that squeak is turned on or off. Now I've read your experiments I can see that this is not surprising. Initially you demonstrated the squeak by rubbing the string. And some picks and angles of attack will probably have the same effect of sliding fractionally along the string during the stroke to set off that harmonic (assuming the string gauge, pitch and VSL match up to create it). Vertical strumming quills (Jean Ritchie style) can be particularly 'sticky' and can set off that high squeak depending on the angle of strum and the quill end cut. I played noter drone dulcimer at 10 or so old time and bluegrass sessions at a music festival over the weekend and I was regularly clipping quill ends or switching picks and strumming styles to get the best tone for each tune.

Volume - We've talked a lot about the volume of the dulcimer over these pages. Again, from a pragmatic perspective, the one dulcimer I seem to end up playing the most at sessions now is the Mawhee. It is a very small bodied dulcimer dulcimer strung with 0.020 piano wire tuned up to Gdd. It has no sustain and so is played with a fast whipped quill in the Galax style - and it really shouts out and compliments the other string band instruments by bringing a completely unique timbre to the ensemble. It is not a dulcimer for quiet solo playing at all - but as a string band or session instrument it works wonderfully. The second instrument I'll carry for sessions is a Galax dulcimer itself - as, again, that sits in a unique place within the instrument mix.

Here's a recording of my Mawhee replica made by Kevin Messenger with guitar backing. You can hear what a unique voice the instrument has - not at all the sound that contemporary dulcimer makers are after :( Which is a shame as the Mawhee Graves family have been playing this design in string bands since the late 1800s !!!



John Mawhee and the Gann Sisters - date unknown.jpg


We are so used to struggling with the lack of volume of the contemporary dulcimer played in contemporary styles that it is easy to forget that early dulcimers (and their European predecessors) were regularly built, set-up tuned and played in styles that allowed them to match the volume of other stringed instruments. I don't play noter drone on old dulcimers because I'm a traditionalist, far from it, I play trad styles on old instruments because they are such practical tools for making music with other musicians. In my opinion, pretty much all of the issues folks have today when playing the dulcimer with other instruments in roots music sessions were solved before 1900 8)
User avatar
Robin the Busker
Super Mbr (501-2000 posts)
 
Posts: 1763
Joined: Fri Apr 17, 2009 5:03 pm
Location: Snowdonia, Wales

Re: An Interesting Dulcimer Experiment

Postby WVluthier » Mon Sep 07, 2015 10:18 pm

Robin ,I can't get enough of hearing you play that MawHee. It is such a great sound mixed in with other instruments.
WVluthier
Junior Mbr (0-50 posts)
 
Posts: 11
Joined: Mon Oct 22, 2012 5:54 pm
Location: Bruceton Mills,WV

Re: An Interesting Dulcimer Experiment

Postby Robin the Busker » Tue Sep 08, 2015 2:43 pm

WVluthier wrote:Robin ,I can't get enough of hearing you play that MawHee. It is such a great sound mixed in with other instruments.


The Mawhee is such a wonderful instrument, and very different from other dulcimers 8) I wish more folks would play these old instruments but there's a fundamental problem with the whole dulcimer learning/playing scene that stops that happening. Basically, the dulcimer has been earmarked as an easy instrument to play. And noter drone has been earmarked as the easiest way to play this 'easy' instrument. Well, that's just not the case at all when you pick up one of the pre-revival dulcimers. They are nearly as hard to get a good tune out of as a fiddle (a lot of early dulcimer players/makers were also fiddlers) so they take some musical knowledge and physical skill to master. And, in general, the contemporary dulcimer playing fraternity is just not used to having to put a lot of effort into getting any sort of reasonable tune out of their instruments. Folks would far prefer to have a low action, easy playing DAd, equal temperament dulcimer with a 6+ than sounds 'sweet' just gently strummed with a guitar pick in easy chords, rather than an instrument you really have to 'work' hard with quill and noter to get any sort of decent tone and rhythm out of it at all :? Oh well, at least it keeps the price of vintage instruments low so I can keep buying and playing them :lol: And thank God for those of you who are passionate about making dulcimers to the old patterns to keep us roots players satisfied 8)
User avatar
Robin the Busker
Super Mbr (501-2000 posts)
 
Posts: 1763
Joined: Fri Apr 17, 2009 5:03 pm
Location: Snowdonia, Wales

Re: An Interesting Dulcimer Experiment

Postby WVluthier » Tue Sep 08, 2015 9:28 pm

I know my building style limits me as to prospective buyers, but, I gett such satisfaction when I build an instrument and then see the amazing things folks like yourself do with them. Thats what keeps me building.
WVluthier
Junior Mbr (0-50 posts)
 
Posts: 11
Joined: Mon Oct 22, 2012 5:54 pm
Location: Bruceton Mills,WV

Re: An Interesting Dulcimer Experiment

Postby rtroughear » Wed Sep 09, 2015 10:10 am

Dulcimer Acoustics: Construction Factors for Bright_or Mellow Sound

There’s a moderate amount of discussion about whether a particular mountain dulcimer sounds mellow or bright; but there’s no real definition of what mellow or bright means. Also some of the construction factors that should make a dulcimer mellow/bright seem to have the opposite effect. The wood won’t cooperate.

Despite the lack of technical definition, we mostly seem to understand what’s meant by the descriptions “mellow” and “bright”. Mellow favours the lower end of the sound frequency spectrum either because the lower pitched harmonics are stronger, or because of the absence of higher pitched harmonics. A Bright sound favours the higher end of the spectrum because the lower harmonics are weak or the higher harmonics are strong. There are no doubt other factors involved in the perception, but the relative strengths of the high/low harmonics is probably the main one.

Construction factors that should produce a particular sound sometimes (often?) don’t. Further back in this thread I reported on three dulcimers that seemed reversed in what they should sound like. I couldn’t really explain why that was.

Factors such as the box cubic air capacity; thickness of the plates; general stiffness; types of wood; size and shape of sound holes and others are given as influencing whether a dulcimer will be mellow or bright when it is finished and strung up. But no combination of these factors seemed to help me in predicting outcomes for my own dulcimers. Making measurements after they are built is fine, but things can’t really be changed by then, and we are a long way from being able to use theoretical values of resonances etc to make accurate predictions about the final sound.

By now I’ve made a hundred or so mountain dulcimers, and I’ve kept detailed records for most of them during construction. I’ve written down my impressions of the sound of each one at the time of construction– some of them I have explicitly noted as sounding “mellow” or “bright” to me. Somewhere in all that data there should be clues as to what is different in the construction of the mellow and the bright instruments.

So I looked back over my records and picked out nine dulcimers I thought of as “bright” and nine as “mellow” and compared the construction details.

And for a change, some practical information seems to have emerged. Some information confirms what seems intuitive, but there were also some surprises (at least to me).

The Factors That Were Looked At
I only considered construction factors that could be controlled during the construction process, meaning that subjective factors such as tap tones, or quantitative factors such as final Helmholtz resonance were not considered. The list of factors is hardly complete – it can only include the things I measured during construction. Others might measure different things. Also the analysis is only for my own pattern dulcimers – I can’t say for sure whether the results apply to other patterns, but it’s a start. All dulcimers in the analysis were full-length fretboard designs.

The construction factors I looked at were:
- Back plate wood density
- Side density
- Top density
- Back thickness
- Side thickness
- Top thickness
- Side height
- Fretboard height
- Fretboard width
- Fretboard density
- Fretboard type hollow/arched
- Completed top/fretboard stiffness (static deflection)
- Bracing stiffness – top
- Bracing stiffness – back
- Scale length
- No. of frets – 2 octaves or 2 octaves+3
- Height of fretboard at strum hollow
- Position of bridge from to the internal edge of the end block
- Internal end block length
- Total instrument weight
- Total area of sound holes

The values for each of these factors for each instrument were collated into two spreadsheets – one for the nine “bright” instruments and one for the nine “mellow” instruments. In some cases, such as bracing stiffness, I had to rank them on a scale of 1 to 5 based on my knowledge of the wood they were made of, and their recorded dimensions.

Analysis of the construction factors
I had hoped that just eyeballing the spreadsheets might clearly highlight differences between the construction of mellow and bright instruments, and while some factors did seem different between the two groups I know enough about comparing small sample sets to realise I had to do some actual statistics. So out came the university statistics texts. Some t-tests and F-distributions later at least I’m satisfied that the comparisons are as valid as the data allows.

In essence I wanted to see, for each factor, whether there was a difference between the mellow sounding group and the bright sounding group, and if there was, how significant was it.

Results


For the eighteen dulcimers I looked at following factors had a significant influence on whether a dulcimer was in the bright or the mellow groups (in no particular order of importance):
Mellow_Bright Table.jpg

Other slightly significant factors were:
- Back/side wood density: lighter wood tends to bright; heavier to mellow.
- Height of fretboard: higher (>20mm) tends to bright; lower (< 18mm) tends to mellow.
- Scale Length: Longer tends to bright; shorter tends to mellow.
- Length of strum hollow: Shorter tends to bright; longer tends to mellow (this may equate to the number of frets: 2 octaves +3frets = shorter strum hollow; two octaves = longer strum hollow)

Factors that did not seem significant for bright/mellow sound:
- Top wood density
- Back/side wood thickness
- Fretboard width
- Fretboard Density
- Top bracing stiffness
- Length of internal endblock
- Total dulcimer weight

Consideration of Significant Construction Factors
Top Thickness
This was unexpected. I have maintained that top thickness does not have much effect on the final sound because top thickness variation translates to top stiffness variation, but the fretboard stiffness swamps the top stiffness for the normal range of top thicknesses. So there may be other processes that make a thin top favour the lower pitches, hence a mellow sound. Earlier in this thread is an experiment to test just this – the effect of top thickness change in two otherwise identical dulcimers. The thin topped dulcimer did sound more mellow than the thick topped instrument, and I couldn’t really explain why. So there is some experimental evidence that backs this statistical evidence, and maybe supports many makers’ gut feeling.

Side Height
This was also unexpected. Since all the dulcimers looked at had the same outline, the height of the sides is really a measure of the cubic capacity of the dulcimer box. Conventional wisdom holds that a bigger box should sound more mellow, but this analysis is saying that for my pattern dulcimers, a bigger box is more likely to sound bright. Two reasons for this spring to mind. First a taller box gains stiffness, which in general favours the higher frequencies. Second, a taller box moves the back further away from the top and may modify the top/back coupling of the wood plates via the internal air. The examination earlier in this thread of three dulcimers, where the lowest sides/smallest box sounded the most mellow, supports this current analysis.

Top/Fretboard Assembly Stiffness
This is an expected outcome. As the overall stiffness of the top/fretboard goes up, the ability to support lower frequencies goes down, favouring a brighter sound.

Stiffness of Back Braces
Another unexpected result. But on reflection it might seem more reasonable. Fixing braces to the back has more of a relative effect than fixing them to the top with its very stiff fretboard attached. The back can go from very flexible (mellow) to very stiff (bright) by the application of bracing. In contrast the top stiffness will not change nearly as much by adding bracing because it is already very stiff because of the fretboard. Note that the effect of the added mass of the bracing (favouring mellow) is not as important as the increase in stiffness (favouring bright), as expected.

Height of Fretboard at Strum Hollow
There has been a little discussion about this, but not much. The reason why this factor is significant is not clear although it would tend to increase top assembly stiffness. I don’t know when strum hollows became common, but Galax dulcimers don’t have them, and they seem to be generally bright sounding instruments.

Bridge Position from Internal End Block

It’s known that moving the bridge off the endblock towards the middle of the lower bout can affect the sound greatly. This is generally achieved by lengthening the dulcimer whilst keeping the scale length unchanged. The end block takes up some space within the dulcimer body, unseen from outside, so we usually eyeball the position of the bridge relative to the end of the instrument, but the distance from the internal edge of the endblock to the bridge may be the more important measurement.

Total Area of the Sound Holes
Again, the effect of sound hole size is generally known – small holes favour mellow; large holes favour bright. I would have thought it was a secondary factor, but in this set of instruments it significantly affected the perception of bright or mellow.

Conclusions
I shouldn't be too dogmatic here, but in in general it seems more likely that:
A MELLOW dulcimer will have a thinner top; lower sides; a more flexible top/fretboard assembly; lighter (or no) back braces (a more flexible back); a low strum hollow, bridge further from the internal end block; and small sound holes.

A BRIGHT dulcimer will have a thicker top; higher sides; stiffer top/fretboard assembly; stiffer back braces; a higher (or no) strum hollow; bridge closer to the endblock; and larger sound holes.

Other factors that are not addressed here, such as the width of the lower bout, may also be very significant.

How reliable are these conclusions? Well, the statistical conclusions on this data set are valid. Whether they apply to the whole population of mountain dulcimers is less certain. But you can say that about any constructional advice. There is a spread of values for each factor within the two groups, and some overlap, so no guarantees. Do some tests and see.

Richard T
Last edited by rtroughear on Thu Oct 08, 2015 10:54 pm, edited 1 time in total.
rtroughear
Super Mbr (501-2000 posts)
 
Posts: 612
Joined: Tue Jun 05, 2007 8:18 am
Location: Brogo near Bega, NSW, Australia

Re: An Interesting Dulcimer Experiment

Postby Robin the Busker » Wed Sep 09, 2015 3:11 pm

That's a great piece of work yet again Richard. The shallow body coming up more mellow sounding that a deep body is a surprise although I would say it is something that I have noticed too. I have noticed that shallow dulcimers as they seem to pack a little more 'punch' than deeper bodies. Capritaurus dulcimers were deep and renowned for being mellow, but I have often thought that this had more to do with the long scale rather than the depth of the body. I have a Presnell that is quite deep and not very wide but it is very mellow - possibly because it has a 33 1/4" VSL !!!

Your observation about moving the bridge forward is interesting. The experience I have had of this (not much) seemed to loose treble rather than enhance bass - in fact the dulcimers I've experimented with seem to loose some bass too although they still sounded 'mellow' because they had lost so much top end fizz. I think I'd plump for saying that by moving the bridge forward it emphasises the mid range and in extreme can actually sound a little nasal (sort of the top and bottom rolled off too much like on a telephone). I can sometimes spot a forward placed bridge by the tone of a dulcimer. Something like Walter Miller's dulcimers are very distinctly 'nasal'. His bridge is in-board and his dulcimers have lost all the bottom end but retain a high ring for some reason:


walter miller - west virginia hills clip.mp3
[ 2.1 MiB | Viewed 4097 times ]


On the 26" McSpadden this rolling off of the top and bottom is quite noticeable compared to the 28.5" - the only difference between the two dulcimers is the position of the zero fret and bridge, which are both moved in board. I've posted this clip before but it is possible worth another listen:


26 inch v 28'5 inch 192bps.mp3
[ 2.36 MiB | Viewed 4097 times ]


Do you have a side by side recording of the bridge move effect because the ones I've had experience of may be anomalies?

Robin
User avatar
Robin the Busker
Super Mbr (501-2000 posts)
 
Posts: 1763
Joined: Fri Apr 17, 2009 5:03 pm
Location: Snowdonia, Wales

Re: An Interesting Dulcimer Experiment

Postby rtroughear » Sat Sep 19, 2015 9:34 am

Robin

I haven't done any proper experiments regarding the position of the bridge. It would be hard to do on one dulcimer other than the open strings, and doing it on different dulcimers opens up too many other variables that might affect the outcome. But it's pretty clear that the bridge position does affect the tone, and this analysis just adds statistical support to what we already know.

Richard T
rtroughear
Super Mbr (501-2000 posts)
 
Posts: 612
Joined: Tue Jun 05, 2007 8:18 am
Location: Brogo near Bega, NSW, Australia

PreviousNext

Return to Making Dulcimers

Who is online

Users browsing this forum: Google [Bot] and 7 guests