[Harp-L] Comb Material Experiment

[Vern]

For the SPAH97 blind test, four of the harmonicas were played on an apparatus that I made from an old tank-type vacuum cleaner and a light dimmer.  The series motor will run at lower speeds and lower pressures with reduced voltage.  A vac on 120V will pull about 80 inches of water. A harmonica likes about 5-10 inches of water.  The lower speed of the vacuum cleaner greatly reduced the noise but not completely.  We put it outside the room and connected it to the harmonica holder with a garden hose.

A little bellows of the type used to test harmonicas in a music store cannot produce a long sustained tone.  However a big bellows as in an accordion could.  The bellows has the advantage of being quieter than any rotary air pump.  

When using a machine to play the harmonicas, the player’s embouchure is absent and the reed may not sound readily or sound different.  A plastic syringe at the entrance to the hole can act as embouchure and can be tuned to maximize reed response.   

If possible, get the test subjects to agree ahead of time (in writing) that your experimental design is fair and adequate and will demonstrate or fail to demonstrate their ability to perceive a materials effect. Try to get them to raise their objections before the test so you can address them in the design.

To encourage participation, we offered a prize to the participant with the highest correct score. The winner’s score was barely greater than random guessing.
 
Vern                              

> On Feb 26, 2025, at 2:32 PM, <meagher at xxxxx> <meagher at xxxxx> wrote:
> 
> Thanks David! I have already observed - not surprisingly - that there are
> some VERY strong opinions on the matter. The complexity of designing the
> experiment is not lost on me, and it seems that no matter WHAT I do, certain
> people will reject the conclusion if it doesn't match their preconceptions.
> Not really my concern, but I'm eyes wide open on that. 
> 
> I have gone back and forth on the human player vs. bellows topic, and I
> think your suggestion of another apparatus - to isolate things absolutely as
> much as possible - is clever. That may be the right way to go! 
> 
> I appreciate your thoughts here, and if you don't mind, I'd love to chat
> further as I go about the design process, get your perspective on things. 
> 
> Evan 
> 
> From: David Wilson <harpoon_man at xxxxx> 
> Sent: Monday, February 24, 2025 10:05 AM
> To: harp-l at xxxxx; meagher at xxxxx
> Subject: Re: Comb Material Experiment
> 
> 
> 
> I'm so stoked to see this topic come up again!!!  I was standing on the
> sidelines eating popcorn during the great comb material debates of decades
> past, and always enjoyed the fireworks that inevitably followed.  I will not
> be at SPAH this year, but I hope this experiment will happen.  
> 
> 
> 
> About methodology, my recommendations would be:
> 
> 1.	Don't use a human player for the comb testing - too many potentially
> confounding variables there.
> 
> 2.	Also don't use a bellows as that does not allow for a note to be
> held long enough for listeners to discern the subtle differences in tonal
> quality. 
> 
> 3.	I'd recommend finding a way to get a hose/tube or other apparatus to
> continuously blow/draw air through a single hole at constant air
> pressure/vacuum pressure.  With this setup, you could play/record long-ish
> samples for each comb material (say, 15 seconds each) to allow the listener
> to listen carefully and hone in on any differences in tonal quality.  Also,
> you probably want to select a lower hole on the harp that would have better
> potential to get the comb material resonating (if there is such an effect).
> I think both a 3-blow and 2-draw would be optimal as it would give you the
> same note, and this would allow for the testing to detect any differences in
> comb material tone between blow notes and draw notes.  Also, to calibrate
> the volume level created by the testing apparatus, I would recommend setting
> up a sound meter at the testing location; having a human harp player play
> long tones on the test harmonica on the test holes; record the decibel
> readings for those long tones; then adjust the pressure/vacuum of the comb
> testing apparatus to approximately match the volume created by the human
> player.
> 
> 
> 
> Good luck!!
>