Scientific experiment -- design phase

Tyghress · Post by **Tyghress** » Sun Jul 07, 2002 11:16 am

I started noodling around with the idea of doing a truly scientific examination of the differences between my various whistles after I found my decibel meter in an old box of electronics. So far I'm just designing the actual experiment, and I'm looking for input. To start with, I will only test soprano D whistles.

First, I wanted to know what the ranking was for volume, because everyone talks about it, but I've never seen any actual numbers. In this fairly uncontrolled test I set the meter up at about 4 feet from my head, and put the tuner next to it, then had to pick a single note to judge. I chose G.

Here I ran into my first surprise. I nearly always overblow my notes. To keep it in tune, I had to play EVERY ONE of my whistles far more gently than I had been. The Burke WBB was the only one that I could adjust so that my 'normal' breathing would hold that G in tune.

At this point I stopped because I wondered about all the other whistle keys. To my utter horror, my absolute fave play alone whistle could not, under any circumstance, be played in tune. Good thing for me, I suppose, that it IS my play alone whistle. Every other whistle could reasonably be played in tune, at its marked setting, by backing off on pressure.

Back to the D's. I have on hand: Burke WBB, Feadog I, Clare, Gen brass, Sweetheart, Sweetone, Meg, Doolin 2-piece, Susato VSB and old Soodlum. Somewhere in the house is a Clarke.

In the original test I didn't have multiple runs, and I wasn't measuring too accurately, more eyeballing the number from 4 feet away. The only whistle that stood out was the Burke which was about 5-8 decibels louder under the above circumstances.

So I have the sketching of the first experiment. I would do each note from D to high D to higher D, maybe five times and have someone else read off the volume and note them down. Is there anything significant that I'm missing in my design here?

Other experiments in the back of my mind:
measuring 'takes a lot of breath', 'chiffiness' (this should be good...we can't really agree on what the word means, no less how to measure it!) 'purity'. For these last two, I expect I'll need to look at the actual wave patterns produced; does anyone know of a very cheap (read: freeware) piece of software to help with that?

Thanks for input. If you AREN'T interested in this experiment, please just don't post to this thread.

_________________
Tyghress
...And I go on, pursuing through the hours,
Another tiger, the one not found in verse.
Jorge Luis Borges

<font size=-1>[ This Message was edited by: tyghress on 2002-07-07 13:17 ]</font>

Isilwen · Post by **Isilwen** » Sun Jul 07, 2002 12:08 pm

Cool Idea!

So basically, you're going to check the tuning as well as the decible output?

Tyghress · Post by **Tyghress** » Sun Jul 07, 2002 12:33 pm

Elf, my ear is atrocious. When running a scale, we really do 'blow notes into tune', so in the most part, I can get the note I want. But until I have a method of controlling breath pressure across notes, I'll have to pass on that.

Any suggestions on how to do that?

MikeClem · Post by **MikeClem** » Sun Jul 07, 2002 2:25 pm

tyghress,

As long as you've got the dB meter, a great thing to know would be: On any given whistle, how soft is the bell note and how loud is the high B (in relation to something in the middle, say G)?

Mike

Sean · Post by **Sean** » Sun Jul 07, 2002 5:10 pm

So it sounds like a great experimemnt. I might suggest that you implement some Control to the whole thing. As you have already noted (no pun intended here) a note is blown into tune, therefore taking into acount other variables; bore size, window width, etc. it is safe to assume that you cannot provide the same air pressure at a constant rate with which you can give a Controlled comparison of volume. I would suggest that you use a regulated air (any type of gas) supply and a given distance from the mouthpiece (probably 0 since it goes In your mouth). Also temperature should be regulated, but since you probably have all of your whistles in one room that may be a moot point. However the flowing air should be at a temp. approximating body temperature as a colder gas will tend to condense in the whistle causing blockage and thereby a quieter sound. Or not if you are using a dry gas (compressed Nitrogen, Argon, etc.). Just an Idea.

Tony · Post by **Tony** » Sun Jul 07, 2002 5:24 pm

Tygress,
What db ratings were you getting? Do you plan to graph each instrument thru their entire range?
You mentioned you have the meter 4' from the whistle... for what it's worth, speaker companies test all their products one meter in front of the speakers using only one watt of power. They acheive sound pressure levels (SPL) as high as 106db for highly efficient units and most are rated between 85-95db.
Many specifications also include a second rating for the speaker when driven at it's rated output... say 155db @ 500 watts.
Some db ratings for reference:
Conversation (whisper) @ 3'= 30db
Conversation (normal) @ 3' = 60db
City traffic (average) = 80db
Rock concert = 150+db
lion roar = 165+db

Daniel_Bingamon · Post by **Daniel_Bingamon** » Sun Jul 07, 2002 6:40 pm

Sean,

Good point on the temperature. You will also find that the metal whistles take a little long for stabilization to occur.

Any whistle that is checked for tuning should always be warmed up first. One way to quicken the warmup is to blow air into the opposite end of the whistle with all the holes covered with your fingers.

You know some people have cold hands and others are always warm. I wonder how much that affects tuning on the high-D whistles.
That also would be worthwhile research.

curioso · Post by **curioso** » Mon Jul 08, 2002 1:16 pm

I've noticed that some of my whistles (notably my Dixon brass telescoping D) have a broader "sweet spot", which I define as the minimum to maximum volume which one can play a note and still be "in tune". On some notes, my Dixon seems to go from very quiet to moderately loud and remain in tune. Other whistles seem to be more touchy. I think it gives the whistle more "feeling". Is there a proper musical term for this?

It might be interesting to see a graph with frequency on the y axis and volume on the x axis for a given note (bell or G). I'd anticipate a response with a plateau around the proper frequency for a given note, with varying sizes of plateau for different notes and whistles.

Cees · Post by **Cees** » Mon Jul 08, 2002 1:36 pm

On 2002-07-08 15:16, curioso wrote:
I've noticed that some of my whistles (notably my Dixon brass telescoping D) have a broader "sweet spot"...

Dixon brass telescoping D? Is this a new kind, or one I haven't heard of? I thought the only telescoping whistle was Noah Herbison's laughing whistle.

About the experiment, Tyghress, great idea. Let us know what you find out.

DanD · Post by **DanD** » Mon Jul 08, 2002 3:16 pm

Tyghress -
You can find a good freeware tuner at:
http://www1.ocn.ne.jp/~tuner/tuner_e.html
It will tell you what note you're playing, if it's in tune, and has a dB meter in the upper right hand corner. It can also show wave forms, although I haven't played with that feature much.
Best of luck with the experiment - looking forward to seeing the results!

curioso · Post by **curioso** » Wed Jul 10, 2002 9:22 am

On 2002-07-08 15:36, Cees wrote:

Dixon brass telescoping D? Is this a new kind, or one I haven't heard of?

Oops! It's not truly telescoping like Laughing whistles. The whistle IS fairly new, but has been mentioned several times on the Board. To be more precise, the proper nomenclature per Dixon's website is "Tunable Polymer Whistles with an internal (brass)telescopic tuning slide". Sorry for the confusion! I didn't mean to activate anyone's WHOA!!! Of course, if you don't have one, they're REALLY nice!

dkehoe · Post by **dkehoe** » Wed Jul 10, 2002 10:49 am

Tyghress -

Great idea! I actualy found a doctoral thesis on the internet by a woman who was investigating the vibrations of a flute body by rigging up a controlled air source (BTW you have to use air, the density and viscosity of other gasses are different) to flutes and measuring the vibrations on the body of the instrument using a phono cartridge. But she couldn't use a real flute, so she put a recorder fipple on the top, and she couldn't use one with keys, so she drilled holes in the side (it ended up looking remarkably like a low-D). I'll post the link when I find it.

She had good ideas on controlling the air stream, although as an improvement, you might run the air through a container with body-temp water in it to get the right humidity. If you were able to measure the air flow, you could determine how much air is required for a given note, how easy the whistle over-blows, how in-tune the whistle stays with respect to blowing velocity. with the spectrum analyzer you could measure the chiff when you interrupt the air flow, what the overtones look like, etc. Finally, some science behind all this "woody", "reedy", "bright", "soft" stuff.