Standartization of linearity measurements

"Little Johnny (Vovochka, Kleine Johnny, ...), what is the horsepower?"
"It is the force produced by a horse of one meter growth and weighing one kilogram!"
"Interesting! And where have you seen such a horse?"
"It is not so easy to see. It is stored in Paris, in the Bureau of Weights and Measures!"

To keep an uniformity of measurements the next points should be determined:

1. The shape and dimensions of hand simulator.
2. Trajectory of movement.
3. Subject to measurement.
4. Easy way to be replicated.

At the moment I start with such the model:

1. "Arm" = set of two rectangle plates connected to "ground" of theremin (maybe via capacitor 100 pF).
2. “Movement” = rotating of "arm" in a horizontal plane (around the vertical axis). The center of wide plate approaches the antenna surface along the arc with radius R.
3. The subject to measurement is an arc lenght from the plate to the “contact point” of antenna. If just the angle is measured it is converted into distance units (1º = 5.585mm with R=32 cm *).
4. The "arm" may be a thin metal sheet or a cardboard/plywood plate laminated by aluminium foil. The "vertical axis " = microphone stand. A joint&driver = step motor from old printer or FDD.

*Note: 1 step of 5’FDD (1.8º/step) = 1 cm (1,005 cm).

ILYA, I'm all for this so don't take my nit-picking to heart.

I've often thought about what exactly "linearity" of the pitch field means.  Since many Thereminists use hand / finger positions to play certain intervals throughout the pitch field, I think it might mainly apply to this?  If so, does bulk movement of the "arm" sufficiently capture this quality?  I'm thinking it likely does, as the fingers and hand are generally closest to the antenna, and proximity is almost everything when it comes to mutual capacitance.

I suppose I'm at the point where an accurate and verified circuit description, when simulated and coupled to measured (my own arm w/ LC osc) / theoretical (via FastCap) antenna capacitance, is enough for me.  This has the advantage of being adjustable after the fact, so that if the Theremin under test was of the "EQ inductor" type and mistuned, one can adjust the sim to give the best possible response in terms of linearity.  I.e., one isn't locked into a particular set of non-ideal measurements when characterizing a particular Theremin.

Also, I'm somewhat curious as to how the volume side responds as well.  The volume side tends to get short-shrift when it comes to Theremin discussions re. characterization & design.

As for the physical arrangement of the experimental measuring apparatus you propose, I think it would be essential to compare the capacitance to a real hand / arm.  The pivoting nature of the "arm" may be problematic in terms of linearity itself.  I'd do a fake arm with hand, forearm, elbow, and upper arm, perhaps with a bit of conductive body attached.  But since I've already pretty much done this with my real arm and via C simulation I guess I'm pretty much over it.  But obviously not over it enough not to nit-pick other's efforts! ;-)

I imagine much of Theremin playing technique (i.e. finger and hand gesturing) is due to the largely fixed high sensitivity up to now (I don't think there is much you can do to get around the 6 to 8 octave range over ~1/2m sensitivity, other than perhaps make it a bit more linear in the near field via interacting LC such as the EQ inductor).  Once digital means (e.g. the Theremini, the Open.Theremin, or hopefully my own designs) become the norm, will players fall back on the use of the bulk hand / arm location rather than the fingers?  I for one have been holding out really learning the Theremin for lack of a decent, less sensitive (or ideally variable sensitivity) instrument to play (or that's my excuse anyway:-).

Ай да забавный прибор для демонстрации линейности терменвокса! (Или для её регистрации и расчётов?) Такой действительно можно поместить в палату мер и весов! ;) На практике же какой бы хорошей линейностью не обладал данный терменвокс - его собрат будет вести себя по другому. Что хорошо для одного - скверно для другого. Разные помещения, устройство сцены, позиция музыканта, конкретная настройка в данный момент времени... Но через секунду всё изменилось до неузнаваемости. Согласен, что некая тренировка рук нужна, но её нельзя стандартизировать. Единственное мерило линейности (или нелинейности) это музыкальный слух и быстрота реакции исполнителя для мгновенной подстройки к нужному тону. Скрипач, беря скрипку в руки, почти наверняка уверен, что возьмёт нужную ноту, а если он начнёт играть на виолончели - вроде всё похоже, но Линейность уже другая и он может промахнуться. А как модель РУКИ ДЛЯ ТЕРМЕНВОКСА - очень интересно! 

ILYA, this is a very good experiment. It is by exploring one thing that someone might discover something else unexpected.

My idea of Pitch linearity is to have five octaves and the ability to play any of the octaves spread out equally and not need  an octave switch.

My working with volume control this week began with interest in volume linearity only to discover what might be more important would be a response that could do a staccato at the loudest volume but also do it on the quiet side.

Christopher

Christopher, as I understand it the control voltage is generated by the chip U3B and the output 12 is fed to U3A through R39 to terminal 1, where a and changes the volume of the signal. So changing the properties of this control voltage, we can change the attack sound. Is this correct?

Hello Valery,

My volume experiment is an original design of mine from a few years ago. I use pulse width modulation which controls the resistance of a vactrol used as a volume control.  I get a controlled resistance range of 2k ohms to 20 meg-ohms. Not sure if this is a better approach but will have a sound sample in a few day.

Then my 4-layer boards should arrive by friday and I will go back to working on my EWS sound.

Christopher

Good luck, Christopher!

Valery
totally agree that the pitch scale of theremin is a capricious thing. But the emphasis of this topic  is not a "interestingness" of the device, and not the device itself, but to come to agreement on measurement methods (so-called "uniformity of measurements"). Just to be able to correctly compare two instruments.

In the near future (until the new year ), I'l try to assemble anything operable.