New Vacuum Tube Theremin Design

Thank you ILYA.  It is a simple schematic.

I used a gas discharge just for fun.  I could also use a vacuum tube for the rectifier, but I decided to use solid state for this theremin.

My next vacuum tube theremin will have no solid state electronics.

Hi all,

I have essentially finished my first vacuum tube theremin.  It was an interesting and incredible learning experience.  I will be posting on my site multiple examples of the tonal range and some 'performances' - I hope.

I am very interested in the huge effect on tone that an air-gap variable capacitor alone has when coupling the two signal grids of the mixer.  I have not added any other tone altering circuitry, so, other than the inherent effects of resistors and capacitors necessary for normal operation, the sine wave output from the oscillators is changed at the mixer solely by coupling effects.

https://markaudiomusic.com/theremin-design-4/

Mark

I like the sound very much.  I like the transition for a pure upper register to a very warm/rich lower register.  Have you graphed out the pitch vs hand distance curve?  

Thanks for your comments.  As regards hand/distance range, I don't keep it constant.  I change it according to the range of the music I am playing.  I can get two good linear octaves with the closed-fist/open-hand spacing, but I prefer the non-linear violin type spacing for most playing.  It just feels more natural, but does require a different hand technique from the linear pitch style.

I have just posted a new wav file and accompanying graph showing 12 different tonal effects (there may be more but I only tried 12) through the use of a single variable capacitor.

The files are at the bottom of this page.  https://markaudiomusic.com/theremin-design-4/

"I chose 5814A tubes for the oscillators.  They are 12AU7s, but are really designed for digital computer use." - markaudiomusic

Yikes!  Sure glad I don't currently live in the tube-based digital computer era (I had enough sense to be born right at the tail of it all :-).

Mark, the coils inside the chassis are really close to the chassis itself.  There is likely some coupling going on between the coil and the chassis, acting like shorted turns.  I recommend you move them at least 1" away from anything conductive.  This is something ILYA pointed out to me on my own prototype.  Having them inside the chassis is probably good as it minimizes interference, but unwanted shorting-type coupling should probably be avoided if possible (it can lower the Q, raise the resonance frequency, and indirectly contribute to drift).

If you haven't played around with a physical coil designing tool you might want to try Inca.  I learned a lot about coupling with it.  (Not sure I'd trust the self-capacitance values from it, however.)  At a distance of 1" coupling is down to a few % or so.

Yep.  One 12AU7 - two triodes - the beginning of simple logic circuits.

The first computer I used at Dartmouth, N.H. in 1967 occupied a whole floor of a building and boasted two kilobytes of hand-wired toroidal core memory!

Thanks for the tip on coil/chassis proximity.  It does have an effect on coupling for sure.  I noticed, even during my breadboarding stage  proximity coupling effects. I will try moving the coils a bit within the chassis and monitor the results. Improving the Q would help solve some minor issues I am having with this version.

At the moment I am concentrating on improving my playing.  My wife wants to play duets with me on the other theremin we still have.

Regarding coil proximity effects.

Early on in my breadboarding stage, I discovered very strong tonal changes in the audio output when I changed the orientation of either the FPO or VPO coil even though they were 10 cm apart.  It produced some beautiful sounds.  I considered and then sadly discarded the idea of suspending the FPO coil on a wooden dowel in some manner so that it would rotate and I could use it as the tone control.  The issues of space and connections, though not insurmountable, led me to the current solution.

Just to be clear, I was talking about the coils coupling with the metal in the chassis, which is an electrical short so it consumes energy, so it lowers Q and lowers the inductance value.  I wasn't talking about coils coupling to other coils, nor about the electrical or magnetic coupling of multiple oscillators in general.

I do think your approach to tuning for Theremins which use EQ coils is good.  An EQ coil Theremin needs at least two tuning controls in order to be able to fully adapt the oscillators to all environmental (capacitance) possibilities.  Pick 2: antenna pad, variable tank pad, fixed tank pad.  The antenna pad is the squirreliest (3x gain) so I'd probably go with the last 2.  The EWS has only the third accessible on the front panel, the first is a bendable wire inside, and the second is a variable L inside - and it all goes to hell when you remove and replace the wooden cover (wood isn't the best material to use in a highly sensitive variable C device).

Yes, I did understand your concern. The coupling effect to the chassis is most noticeable with the VCO, where I need the Q to be very high.  It seemed to suffer the most in the move from the breadboard to the chassis.  

 I still wonder if playing with the direct coupling of the FPO and VPO coils might have a practical use in theremin design of any type.

 Saw you added some stuff.  Yeah, my second theremin - basically an EWS copy, suffers a lot from removing the cover.  

 Yesterday I tried bending the wire on my vacuum tube theremin from the FPO antenna to the Antenna coils to lay along the chassis side.  I had to add 2.5mH to compensate.  That's a lot.