# Relationship between Series LC and Parallel LC

Posted: 1/23/2014 3:52:39 PM

From: Thunder Bay, Ontario

Joined: 1/21/2014

So I am modelling the EW theremin and using the following simulation;

I get close to the frequency output range 283kHz to 285kHz.

My questions are, can a person actually add 40pF or more into the equation?

Is there a ration between the Series LC and Parallel LC that needs to be maintained (25:1) etc?

Is the length to frequency matching of the antenna more important than the relative capacitance of the antenna?

thx, any response is greatly appreciated.

Posted: 1/23/2014 4:43:21 PM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as WaveCrafter.com . ...................................

Joined: 12/7/2007

Thierry explains the relationship between the oscillator LC (tank) and the antenna series LC.. Its one of the most difficult aspects to grasp, and Thierry has explained it better than anyone else!

Simulation.. IMO, forget it! - This part of the circuit is almost impossible to simulate correctly - Or to put it another way - only if you set up the initial conditions absolutely correctly to place the oscillator operating point in the right place before you run the simulation, and have C5 set for a small change about the correct operating point (LC) will you get any usable results.. And then you need to convert C5 into a realistic table of distance <-> capacitance (accounting for background / unchanging capacitance) in order to do any useful plots..

Unless the players hand is really close to the antenna, one can expect change of about 2pF over the entire theremin playing range - and this change is not linear - a few hundred femto Farads change when you get further than about the mid point (say 30cm) from the antenna.. So no, 40pF is what you may get if you get 1mm from the antenna or thereabouts - not the kind of capacitance seen in normal playing.. One is looking at perhaps 10pF to 15pF antenna capacitance (I dont know the EW specifics)

*numbers given may be miles out - all sorts of factors come into play...

* Antennas are not "antennas" - they are merely capacitance sensors / plates - there is no "tuning" in the RF sense - the larger the antenna, the greater its capacitive coupling to grounded objects an the greater its sensitivity to the player.. The balance is to make its area large enough to provide sufficient change in capacitance to give adiquate pitch variation when being played, but small enough that background capacitive coupling doesnt 'swamp' the player coupling... In practice, its determined by the circuitry so that it provides, in combination with the circuitry, the correct frequency (and deviation) chosen by the designer - theres no magic number..

Oh - Just noticed the frequencies on your schematic! (7.5MHz to 3.55MHz) - Not sure whats going on with that simulation.. and BTW, which simulator is it? Looks a bit like LT Spice, but not quite...

Fred.

Posted: 1/23/2014 9:08:18 PM

From: Thunder Bay, Ontario

Joined: 1/21/2014

Oh - Just noticed the frequencies on your schematic! (7.5MHz to 3.55MHz) - Not sure whats going on with that simulation.. and BTW, which simulator is it? Looks a bit like LT Spice, but not quite...

Thanks Fred, I'll double check that, but I used 1/(2pi(sqr(40uH*10pF))) and 1/(2pi(sqr(40uH*50pF))) to calculate the Fx.

I am using LTSpice but with the background colour changed. :)

Thank for the link, now I have some bedside reading to do ;)

Posted: 1/23/2014 9:27:25 PM

From: Colmar, France

Joined: 12/31/2007

That's a big mistake! It shouldn't be 4 x 10uH but 4 x 10mH!!!

And a realistic value for the antenna capacitance can only be simulated by putting the 10pF static capacitance in parallel with another C for the hand {Chand}. A more realistic simulation command would then be

.step oct Chand 0.040625p 1.3p 1

which gives 6 simulation runs covering an about 5 octave playing range.

And don't use the standard NPN model for the transistors, but the 2N3904 to come closer to internal capacitances and hfe.

Posted: 1/24/2014 3:51:42 PM

From: Thunder Bay, Ontario

Joined: 1/21/2014

That's a big mistake! It shouldn't be 4 x 10uH but 4 x 10mH!!!

DOH!!

I also made the suggested changes and now get a oscillation change of 337Hz, that seems a little narrow, not much of a note range in there.

Posted: 2/2/2014 12:50:49 AM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as WaveCrafter.com . ...................................

Joined: 12/7/2007

There are some LT-Spice simulations of the EW front-end and oscillator etc here on Element 14, was a long time ago that I did these, wont say theyr great (I really dont remember, LOL ) but you may be interested.

Posted: 2/2/2014 2:14:07 PM

From: Colmar, France

Joined: 12/31/2007

A pitch variation of about only 337Hz shows that the model is not yet ok.

Which exact parameters did you select for the coils L3 to L6? Their SRF is an important factor in that kind of circuit. The Miller or Hammond coils have about 4.65pF parallel capacitance and about 30 Ohms series resistance.

You should run the simulation always twice: once without the whole antenna circuit (starting with L3) connected and a second time with the antenna circuit connected. In the second run, the oscillator's frequency should go up by about 3.5kHz. Tune the free-run frequency by increasing or decreasing the inductance of L1 until you obtain the desired behavior. That's called "tuning the oscillator of a theremin". If well tuned in the way I described above, you should reach a pitch range of about 1700 Hz with the indicated C_hand values.

Posted: 2/2/2014 5:56:40 PM

From: Thunder Bay, Ontario

Joined: 1/21/2014

Thank you all for the responses, they are very helpful. I got the specs for the coils and in reviewing my background material from the Moog etherwave realize the schematics in the two documents have different values for the components, I must have done my stuff with a bit from one and a bit from the other. Ya, I'm a noob ;)

Posted: 2/7/2014 2:36:19 PM

Joined: 2/2/2014

For those interested on simulations, here you will find the LTSpice files for Theremino CapSensor and Moog Etherwave: www.theremino.com/wp-content/uploads/2012/10/Etherwave_and_CapSensor_LTSpice_Simulations.zip

With our libraries you can open any simulation without problems and you benefit of thousands of transistors and diodes. Although all operational amplifiers will be available by selecting them with the most used and finally OpAmp5 and schematics are no longer strange with huge resistors and "bizarre" transistors.

These simulations do not include the temperature dependence of the inductor, but focus only on the influence due to the active components, FET and Transistors. An unstable inductor affects greatly the frequency and can make not relevant, these simulations. Thank you to dewster, to have pointed it out.

Please notice the absolute temperature stability of our CapSensor, also on strong temperature changes (10 to 50 degree).

Another important characteristic (when building digital Theremins) is to generate a high frequency (3 MHz instead of 300 KHz) to get more resolution, when digitally counted.

A third important characteristic, is the low power that minimizes temperature changes in the active element (Transistors or FET)

Finally, use low noise elements to increase the stability, when the hand is far from the antenna.

Posted: 2/7/2014 4:18:15 PM

From: Northern NJ, USA

Joined: 2/17/2012

livio:

1. You didn't include the "EQ" inductance and antenna / hand capacitance in your Etherwave circuit.  This LC resonance dominates that of the tank LC.

2. What is the voltage swing at the antenna of CapSensor?  The theory around here is that a large voltage swing helps to swamp environmental noise.  The Etherwave has a fairly healthy voltage swing at the antenna.

3. What are you doing with CapSensor to protect against ESD?

4. Your waveforms seem to show phase offsets at various temperatures for the Etherwave but I don't know how to interpret them.