RCA - Signals / Waveforms etc - "Real" .

Thank you, Charlie!

This looks, to me, like its the mixer which creates the waveform - Just as Bob said!

The V7 grid signal has the sort of high harmonic content shape that, when driving the RCA 106 speaker could give the sort of waveforms I am seeing in samples taken from mike recordings I have been looking at recently..

I bet if one was to feed the V7 grid signal into a modern amplifier and Hi Fi speakers, it would sound bloody horrible!

We need more samples and waveforms rrom other RCA's for comparison and confirmation, but this is a huge step forward -

I have been in contact with some RCA experts, and hope that we can get some waveforms from some historic and well maintained Lev Theremins from them.

IMO it is really sad that for so many years, Lev's brilliance has not been fully appreciated, or that the real elegance of his instruments has not been fully understood - even by those "skilled in the art"... IMO, it is almost disgraceful that an institute like the Theremin Center of the Electroacoustic Music in Moscow has done (or seem to have done - perhaps there are archives in Russian which have not become translated / published) so little over such a long period of time to explore and document Lev's creations.

Its time all that changed.

Fred.

" could only get three of thetest points  at the moment.  The other 3 are more difficult to access to access externally"

Charlie - You probably got the most important test points! - V4 Grid is probably the next most important.. If the waveform here is similar to the waveform on V7 grid then we can probably eliminate the transformers as a major player in the sound creation.. If they are similar, testing V4 Plate is probably not needed.

V2 plate will most likely be something of a 'mush' - I expect you will see HF 'encased' in a waveform resembling the waveform you see on V7 grid..

 A lot will hang on what appears at V4 grid - is this going to be similar to whats on V7 grid, or similar to whats on V2's plate? (you may not see much on the plate - a lot will depend on the resistance and charactaristics of the winding its connected to.. You can only view voltage.. you really want to view current here, but unless you have an expensive current probe there is no safe way to monitor the current, as you would need a 'scope isolated from ground and to fit a series resistor to the plate and look at the voltage across this resistor)- or is there going to be a surprise?

Heres are my guesses..

The output of the mixer is going to be (I believe) a current which is the sum component modulated by the difference (wave shape) component - this difference is not going to be a mathematical multiplication of the two input waveforms - tis going to be a function of the biasing / saturation (distortion) caused as the two signals fight (or work harmoniously together - either analogy probably works ;-) over control of the current through V2.

This resulting mix will pull current through the "primary" winding of T7 - and this is my major source of uncertainty.. I suspect that the inductance of this winding (or more particularly, the frequency response of its coupling to the secondary) will integrate this current, and that only (or at least the majority of) the signal which is passed to T7's "secondary" winding (the one connected to V4's grid) will be the difference frequency waveform, stripped of most of the high frequency sum components.

If the above is correct, you should see an audio waveform on the grid of V4 - perhaps with some "fuzzyness" due to some HF getting through.

The gain of V4 will be determined by the volume current from V5, the signal (voltage) on V4's grid will produce an AC (Audio) current through T4's "primary" and the inductance of T4 will further attenuate any HF in the signal - AC Current through T4's "primary" couples to its "secondary" to give the signal you have shown ob the gate of V7.

.. Time will tell ;-)

Fred.

Charlie, you are THE MAN!

My guess is that the transformers will not make any difference. When they are operated near saturation, some sweet sounding distortion can happen, but these look far to large to saturate at the levels we see  here.

RF filtering, maybe some integration, but no magic.

" but it was really hard for me to keep the pitch stable, adjust the scope level, and take a picture   " - Charlie

OMG! - You were using the antenna to set the pitch !? ..

Take the antenna off, and put a capacitor or two from the antenna connection to ground - anything between 6pF and 14pF - by stringing capacitors together you should be able to tune it to any pitch..

Mind you - I would imagine that trying to hold a steady pitch while taking measurements is the ultimate in theremin challenges! -

I doubt if even PP or Lydia or (dare I commit the ultimate blasphemy ?) Clara could do that!

;-)

Fred.

"RF filtering, maybe some integration, but no magic." - w0ttm

Oh, there is magic!

For a waveform like to one seen on V7's grid to become the sound we hear from the RCA - well, thats magic!

I have always suspected that the instrument was designed with a completely different perspective and "mind set" to the way electronic instruments are designed today.. Back then, there were no Hi Fi seperates, no standardised "line levels" - one could not make any assumptions or design for interfacing to "standard" external equipment.

In those days, whatever one was designing, one was designing the whole item or "system" - Be this a gramaphone or a theremin, it encompassed the signal source, the amplifier, the loudspeaker - and one tailored each bit, crafted each, so that what came out of the loudspeaker was what you wanted to hear.

These days we design sections - a system most often contains just the signal source and any shaping / equalisation / colouring such that, when connected to a standard "perfect" amplifier "perfect" speakers, we get the sound we want.

The RCA was designed to USE the charactaristics of the speakers it was designed to work with - I believe that the pulse nature of the audio signal was crafted to excite the speaker so that its "imperfections" and (by todays standards extremely) limited frequency response were actually exploited as an "electro acoustic resonator / complex filter" -

What we hear from the RCA theremin is the result of this crafting - Take an RCA, give it a line-level output from the V7 Grid or Plate signals, drive this signal into a modern Hi Fi, and it will be HORRIBLE! - Just look at that signal! An Enkelaar switched to its "wasp in a jar" sound will be smooth and sweet by comparison!

But look at what comes out of the RCA speaker - You see complex resonances, HF attenuations, and acoustic features you do not usually get from electronic instruments.. Look at the spectrum from these waveforms, and you see a sharp cut-off as the speaker attenuates harmonics it is unable to respond to - and this cut-off is quite low in the audio spectrum - quite "laughable" by todays Hi-Fi Standards ... But exactly what one needs to cut away what would be extremely unpleasant harmonics fron the RCA if they got through.

There certainly is MAGIC! And I do not think that we will replicate this magic simply by cloning the oscillators and mixer - We need to replicate the RCA-106 charactaristics as well - If we dont do this, we may as well just play with an Enkelaar.

Fred.

To me, the waveform on V7:G is quite similar in appearance to the raw waveform from a larynx - its the kind of waveform I have fed to vocal formant filters.. I suspect that the spectral distortions from the RCA-106 and this "vocal excitation" signal is part of the reason for the RCA's tone.

I belive that you are partially correct, fred.   The speaker adds a lot of warmth and character but I dont think that it is exclusive to 'the Sound'.      Ive played mine through modern speakers either via an impedance matching tx,  well as through external amp (connected before final amp stage.)    The 106 is definitely preferable, not necessarily because it 'adds' to the sound-  in fact it actually takes a lot a way -   Modern speakers seem reveal a  harsher, and generally, more 'raw' sounding tone. Louder and  almost too much character. It is not terrible....more like RCA with overdrive...-sure that this could be corrected w/ a well trimmed mixer.    I havent heard an enkellar personally...but Id wager that I would still prefer my RCA.       

ALSO!   -AFAIK,  the Hoffman (RCA) theremin uses a modern speaker - or at least it had been modified w/external xlr jacks to more easily accommodate recording/studio sessions. 
  It is probably the most recorded/ most heard RCA theremin.  Possibly @ line level?

"Modern speakers seem reveal a  harsher, and generally, more 'raw' sounding tone. Louder and  almost too much character. It is not terrible....more like RCA with overdrive...-sure that this could be corrected w/ a well trimmed mixer. " - Charlie

Ok, then there is no magic - w0ttm is right :-(

A simple HF roll off, or parametric EQ .. No magical resonances in the 106 , no fine tuning of the waveform to exploit these..

Oh well - It was a nice delusion while it lasted! .. Hell - it might even be that the wonderful sound from Clara's theremin really has almost nothing to do with the waveforms eminating from the speaker - Perhaps it was just Clara's mastery which made it.. or the fact that the recordings we have had the added warmth of tape charactaristics..

Anyway - if its as simple as that, then from the waveforms you have presented, cloning a solid-state RCA shouldnt be hard..

Fred.

... Alternatively, just perhaps, its down to perception..

You are used to hearing the lovely sound of your RCA clone throught the 160 speaker - this sound is imprinted on your memory and encoded in your synapses.. So when you play your RCA through a amplifier / speaker which does not distort / limit the HF, you notice the difference, the increased harshness, but your "interpretation" of this sound may well be different  (possibly less objectionable) to how others hear it.

Just a perhaps.

Now we are entering familiar ground for me.

There are circuits that take a raw guitar output and shape it so it sounds like it went through a tube amplifier when sent to a full range PA system.

They are just soft clipping and EQ. LED's in the feedback circuit of an op amp are impressive, getting very close to natural tube distortion.

They can also give the asymmetrical wave form we see from the RCA with very little effort.

With a reasonably decent signal from the mixer, only the sharpest of ears will be able to detect a difference from the real thing, and Spock has not logged on for a while....

w0ttm -

If I read you correctly, you are saying that a major factor in the RCA sound could be the charactaristics imposed by the tube power amplifier (V7) - I presume that when Charlie took his output via a matching transformer, he took it from the speaker terminals..

(Ouch! - I have just noticed that one of these is connected directly to 190V DC - not even via a limiting resistor! - POTENTIALLY LETHAL ! - YOU DO NOT WANT TO TOUCH THE OUTPUT JACKS! THEY PROVIDE TWICE THE VOLTAGE AND MANY TIMES THE CURRENT REQUIRED TO KILL YOU ! )..

Certainly looks to me like "safety critical design" had not been invented yet! LOL - But even so, having that kind of voltage on accessible plugs is utter madness! - particularly as one additional transformer on the output would have eliminated the hazard..

We have not yet seen the waveform on the plate of V7 - but it may well be that the waveform here is softer than that on V7's grid.

And I think, in terms of a "clone", you may well be right - Soft clipping or at least deliberately introduced non-linear transfer function, followed by some filtering, may be enough.

Fred.

I actually find it bizzare that 110V AC was chosen as the mains supply for America, and my understanding was that one of the reasons was safety - It was (almost correctly) believed that this voltage was just on the threshold of safety, providing less than 12mA through a body with resistance not lower than 10k ohms - which was deemed "normal" resistance unless the victim was wet.

- And yet RCA was able to manufacture an appliance which supplied 190V DC to a domestic loudspeaker, and was not even compelled to attach a large warning sign on the instrument..

CRAZY!

Yes, the power amp and speaker will modify the sound quite a bit, compared to modern, full range systems. I wouldn't say that's the prime engine, just a contributor. My guitar amp speakers have limited response, with various peaks and valleys, and I'm sure the '106 has this trait in common.

Just like with guitar, the signal source is important, and the waves from Charlies clone will come in very handy.

The RCA power amp is a single ended triode. These produce significant, asymmetrical distortion, especially at full throttle. Lousy for hi fi, but great for an instrument.

Yes there is LETHAL VOLTAGE at the speaker terminals.