RCA Sounds / Recordings / Samples

This is a relocation of the sub-topic (hijack) from here (http://www.thereminworld.com/forum.asp?cmd=p&T=4213&F=715&p=2)


[i]"I'd hate to see some day in the future when all the first theremins are dead and gone and someone goes "Man, I wish someone had made a really clean, technical recording of one of those". - Dae23 [/i]

Those were part of the reason I raised this issue - the other reasons include wanting to 'clear up' the issue of what a "real" Theremin "should" sound like, and wanting to have a "RCA" preset in my Theremin which was at least close to the real thing.

I do realise that there will never be any "absolutely definitive" sample.. I am sure that no two RCA's sounded exactly the same - that even ageing of the 'tubes will greatly affect the tone..

As I see it, we have (or had) several ways of listening to an RCA..

1.) Listening to a live performance of a new-ish RCA played through its speaker, back (say) Pre 1960.. Now impossible without the invention of a time machine.

2.) Listening to a recording of [1] above, made on recording equipment of the time, complete with the 'degradation' this equipment introduced.. I suspect that most of what we percieve the "RCA sound" to be comes from this source.

3.) listening to one of the remaining RCA's with its original speaker.

4.) Listening to [3] above, recorded using modern equipment, which will be quite close to what is heard in [3].

5.) Listening to the signal from an RCA taken directly into modern equipment.

It is not (or, I think, should not be) too difficult to process the signal from [4] to give the coloration added by old recording equipment ("Vintage radio").. And if one has samples of [5] and [4], it should even be possible to derive a 'model' for the RCA Speaker..

As I see it, the best starting point is probably to get samples of [4].. This would be (I think) the closest to the 'original' sound as heard by the audience.. This could be processed to get clean 'vintage' recording sound, which would probably be the sound we all know and love..

Of course, none of the above will do anything for our (or certainly my) playing technique or skills - And none of us is going to sound like Clara even if we had a perfect RCA emulation or a time machine..

[i]"Nothing had been changed. The "vintage" sound all had to do with the equally vintage recording equipment used in the 40's & 50's. In order to emulate that sound, I put the modern digital recording through an FX program called "Vintage Radio" and the RCA sounded exactly like it did 50 years ago." - Coalport [/i]

This is most interesting..
Was the recording you processed with "Vintage Radio" a recording of the RCA played through its original speaker 'miked up, or was it the RCA recorded directly?

Fred Mundell
Fundamental Designs Ltd.
Electronics Consultant.
<- See Profile Image for Email.
Designer of Theremins and other alternative electronic music controllers and instruments.

I would have thought that one of Lev's hand-built originals would be the prototype of what a "real" theremin sound "should" be like. I'm going to go over my copy of "Electronic Odyssey" and listen for clean samples that could be pulled out of it.

What I'd really like to have is a O-scope trace of what the waveforms are. Then I could reverse-engineer how to get that out of a modern box.

FredM asked:

Was the recording you processed with "Vintage Radio" a recording of the RCA played through its original speaker 'miked up, or was it the RCA recorded directly?

***********************

First of all, Hoffman's theremin does not have its original speaker. The speaker in the instrument when I got it in 1998 was a GE replacement speaker that Hoffman installed in the early 1950's. I have no idea what sort of speaker was in it before that.

The GE was very dirty and covered with a dark brown glaze from years of smoky clubs and recording studios. I carefully removed it and replaced it with one of Reid Welch's CLARATONE speakers, custom built for theremins. This improved the sound greatly. I put the old GE away but could reinstall it anytime if the need were to arise.

I can't remember whether I miked the RCA or plugged it directly into the console when I recorded SOMEDAY MY PRINCE WILL COME.

All right, now I have to ask about a Claratone speaker. Custom built for theremins!?

Do tell.

Fred - you said specifically no vibrato, and I understand the reasoning for that, but I have to wonder; given that pretty much everything (up to and possibly including the migratory patterns of the Bulgarian Soup Weasel) affects theremins in complicated ways, it it possible that [i]moving[/i] your hand in the pitch field could affect the waveform in non-obvious ways (i.e other than just changing its frequency.)

[i]"I would have thought that one of Lev's hand-built originals would be the prototype of what a "real" theremin sound "should" be like." – Joe Max[/i]

Probably - As I understand it (correct me if im wrong) the RCA was designed as a low cost version of Lev's "original"..

However, I think it is the sound of the RCA, and particularly Clara's performances on this instrument, which we associate with the "true" Theremin sound.. We really need samples of the Clara-min..

[i]"is it possible that moving your hand in the pitch field could affect the waveform in non-obvious ways (i.e other than just changing its frequency.)" – Gordon C [/i]

Yes, I suppose it is "possible" - but unlikely.. Adding vibrato to any waveform DOES give "distortion" to the waveform during the period where the frequency is changing.. As the frequency changes, the cycle-cycle waveform MUST distort in order to 'become' the new frequency.. As to whether anything 'more' happens to a Theremin waveform..?..

One main difference from waveforms produced by heterodyning instruments would be the fact that it is a high frequency oscillator who's waveform is changing as a result of vibrato - and this changing waveform and its changing harmonics is then heterodyned to produce the audio.. So I imagine that perhaps this could give a different distortion profile to what one would get if directly modulating an audio signal..

But I really dont know.

Ok, so we need a few samples with vibrato as well... Frequency of vibrato? ..?.. Fast vibrato should highlight any dynamic waveform changes more clearly.. Depth of vibrato is probably not too important, as long as there is enough change in frequency to be able to capture waveshape changes over a few Hz.

Oh hell - I need to get a research grant and buy a RCA, then take two years lessons from Lydia, set up a studio exclusively for the RCA, having (as yet unavailable) technology so that the HF oscillator waveforms, the post-mixer waveform, the audio output, and the 'miked speaker can be simultaneously recorded..

More seriously - Someone somewhere must surely have done some research to preserve the ideas behind the Theremin, and its sounds.. I think its time to contact the Theremin institute in Moscow..

What I think I really should do is not be so pedantic.. Until "better" samples are available, I have the old recordings to work with - I will do a RCA preset based on these (and probably get a lot of moans, because as far as I can tell, the "RCA sound" [as heard on old recordings] has a LOT less harmonics and is much more sine-like than people want to believe..) .. I could modify this preset later - or add another - if "real" samples become available.

Fred Mundell
Fundamental Designs Ltd.
Electronics Consultant.
<- See Profile Image for Email.
Designer of Theremins and other alternative electronic music controllers and instruments.

I'm coming at this conversation from FredM's point of view, that part of the original theremins sound was due to the fact that it's output was full of upper harmonics that were tempered by the frequency response of the 12 or 15 inch "full range" speakers they were played through (and they're respective cabinets/baffles).

I first started thinking about the whole thing to try to make my Etherwave Standard sound better (and look cooler, with a diamond baffle).

However, the elephant in the room (as has been discussed before) is the fact that the theremins we're talking about are tube driven. Which is basically the sound everyone is looking for, no matter what electronic instrument it is. Usually this means guitars and they're kin, as far as AMPLIFIERS go. First generation theremins have the blessing (or curse) of tube sound generation. Which most everyone seems to agree sounds just plain better.

Granted, folks like Moog were from the time that tubes were tired and old, and were glad to leave them behind (Imagine a tube Minimoog).

But picture a Minimoog (or guitar) through a high quality tube amp vs. similar quality solid state amp and one can usually hear the benefit of tubes regarding audio. Especially when they start abusing waveforms.

My question is:

Why doesn't anyone make a damn tube theremin!

Arrrggg....


That's my rant for the day. I just hate the fact that the only repository of tube knowledge is wrapped around guitar amps.

As I understand it, the reason why tube theremins sound "better" to many people is that the qualities of the tone actually change depending on the register the thereminist is playing in. It all has to do with the heating and cooling of the elements inside the "magical Inca fire bottles". Transistorized theremins are less interesting to the ear because the wave pattern is fairly consistent from the bottom to the top of the range.

This shifting tone more closely emulates the sound of a human voice (every aspect of which is constantly changing as the singer changes the shape of the throat and all of the various resonating chambers of the vocal instrument).

One thing you can do to emulate the sound of a tube theremin if you are playing a transistorized instrument, is to put the signal through a tube processor of some kind prior to amplification. I use a MILLENNIA ORIGIN STT-1 (http://www.mil-media.com/stt-1.html)

The problem with a unit of this kind is COST. Even on eBay, one of these babies is going to run you a couple of grand which is more than most people payed for their theremins. It is probably not worth it unless you are a fairly busy, full-time professional musician. Devices of this kind are not made for hobbyists and in order to get the most out of them your entire kit has to be high end.

The MILLENNIA has two circuits in it, a solid state circuit and a totally separate tube circuit. You can A/B from one to the other with the touch of a button and hear the difference between the two. If you play a solid state theremin but you want the sound of a vintage tube instrument, this is one possible way to approach it.

About the CLARATONE speaker. Theremin expert and vintage enthusiast Reid Welch created these back in the mid 90's to emulate, as closely as possible, the sound of Clara Rockmore's custom theremin. He didn't make very many of them, and I bought the last one (which I installed in Samuel Hoffman's RCA). Occasionally one of them comes up for sale so if you're interested keep your eyes peeled.

[i]"Why doesn't anyone make a damn tube theremin!" - Dae23 [/i]

They do.. There are still several tube Theremin manufacturers - Thierry pointed one European manufacturer to me some months ago, but I cant find the link now..

Ok - I know 'tubes are loved and adored for (often) good sonic reasons.. BUT.. A tube MiniMoog would never have resulted in the explosion of analogue synthesis we saw as a result of the fact that using semiconductors it became possible to make affordable instruments ("affordable" being relative.. I could never have afforded a new MiniMoog or even the cheapest ARP when I was a teen - But I could just about scrape enough money together to buy parts and build my own synthesisers).

The reality is that if one is making good affordable Theremins, FULL tube Theremins are probably not the right route.

BUT.. I cannot see that MOST of what a 'tube design can sound like could not be produced using semiconductors..

Lets think about the major components of the "vintage recording" sound..
1.) Bandwidth - 20Hz to 20kHz was pure fantasy then..
2.) Noise - I dont really think we want to duplicate that.. Although, due to many factors (including bandwidth) the noise was much less harsh than modern noise.. ;-)
3.) Saturation / Clipping - This, I think, is possibly the most important charactaristic of old tape recordings.. The fact that often the higher amplitude signals were recorded in the extremely non-linear zone of the tape, which gave a 'soft clipping' effect.. This whole area is worthy of a dedicated thesis.. but the complexity of tape recording is absolutely staggering (I worked aligning old tape recorders many years ago in what seems like a different life..) - One has both the recording process and the replay process to think about - a tiny error in the asimuth adjustment of either record or playback machine has huge effects on the frequency response.. The bias level settings are critical, and vary from one batch of tape to another - particularly in the early years.. etc
.. There is also the mechanical aspects of the transport system - Wow and Flutter are well known about - but what about W+F which dropped below unnaceptable levels, but was there nonetheless - adding frequency dependent waveform alteration to the audio signal.
4.) THEN the playback electronics comes into the picture - but the signals have already been 'modified' substantially long before the playback equalisation attempts to restore some resemblance of the original.

As I see it, "Vintage" equalisation to limit the bandwidth from a modern Theremin, coupled with "vintage" compression and either a FET circuit to emulate tube soft clipping, or one real tube to perform this function should be all that is needed to get close to what we think we want.. ;-)

I am thinking about perhaps having a "time machine" knob on my Theremin, which will be scaled "1930" through "1960" at centre position, to "2010" at the other extreme.. and this knob simply tailoring the bandwidth and compression for any sound from the Theremin..

I think that to do the job well, one would be best using a DSP - but I wont go that path (for one thing, I am not competent enough to do it).

[i]"As I understand it, the reason why tube theremins sound "better" to many people is that the qualities of the tone actually change depending on the register the thereminist is playing in. It all has to do with the heating and cooling of the elements inside the "magical Inca fire bottles". Transistorized theremins are less interesting to the ear because the wave pattern is fairly consistent from the bottom to the top of the range." - Coalport [/i]

This is interesting - BUT - I find it hard to believe!

There may well be some mechanism in 'tubes which gives variable wave shape - But I cannot see how this could be from any thermal effects - the thermal latency is far too slow to be able to respond to changes of register or notes.. we must be talking about a LONG time (in musical terms) before any significant change in temperature was to result from even quite a dramatic change in current - certainly 100's of milliseconds.*

Perhaps the mechanism (if such a mechanism exists) is more likely to be due to depletion of the 'space charge' arround the cathode.

Whatever - Altering the waveform as a function of pitch (or for that matter volume, or any combination of pitch and volume) is quite easily achieved with modern circuitry when one has voltages which accurately track both pitch and volume.. This is the method I am using (applied to pre-heterodyned waveforms) to give dynamic tone variation.

When one knows what the waveshape is for any given frequency or amplitude, it is (theoretically) easy to tailor the filters (VCF's) etc so that the charactaristics can be closely duplicated..

All I need is a good, complete set of samples..

;-)