I dunno if you guys have seen the above but I was quite amazed at the tone of it the moog melodia, it sounds pretty darn good with just 4 transistors all be it 1960's ones that are germanium (according to Fred on the other thread "Kustom =Melodia").There must be an equilavent transistor today like those fat old 1960's ones.
The 2N68 are Germanium transistors which are very rare today and, still worse, NPN types which were still more rare already in the ancient times. It seems that PNP types were easier to manufacture.
I did some research about the small signal parameters of the 2N68 and remembered that there was a European Ge-NPN transistor with about the same parameters: the AC127, made by Valvo and Philips. The latter can still be found on eBay for not too much money and I think that it's definitively worth giving it a try.
In moments like this I'm almost glad to be old enough to be still familiar with these old beasts, but young enough to remember them. :-)
Hello Thierry & Dewster,
The video of Kip was excellent but did not reveal where the audio went after it left the theremin.
Below is the parameters for the NPN version. I would be curious as to what particular characteristic you would think has the most influence on enhancing a good sound. This data is basic but some datasheets go way over my head.
Thermal drift may be a major issue I agree, but what particular character would you focus on for a special sound. (I know these are not available)
Type Designator: 2N95
Material of transistor: Ge
Maximum collector power dissipation (Pc): 2W
Maximum collector-base voltage (Ucb): 30V
Maximum collector-emitter voltage (Uce): 15V
Maximum emitter-base voltage (Ueb): 15V
Maximum collector current (Ic max): 1.5A
Maximum junction temperature (Tj): 75°C
Transition frequency (ft): 175KHz
Collector capacitance (Cc), Pf: -
Forward current transfer ratio (hFE), min/max: 40T
Manufacturer of 2N95 transistor: CSR
Package of 2N95 transistor: MM1
Application: Power, General Purpose
Transition frequency (ft): 175KHz
This might influence good sound by not producing any sound at all! ;-)
Rather than spending time looking for exact ancient replacements, one's time is probably better spent figuring out how to get a similar sound with silicon transistors. Same goes for tubes.
Dewster, Ge-semiconductors have a junction forward voltage of about 0.15V while Si-semiconductors have 0.6V. At least the VCA stage won't operate without modifying the whole DC working point.
Ge transistors are still available as I wrote above.
I just wonder why RS theremin cited the 2N95 (which was a kind of power transistor in the time) datasheet while Moog's original circuit diagram shows clearly 2N68 which was a low power transistor with a transition frequency > 400kHz which is enough high for the pitch oscillators to work at 150kHz. A short and rough calculation of the voltages and currents in the Melodia circuit also show that the latter's parameters are more than sufficient.
Thermal drift risks naturally to affect the oscillators. But since both pitch oscillators are almost identical, they are allowed to drift since they will be at about the same temperature and thus drift in the same manner which will be nulled when detecting the difference in the mixer. And oscillators with si-transistors will also drift. In the Melodia circuit, the oscillator transistors work with a collector current of 0.5mA and a Vce of about 4.5V which gives a very low power dissipation (2.25mW) which will almost not contribute to thermal problems.
Why still discuss and speculate after an old and experienced geezer like me has already given a working solution with the AC127?
I suspect the sound on that video had little to do with germanium - I suspect it may have more to do with microphonics.
IMO, this may be the great "dark art" to why some theremins sound "better" than others - and similarly perhaps one reason that tube designs sound "better".
If one looks at the physical layout of the Melodia, one sees a plethora of special inductors with their slugs, one sees a construction far less rigid than present PCB assembly, and one sees all this mounted in a lovely wooden box... Note also, a metal sub-chassis fixed firmly to the wooden enclosure (good acoustic coupling) and the EQ coils (both pitch and vol) "suspened" between this chassis and the respective antenna connections..
What I believe may be happening is that the sound causes resonances in the woodwork, these vibrations cause (tiny) movements of components which affect the frequency of the oscillators, this gives (eventually) a change in the heterodyned waveform ( greater 2nd harmonic which may, as audio, couple back through the same acoustic mechanism, to give 4th harmonic and so on)... Also, (particularly with its unusual vol antenna) the volume antenna will be modulated with audio, which should become AM - but AM at the audio frequency with respective distortions and phase differences.
I suspect the above because I have heard the same sort of sounds from my theremins - My H1's were constructed without anything "special" (no germaniums or tubes ;-) .. I set them up on my bench mostly listening to the sound via cans - they sounded great (well, I would say that! ;-) But when I put them into their boxes (which had two good speakers built in and 20W RMS, so it could go loud and really pumped air out of the ports) the sound changed completely, Richer and more "acoustic"... It wasnt the amp or speakers - I tested using my Kenwood KA-3020 and my Tannoy Golds, which usually make everything good sound good.. But these theremins definately sounded better when playing through their acceptable but cheap Maplin amplifier and speakers in which they were housed.
I now know why. My antenna was on top of the speaker box, below it was the bass port (a piece of plastic pipe).. I glued my antyenna wire with a small magnetic connector onto the top of this port - My antenna was an easily removable (but fixed with 4 screws) assembly and the EQ inductor was bolted to the antenna studding and had a magnet on the other end - This allowed me to remove the antenna assembly easily, and more important, put it back easily because the connections found each other without me needing to actually get inside the theremin.
But, the above allowed the EQ coil to dangle being vibrated by the audio - it was close enough to the speakers magnetic fields that Frequency Modulation was occurring. When I was at the show I wondered why the theremins sounded "ordinary" when I listened on headphones (put it down to the crap quality of the cans) and why the volume seemed to affect the theremins tone, sounding best at mid volume (put that down to the amplifiers / speakers)
But I now KNOW that acoustic coupling was a major contributor to the sound, and was playing with sound modulated capacitance plates to connect across the antenna circuit..
I have done electronic frequency modulation of the VFO from the audio, and this does improve the sound.. but not nearly as much as if the sound is coupled through some good wood!
ps .. I would love people to independently test this - move your amplifier close to your theremin, crank the volume up, and see if you can hear a difference. I think with modern construction, this effect will be less (or perhaps not at all) noticable... My theremin was, technically, badly constructed at the antenna end - If I had had more than a few weeks to build 16 of them, I would have been able to take more care and found a different way to facilitate the antenna removal - I have never seen any theremin so optimally "designed" to be affected by microphonics! (LOL)
Microphonics .. Probably for geeks and nerds only
Hijack continued.. LOL ;-)
My first thoughts (which I posted shortly after I thought about theremin microphonics following HO-2010) was that the antenna was the 'vibrating' component - but this hypothesis failed because I noticed more harmonics the further my hand was from the pitch antenna (at the bass end) - at first I put this down to frequency response of the circuitry and my hearing (the higher the pitch, the less audible harmonics are 'available').
But experiments showed this did not seem to be the mechanism.. The mechanism(s) seems to be that, as the hand capacitance reduces, the tiny changes in capacitance (unmeasurable except by monitoring the oscillator waveform) become more significant.. removing my hand and tuning the reference oscillator to vary the audio frequency, I could see distortion on the VFO waveform which was proportional to the volume level - and sweeping this frequency I was able to find audio frequencies where there was more distortion than others - Even when cranked up loud, and at 'optimal' frequencies, the distortion was difficult to see (particularly on the heterodyned audio) and it really surprised me that it had made such a difference -
Also, I found that if I held the pitch constant, even at 'optimal' settings, I did not hear any 'improvement' in the sound even though I could see a tiny change in the waveshape.
BUT - If I changed the frequency, I could certainly hear something different - this 'acoustic' quality that I believe I can hear in that melodia video -
Here is my hypothesis:
Microphonics set the cabinet resonating, lets say that the resonances are complex - one gets probably a major peak, lower peaks and harmonics of these, entirely dependent on the acoustic properties of the cabinet.
Lets say one had a major resonance at 200Hz, another lower at 400Hz, then some other resonances from the side panels giving smaller peaks at 1k, 2k etc - then one gets phase cancellations depending on where susceptable components are placed..
Now, the audio is at 200Hz, moving up - At 200Hz one gets a lot of FM, but, as the frequency moves up, it does not kill the 200Hz oscillations in the cabinet immediately - so one gets say 250Hz being FM modulated with a decaying 200Hz signal - at the same time one picks up any other resonances, and ends up with a complex FM "delay line" which 'retains' qualities from the notes one has just gone past..... This is exactly the type of interaction which I suspect gives a sound a more 'acoustic' quality - too tiny to seem to make any difference, almost impossible to see with a scope, it really shouldnt make any difference...
The same is true for tubes - Microphony was seen as a curse, and tubes were (and are) built as much as possible to eliminate this curse.. But all tubes are still microphonic to some degree. One gets tiny mechanical vibrations of the grid WRT the cathode in particular... Not enough to see anything with a 'scope mostly - but perhaps enough to hear? IMO, Tube amps always sound best when the volume is cranked up - and, IMO, the best sounding tube amps are the ones which are built into the same enclosure as the speaker.
I think that perhaps we have an aspect of our sense of hearing which we have 'forgotten' about but which is still active - it probably evolved for some purpose other than to help us identify the difference between acoustic and electronic instruments (LOL ;-) but I suspect it is probably to do with determining spacial aspectsof our environment - subtle resonances and the like giving us subconcious clues about the matter arround us.. In musical instruments we can hear the difference between a acoustic grand and any electric piano - and I think this will probably always be true - To make a real total simulation, one would need to model every interaction of every micro and macro component with each other, and you would have to make every component responsive to every sound and vibration of its environment.. Even when not being played, the mere presence of a grand piano on a stage will alter the sounds that the audience hears when the rest of the band is playing.
FredM said: IMO, this may be the great "dark art" to why some theremins sound "better" than others - and similarly perhaps one reason that tube designs sound "better".
I have read people saying they swap tubes to find the one that sounds best, likewise is there a characteristic about Germanium or Silicon that would change the sound. I don't think so. Set aside how the sound is played, wave shape or acoustics. The black arts must be those things which cannot be found in modeling.
I have tried various acoustic enhancement but these method are way to difficult to have repeatable sound. It must be done electronically. I have tried several optical enhancements of the audio wave , some with promise, most not. One word that always came to mind is scintillation, I don't know why, but many years later I have yet to figure out what the voice in my head wants me to do!@#$
I am not a dang scientist!
Only in the past year would I say I have the complete theremin package but there is something I am still missing, this is why I continue to evolve.
Fred the reason I used Radio Shack "only" was for the discipline of using limited little pieces. I wanted to develop an understanding of the theremin principles stage by stage. I had used the online stores for years. I never looked at another theremin design and so my approach came from practical experiments. I only had a memory of the Beach Boys at an interview around 1964 and they had a plate antenna theremin.
I put my work on the web because I thought people would chime in and give me ideas during my learning curve. I found not many existed. The best piece of info I ever got was from Don djpb_designs at TW on the versatility of the Vactrol, something I did not know about and never seen mentioned again.
"One word that always came to mind is scintillation, I don't know why, but many years later I have yet to figure out what the voice in my head wants me to do!@#$" - Christopher
Ask the voice this question:
Ionization, Radio (radar), or Interplanetary ?
"magic of the Vactrol"
This has been mentioned here - perhaps called it something else, but I did a posting referring to it once or twice, and warned of its possible demise due to Cadmium and RoHS. Also spoke about the quite similar H11F1 Opto-Coupled FET.
Mentioning Don .. Havent seen him here for a long time, Hope he's ok.