Low voltage tube mixer ?

Posted: 4/6/2014 1:20:25 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

Sorting through the last of my collection of electronics mags I am having to dispose, I came across a DRM mixer circuit in Elektor 7-8 2005..

I was puzzled because it didnt have any HV, so checked the data sheet for the tube, and it looks like it might just work down at 6V..

My thinking is on these lines - If it works at 6V, it should certainly work reliably as a mixer at 12V.. Two tubes with their heaters wired in series, one as a mixer, the other as a pre-amp (or VCA?) driving a small audio transformer (perhaps also replace R4 with a small aidio transformer) , and one may have a really simple way to the "valve sound" without all the hassle of a HV supply (one could run the whole thing from an external 15V or 18V SMPSU  fed into filter and 12V linear regulator)..

You would need transistor oscillators etc to make a full theremin - but the sound IMO is mainly the product of the mixer and following stages..

(oh - this tube [ EF95 / 6AK5 ] seems quite common - I found some NOS @ £6 each, and plenty NOS @ <£15 )

One other thing - My knowledge / expierience of tubes is minimal (perhaps even dangerous!) so anything you do relating to the above is down to you! - I can offer no help, advice or consolation! ;-)


ps - according to the article the antenna signal in the above circuit is  tuned to 5990kHz, mixed with the 6MHZ local Xtal oscillator and the output is 12kHz (?) .. Whatever - I suppose its close enough to the expected 10kHz, or they aproximated the frequencies a bit... R1 can be between 0 and 1k, and sets the operating point / grid bias..

pps.. Looking up RTL2 DRM to find out what it actually was (ie, understand it, LOL ;-) I actually found the schematic from the Elektor article..


Posted: 4/6/2014 5:05:12 PM
RS Theremin

From: 60 mi. N of San Diego CA

Joined: 2/15/2005

Hello Fred,

Back in the 1950’s I remember my neighbor had a remote control airplane hanging in his garage and it had these miniature peanut vacuum tubes in it. Later I would buy these tubes with spaghetti wires hanging from them at WW2 surplus warehouses. I did not do anything with them but they fascinated me at 12 yrs old. Maybe I needed to see a doctor (-‘ The only reason I experiment with electronics today is the staring at little colored pieces, eye candy. I have always wondered if the Tvox Tour was using a peanut tube/valve on the front end and that is its secret?  Thierry hello…


Posted: 4/6/2014 9:29:42 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

Utterly and completely OT ;-)

Hello Christopher,

Yeah - there were little wire ended tubes back vefore transistors (which originally were built into little glass "test tubes" - I have a few germaniums like that) and some were low voltage..

But they are gone -

I have looked for these small  low voltage wire-ended tubes in the past, (thinking they would be ideal for hybrid instruments)  and not found them

I am sure the Tvox doeant have a tube! .. For one thing, its got a lovely sound, but its certainly not a tube sound! ;-)


"The only reason I experiment with electronics today is the staring at little colored pieces, eye candy."

I know that feeling! - I was into chemistry for that reason - all those lovely colored crystals and stuff.. Cobal chloride was red crystals, copper sulphate was blue - then there was iodine crystals one could heat and it precipitated into this 'sheen' (and use to make powder that exploded when people walked on it..) I had this book on how to make fireworks .. Even childrens chemistry kits in those days were damn dangerous and contained substances you cannot buy as an adult today! - and then the colors one got when you mixed these with magnesium and potasium nitrate and sugar and set fire to it... ;-) ... The fire brigade paid regular visits to my local park (well, that was actually a small field next to the old Greenpoint Stadium)  top left of this picture)  after I had tried to launch some rocket propelled with various mixtures of solid and liquid fuels.... It was a great place to experiment because one could hide the kit in the stadium (which was rarely used and quite run-down at the time, with no security) and hide there when the cops came..

I played with stuff that today are banned - Benzine and Triclorethelene, mercury and mercurichrome.. There was an accident once just outside our school, a vehicle with a load of mercury spilt its cargo and there were buckets of the stuff on the road - All the kids grabbed any jars and bottles they could find - we were scooping it up with our hands and bits of paper - I ended op with a coke bottle full.. Tiday theres a panic if one breaks a small mercury thermoneter!.... But the stuff never did me any harm!  - Its not like I turned mad or anything!!!   ;-)

I was about 11 when those facinating electronic bits caught my eye and my career direction turned away from chemistry.. If I could have afforded a saxophone I might have ended up a musician - They really appealed to me because they looked beautiful and shiney (and sounded great, but that was probably less important! ;-)

(I never gave up completely with chemistry - My last act as a "chemist" was making a gunpowder + iodine flash powder "putty" mixture which I put into the bells in my school [Camps Bay High] and they all blew up when they sounded "End of lunch break" ;-)

Posted: 4/7/2014 4:33:27 AM
RS Theremin

From: 60 mi. N of San Diego CA

Joined: 2/15/2005

Fred stop mentioning blowing things up, the internet is monitored! NSA

Here is the data sheet on the 6AK5 that states the typical plate voltage is 180 volts. In my design I also use a pentode but for the oscillator and run a plate voltage of 112 volts @ 5 ma.  In one experiment I dropped the 12v-heater voltage down to 7 volts and found it still worked. I posted my Hartley oscillator circuit at TW a few years ago and there was no interest. Also in my design if I use glass tubes, which are purtier, instead of metal enclosed I pick up too much 60 Hz hum from the environment. All the theremin tube/valve guys seem to have faded away. The main thing is, it is not a practical approach cause when it breaks who ya gonna call... Theremin's Ghost Buster? (-'



Posted: 4/7/2014 10:47:44 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

ADDED -> I just found this article on using these tubes at 12V

It looks like one should be able to build a full theremin using tubes running at 12V !  ;-)


"stop mentioning blowing things up" -

Oh, that was a long time ago - back before 1971 ;-)

Yeah, I looked at the data sheet - but thats the thing about tube data - they give maximum ratings, but tubes work well below these - I just wish I could remember all the theory.. I did a massive lot of studying on valves (even though everyone was going over to solid state, the textbooks were about 50:50 - So I studied the physics of both - the space charge 'rounf the heated cathode is about all I can remember ;-)..

I think it more-or-less comes down to the effect of cathode-anode potential and the effect of this on transconductance - and I guess that pentodes could operate at lower potential because of the screen grid - but I really dont know..

Elektor circuits tend to be more reliable than some other publishers I think - And this circuit seems to have been picked up and reproduced, with people claiming that it works.. I dont much trust the idea of 6V - AFAICS there isnt enough grid potential available to be safe (Gm chart seems to drop off at about 6V) but 12V looks to be getting into some reasonable operating area - perhaps.

Ive got to try it - Just putting my modular together now (an EW with chunky buffers on the oscillator outputs - this will drive an "expander output" onto which one can plug whatever mixers you want.. A perfect jig to test a tube mixer - I didnt design it for this purpose, I designed it so I can compare various mixer topologies including "E-Pro like" "4Q" and "Lev like (tube emulation)" .. My thinking is that a beefy buffer for the EW, which drives its internal mixer but is also taken to a VGA socket at the back of the EW, so one can use a VGA lead to connect to an external "expander" box fitted with whatever "voicing" circuit you want (or I provide ;-) - Be this a tube mixer, or a register switching mixer, or an additive engine - complete with vocal formanter or whatever - might just be the most viable route to make any money from theremins ;-)


Posted: 4/8/2014 8:52:53 PM

From: Texas

Joined: 2/22/2010

Greetings Fred,

Interesting thread.  Have you looked at these low-voltage Russian subminiature tubes?  They were designed for low voltage applications, and plenty of NOS tubes are available from $5 to $10 I think.



Posted: 4/8/2014 11:25:43 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

Hello BB,

I am a little worried about using Russian NOS - Been there before with ferrites, was told these were available in the quantities I might need, but whan I came to ordering 10 I waited months before being told they were "discontinued" - LOL - I knew that...

The tubes I am most interested in are the Chinese 6J1 and 6J2 parts - These are apparently in production (New, not NOS) http://www.ak-modul-bus.de/stat/roehre_6j2.html and cost 5.70 euros. I have ordered a few of these to play with (and a few sockets ;-) Reading about these tubes in the dot amp forum, they werent highly rated (tubes costing 3*+ the price got highest ratings) but I doubt that any of these tests/comparisons will be applicable for tubes run at 12V .. I imagine even the best tube will sound shit under these conditions unless some extra "twiddling" is done to the circuit.

To me, the biggest reason for not going near tubes has been the power supply requirements - As soon as one must build a mains input power supply into a product, things get horribly complicated from a legal perspective - One can get away with self-certification based on in-house testing and one visit to an EMC lab, but as soon as you have mains into your product you need the full certification covering Low Voltage directives etc - and you need to administer all the required formalties about getting certificates from part suppliers etc..

If I can have tubes powered from a standard available external supply (I am thinking of SMPS designed for laptop computers, giving 18V at several amps, and costing < £15) and have linear filters and regulators dropping this to a clean 15V or 12V DC inside my equipment, then production becomes feasable. Wiring the tube fillaments in series and having a LM317 regulating the current at 170mA with slow start-up and turn-off should give minimum dissipation in the regulator if powered from 15V, and preserve tube life..

All that remains then is to see if the tube sound I can obtain from an anode voltage less than 15V is really worth the effort.



Posted: 4/10/2014 4:30:08 PM
RS Theremin

From: 60 mi. N of San Diego CA

Joined: 2/15/2005

Fred Said: “If I can have tubes powered from a standard available external supply (I am thinking of SMPS designed for laptop computers, giving 18V at several amps, and costing < £15) and have linear filters and regulators dropping this to a clean 15V or 12V DC inside my equipment, then production becomes feasible”

Somewhere I mentioned having a low end wobble in my tube/valve hybrid due to AC power line voltage fluctuation. In researching how to overcome this in a simple way, I think Fred and my journey has met up. My thought was how could I regulate the anode/plate voltage better. This idea may not work do to noise but I am always game to try anything. There are 1000 reasons a theremin will not cooperate and I have met most, I am due for a breakthrough. LOL

First I will start with a well-regulated 12-volt supply to power everything, including the 12v filaments. Then I am going to try a 12-volt dc to 120-volt ac inverter. My entire Altermen setup uses 13 watts total and I am almost sure I can discard the two power transformers I use. My 110v voltage section already uses a capacitor-isolated ground so that is taken care of. I should have an answer in a couple of weeks. Darn I need to finish the build, always something more interesting to explore.

A 12 volt tube/valve theremin, did someone say lets go busking?

Lev Sergeyevich would be jealous as we have so much to work with today.


Posted: 4/11/2014 12:15:18 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


I am NOT a tube expert - I dont even qualify as a competent tube hobbyist! - So everything which follows may be wrong, and is based on ideas and memories which could be completely wrong..

1.) I am looking at driving the filament/s with DC, regulating the current rather than the voltage (if the current is constant then the voltage across the filament should be constant if its resistance doesnt change - I am looking at setting the current to provide the correct voltage when the tube is hot, this will limit the start-up current when the tube is cold.. I also intend to ramp the current up at switch-on, and slowly fade it down at turn-off, as I believe this should make the tube life longer)

Regulating the current makes everything a lot easier - An LM317 looks ideal for the job.. And this scheme gives protection without any extra circuitry.

2.) I have a niggling worry about driving filaments with DC.. I think I remember tales of 'migration' in filaments that were DC driven (elements in the filament migrating to the +Ve or -Ve side, weakening the wire, and causing early failure) but cannot find any reference to this, so its probably a myth (although I can see that it may not be).

3.) The filament has an 'averaging' function which means that noise / AC on it shouldnt be that serious - I dont like having audio frequency AC (50/60 Hz and harmonics) anywhere near any signal path, so getting rid of AC filament drive suits me fine.. But technically, there shouldnt be any significant "transfer" of filament AC to the signal path - it should just come down to the filament temperature, and I cant see this fluctuating at 50/60 Hz. I suppose (depending on how configured) one could get coupling from the filament to Cathode.. Whatever .. I will use DC.. But I mention this as a preface to the next....

4.) As I see it, the primary source of supply induced 'noise' is more likely to be on the Anode - Cathode circuit (and of course anything coupled to the grid/s). For this reason, I think that if you use any kind of inverter to provide the HV supply, filtering this supply so that there is no AC on it will be essential.

5.) Re the above (4), I would avoid using a low frequency (50/60 Hz) inverter to generate the HV supply - I would look instead at a high frequency DC-DC step-up converter - Its a LOT easier and cheaper to apply good filtering and linear regulation to the output from such a converter rather than to rectify and filter 50/60 Hz HV AC.

6.) Simple SMPS with a small step-up HF inductor / transformer can be made with a readily available cheap IC.. The only area of some bother is that the HF needs to be "contained" and fully filtered, particularly with theremins.. But IMO this is still a LOT easier than getting rid of mains frequencies. I have never used such a step-up on tubes, but I have used it to provide HV for antenna shielding (right at the most sensitive part of the theremin) and I was able to get rid of all HF on the supply with a passive LRC filter followed by a linear regulator. [ This is a SMPS family that looks ideal for the job - output voltages up to 350V DC can be obtained on some parts, more if you use a transformer.. But best, you can synchronise the switching to some external clock like the reference oscillator, which eliminates any possibility of ghost tones. http://cds.linear.com/docs/en/datasheet/3758Afd.pdf  - I must just say that I didnt use this part, Its SMD - I used a basic unsynchronised SMPS IC and added more complex circuitry to get what I needed]


Unlike you, I am not absolutely convinced that tubes add anything which cant be created more simply using modern components.. I have no interest in cumbersome electronics and HV supplies etc unless these are absolutely required for "good" sound.

My thinking is in the lines that IF tubes are "needed" then they will be needed probably for the mixer and (perhaps) output stages only - the mixer being by far the most likely important role. I cannot see any sonic benefit in replacing oscillators with tube ones, but there might be a thermal / stability benefit.

I also think (and Thierry is the person who got me to look more closely at this) it likely that to get the "classic" mixer sound, the mixer needs to drive a transformer, and the charactaristics of this transformer (and related resonant components) will be critical in 'coloring' the tone.

So my experiments will be on the following lines:

A:) Build a 2 tube mixer / amplifier using 6J2 tubes with a 12V "HT" supply, use small available audio coupling transformers, drive the mixer inputs first from my signal generators (and play with the circuit, messing about with the grid drive levels, capacitances on the transformers, that sort of stuff) and if anything promising shows up, proceed - otherwise it goes into my big box of other time-wasters.

B:) If I get past A, link up the buffered  EW oscillators to the mixer (with whatever scaling / biasing I have determined from the sig-gen experiments) and give it a spin.

C:) IF its really good, build a little buffer board to fit in the EW, with a socket taking the signals (and the volume and perhaps EW audio) to the external tube box (which would be seperately powered).. Add a (probably H11F1) VCA to the box, and get some thereminists to evaluate it before I do anything else.

In truth, had it not been for seeing a tube mixer powered from 6V, I wouldnt even have looked at tubes - I have a load of lovely mixers.. But tubes are probably  the only mixer type I havent explored (Almost in order of preference, but their use is application specific..  Ive built Fet,Linear CMOS, AD633 4QM, MC1496 4QM, BJT, Optical, XOR/Mixed Signal, Diode) So I need to try a tube mixer at least once!


Posted: 11/19/2018 11:43:06 AM

From: Switzerland

Joined: 11/5/2018

Valves and transistors are very different from a practical point of view. Tubes works better with relatively high voltages and low currents, which mean to drive high impedance loads, when transistors works better at relatively low voltages and high currents, which mean to drive low impedance loads. That said, some tubes was optimized to work from a car battery, and if I would use valves at such supply voltages, I would utilize them. But I would prefer to use a DC to DC converter in order to get a high voltage from the 12VDC, which will let me use easily available valves.

An issue with tubes working at very low anode voltage is the influence of that voltage on the mu of the valves: the lower Ua, the lower mu. This is especially true for tetrodes, pentodes and the like because with them, the gain of the tube is fixed by Ug2 and is almost independent of Ua. 

But it is another factor that influence all vacuum tubes: the attraction of the electrons in the space charge by the anode. The heater warm up the cathode which make a space charge around the cathode. If the anode is positive, a current can flow between that space charge and the anode. Higher the anode voltage, higher will be the velocity of the electrons and the current. The maximum current will be limited by the space charge, more exactly by the capacity of the warmed cathode to supply free electrons to the space charge. It is dependant on the cathode material and temperature. Hopefully, most tubes and applications are designed to work below that limit.

It is another limit: at very low anode voltage, the anode will have a very weak influence on the space charge. If you look at the Ia=f(Ua) characteristic of a diode (fundamentals - valve wizard), you will see that at low Ua, the slope of the curve will increase slowly, which imply this is a region that is highly non linear. As Ua increase, the tube will reach a point where the slope of that curve is constant. That region is linear. Then Ua reach another point where the slope will begin to decrease, that until it become equal to zero when it reach Ia max (where Ia will be limited by the cathode temperature as explained before). As all other valves are diodes with additional grids, this apply to all valves in existence. It is why I don't recommend to use ordinary vacuum tubes at low Ua voltages, with the exception if you want the tube to work on such non linear fashion.

You must be logged in to post a reply. Please log in or register for a new account.