Jul 5, 2018

Avalance VCO

Thanks to Look Mum No Computer, a super simple oscillator circuit has been popular lately based on the reverse avalance breakdown effect in normal transistors. I decided to experiment with this idea myself, and in particular see if I can add basic V/Oct control to it.

The circuit


The oscillator core cosists of the timing capacitor C1 and in parallel transistor Q1 connected "in reverse". The base lead of the transistor is not used and should be cut off as short as possible to avoid picking up noise. The circuit operates by charging up C1 until it reaches the reverse breakdown voltage of Q1. When Q1 reaches breakdown it will suddenly allow current to flow through it, quickly discharging C1 again. Then Q1 resumes normal operation and the next cycle of the oscillator begins. Thus it is a saw core, and Q1 assumes the function of both the comparator and the reset switch found in traditional saw core oscillators.

I shifted the topology around a little bit, and referenced the oscillator core to +12V instead of ground. This way a normal ground-referenced transistor can act as a current sink to control the oscillator frequency, much like in traditional saw-core VCOs.

I used the exponential converter found in the Moog transistor ladder filter. It consists of a PNP buffer (Q5) and a NPN transistor (Q4) as exponential converter and current sink. Q4 and Q5 have roughly opposite temperature coefficients, providing a very rudimentary form of temperature compensation. In order to isolate the oscillator core from output loading, I added a darlington buffer to the output (Q2 and Q3). Finally the output is AC coupled to remove the offset voltage.

As others have investigated, many transistors are suitable in principle as avalanche oscillator, but their threshold voltage and other behavior differs. Of the transistors I had at hand, I found the 2N3904 has a threshold voltage around 9V, giving a suitable amplitude close to the 10Vpp modular standard. A list of suggested transistors can be found in Kerry Wong's article, but note that the voltages listed there are 3V higher due to the LED in series with the transistor.

The capacitor value sets the frequency range, I went with a smallish value to reduce the current needed from the exponential converter. Smaller values don't seem to work at all, which is a pity since reducing the current further would help with tracking and stability. A larger capacitor such as 1uF could be used for LFO-range operation (for electrolytic caps the + lead should connect to +12V).

I made a stripboard layout that fits neatly behind a 4HP eurorack panel, and a single drawing with all info you need to make your own:


You can download a PDF version here, and the front panel layout here. I made the front panel out of 2mm plywood, and the final result looks like this:



Tracking


So, you ask, how well does it track 1V/Oct? Well, for a synth VCO not very well at all, but for such a simple circuit I think it's not bad. I got it down to the following tracking error:


While the low and high end are horribly off, there is a range between about 30 Hz and 500 Hz where one could pull of a half-decent bassline or some drifting drone tunes. Drift, indeed, is probably a more serious concern since temperature stability is very limited. At higher frequencies self-heating (presumably of the expo converter Q4) is the biggest issue, causing the frequency to drift up when tuned to high frequencies.

40 comments:

  1. Just modify the expo converter to use an lm3046 chip and it works better.

    ReplyDelete
    Replies
    1. Has anyone tried doing it this way yet? How would that alter the schematic?

      Delete
  2. Very very interesting, thank you! I knew only about negistor oscillators and built some, time ago.
    I don't look for 1V/oct tracking, but, what's it's max frequency range?
    And what CV range are you using? 5Vpp? 10Vpp?
    I'd use 10Vpp CV so I'd like to know which freq range is atainable with it.

    ReplyDelete
    Replies
    1. As you can see from the figure, the range I tested is about 10 Hz to 5 kHz, which is roughly 9 octaves so 9 Vpp since it is tuned to 1 V/Oct. This is more or less the useable range, I think you can go to 10kHz but not much further. Of course you can move the range lower/higher by using a bigger/smaller capacitor for C1.

      Delete
  3. Worth reading article. Well done ! Kind regards R.

    ReplyDelete
  4. I've no problem getting the LMNC format reverse avalanche working but your superior CV in handling appealed to me. Unfortunately I've literally been unable to get your circuit working on the breadboard. I've actually managed to burn out two potentiometers trying, and I simply haven't been able to get it to work. The darlington pair works (tested that individually) beautifully, but I can't get the oscillator core or the expo converter / CV/pitch control to work.

    I've no idea what I'm doing wrong as it seems pretty straight forward.

    ReplyDelete
    Replies
    1. I finally got it oscillating (I couldn't get it to work with non polarized caps, but I could with an electrolytic and LED like the other version). Now the problem seems to be a dead zone from about halfway to 0 on the freq potentiometer where it turns off entirely.

      Delete
    2. I can eliminate the dead zone by dropping the 220k pull down to -12v.

      I'm curious if maybe this is because my electrolytic+diode combination.

      I couldn't get the ceramic/nonpolarized to work anyways so that was weird

      Delete
    3. Have you tried different transistors for Q1? The avalanche voltage may differ per transistor quite a lot, and at lower currents (lower frequency settings) the expo transistor Q4 takes up more voltage, which might cause the dead zone. Possibly you can also adjust the range with the trimmers to get rid of such a dead zone.

      Did you try the electrolytic capcitor without LED? I think leaving out the LED should make it more likely to oscillate, independent of what kind of capacictor you use.

      Delete
  5. Would using a +/- 18V power supply be a problem? if so how would the schematic change?

    ReplyDelete
    Replies
    1. Should be fine, maybe change R8 to 220k to keep the tuning range similar.

      Delete
  6. What kind of capacitor is the 10nF? Can't seem to find it anywhere.

    ReplyDelete
    Replies
    1. Any film capacitor will do fine.

      Delete
  7. This is so cool! Have you tried the circuit out with the SS9018 transistors that Look Mum No Computer used for the later revisions of his circuit? (Since they oscillate at much lower voltages, 8 or 9 volts.)

    ReplyDelete
  8. I am a retired EE circuit designer and have been using transistors since 1954. Generally, the emitter avalanche area of operation was avoided as it would destroy the junction transistors of the time. The problem was related to the uniformaty of the junction. If the avalanche current was concentrated in a tiny area of the junction, the temperature there could get high enough to melt the spot, damaging the transistor. Apparently, advances in transistor fabrication have resulted in more uniform junctions.

    ReplyDelete
    Replies
    1. I have tested only a few transistors, but Sam from Look Mum No Computer has built a system with 1000 reverse avalanche oscillators: https://www.youtube.com/watch?v=c3wk9WWTfNs
      He never mentioned any failed transistors, so indeed it seems fabrication has come a long way!

      Delete
  9. Hi is there a recording of this working? That fact that it would be really wonky and off is far more interesting than just another vco. Also just out of interest, if you were to have this vco x6 for polyphony how would you implement the expo converter? do you have to copy the circuit x6?

    ReplyDelete
  10. Hello I have a problem with the module, when I use the fine potentiometer, about 300hz towards 0hz it turns off, I can only get medium and treble sounds, what problem can it be?

    ReplyDelete
    Replies
    1. Hi! I suspect you happen to have a transistor that is just on the edge of having too large breakdown voltage for this circuit at 12V. The breakdown voltage is not a normally used feature of transistors, so the manufacturers don't specify or guarantee it to be any specific value (they only give a minimum value).

      To try this, easiest would be to power it from a bit larger voltage, for example 15V. If my suspicion is correct, it would work down to lower frequency at larger supply voltages.

      Another way is to try a different transistor for Q1, even one from the same batch could have different breakdown voltage if you are lucky.

      Delete
    2. I will try, also in the lookmumnocomputer schematic, I used other transistors like the BC337, could I replace it with one of these? I only have a 12v power supply. thank you very much for your answer.

      Delete
    3. Yes, you can try BC337. I think the pinout will be reverse of the 3904, so double check which way around it should be :)

      Delete
    4. Finally works! It was the transistor Q1 thank you very much!

      Delete
  11. I've tried to get this work, but q5 fries every time. Any advice where I should start troubleshooting?

    ReplyDelete
    Replies
    1. Also, can't get any sound out of it.

      Delete
    2. Did you follow the stripboard layout or build from schematic? Perhaps the power is connected directly across Q5 by accident, is R2 installed correctly? You could try removing Q4 to see if that stops Q5 frim burning out.

      Delete
  12. I followed the veroboard layout and tried to check everything with schematics. And also checked every connection with DMM. This is my second attempt with this, and still no use, so... I think must have read the layout wrong and put something in wrong place...

    But I'll try what you suggested and take Q4 out and see what happens then.


    Thanks.

    ReplyDelete
  13. I suspect you could fix the temperature drift the same way that Moritz Klein did for his Schmitt-trigger-inverter VCO: https://www.youtube.com/watch?v=dd1dws6pSNo I wonder if this would also help with the audio range?

    ReplyDelete
  14. Is a Film capacitor necessary?

    ReplyDelete
    Replies
    1. Hi! A ceramic capacitor will also work, but some types are have big changes with temperature. Best ceramic types are C0G or NP0, they are very stable.

      Delete
    2. Got it! Thanks a bunch!

      Delete
  15. This comment has been removed by the author.

    ReplyDelete
  16. Why the fuck does Q1 have the arrow pointing the other way if it's the same transistor as the rest of them? And why the fuck does it say '1', '2' if '2' is the base which you supposedly clipped off? Makes no sense. I'm retarded or you're retarded

    ReplyDelete
    Replies
    1. Dear For Fuck Sake,

      I feel your frustration! You are totally right that it should say '1' and '3' for Q1. The ´1´ and the arrow are in the right place though, they show which way around the transistor should be connected. So the emitter (arrow) goes to +12V.

      It's very unusual to connect a transistor this way, as I mention in the text it's "in reverse" which gives the avalanche breakdown effect.

      Happy building, hopefully it works out!

      Delete
  17. Im a bit confused, with the whole cut off the base thing, it seems if its unconnected it simply makes no difference than removed, and furthermore, if connected to a voltage source in the otherwise layed out oscillator core, in my.. experience... it acts as a CV in for the pitch...that im pretty sure can be tuned via resistance/appropriate dividers from said voltage source to pretty reasonable volt per octave all in itself...

    ReplyDelete
    Replies
    1. Hi! Indeed this circuit is confusing.

      The reason to cut off the lead is that it can otherwise pick up interference, act as a small antenna.

      I'm not sure if the reversed biased transistor could work as an exponential converter at the same time. I suppose in principle it could, due to the partial symmetry of the NPN transistor it should work as a poor transistor even in reverse bias. But one would have to control the base-collector votlage, so the collector would have to be at 0V for that to work without complicated biasing. Anyway, a nice idea!

      Delete
  18. perhaps in reverse avalanche the exponential behavior before fully active still exsists concerning the reversed current,

    ReplyDelete
  19. I built this vco point to point with stuff I had laying around. No breadboard (no board of any kind), and resistors tied together to get values close enough to the schematic. It works just fine with a baby 8!

    ReplyDelete
  20. Since the expo/buffer is like the one René uses in his 4069 VCO, you can try thermal glue and 10k thermistors as in his circuit...

    ReplyDelete
  21. BTW, surprised you get a saw from the collector since the usual negistor oscillator outputs a thin pulse there and the saw from the emitter

    ReplyDelete