FSH-1 random speed change

[Marcvv]

Hi,

I Have build a FSH-1 with a tonepad PCB.

Works great! (Thanks FP)

I was wondering if it would be possible to create a random speed control of the sample and hold side.

Would this even be possible by using the CV of the S&H CV?

Thanks, Marc

[Ripthorn]

I asked the same question in a thread titled (roughly) "FSH1 has strong ticking" and the answer was "yes it is possible, but requires lots of extra components and work".

[MarcoMike]

perhaps a simple oscillating speed would sound like random associated with S/H... that should be easier...

[frequencycentral]

Well, first of all you would have to make the LFO into a voltage controlled LFO (VCLFO) ie use a different LFO, possibly a LM13700 based design. Then you would need an second S/H circuit being fed white noise and running at a high speed from a second LFO. The CV output of the second S/H circuit would feed into the CV input of the VCLFO. You could even have a slew control to allow the speed changes to slur. That seems like a lot of extra circuitry to me. Easily done on a modular synth though.

[soggybag]

I think the easiest solution would be a micro controller. You could program a randomized LFO onto a 8 pin DIP PIC or AVR. IO suspect that an LFO made with a microcontroller would avoid ticking problems.

[Marcvv]

Hi all,

Thanks for responding.

Well, first of all you would have to make the LFO into a voltage controlled LFO (VCLFO) ie use a different LFO, possibly a LM13700 based design. Then you would need an second S/H circuit being fed white noise and running at a high speed from a second LFO. The CV output of the second S/H circuit would feed into the CV input of the VCLFO. You could even have a slew control to allow the speed changes to slur. That seems like a lot of extra circuitry to me. Easily done on a modular synth though.

Is this not what is already available in the Maestro FSH-1?

Would it be possible to get a led or something like that going on the CV that is in the circuit and drive an LDR from there that is connected to the speed pot?
Or is this a too simple thought??

Marc

[frequencycentral]

Hi all,

Thanks for responding.

Well, first of all you would have to make the LFO into a voltage controlled LFO (VCLFO) ie use a different LFO, possibly a LM13700 based design. Then you would need an second S/H circuit being fed white noise and running at a high speed from a second LFO. The CV output of the second S/H circuit would feed into the CV input of the VCLFO. You could even have a slew control to allow the speed changes to slur. That seems like a lot of extra circuitry to me. Easily done on a modular synth though.

Is this not what is already available in the Maestro FSH-1?

Would it be possible to get a led or something like that going on the CV that is in the circuit and drive an LDR from there that is connected to the speed pot?
Or is this a too simple thought??

Marc

Hey - you might be onto something there - perhaps an LDR in parallel with the speed pot would do the trick, coupled to an LED derived from the S/H. You would have to regulate the brightnes on the LED with a resistor, and use a pot to dial in the amount of randomness. Sounds interesting - and easily do-able.

[Marcvv]

Hey - you might be onto something there - perhaps an LDR in parallel with the speed pot would do the trick, coupled to an LED derived from the S/H. You would have to regulate the brightnes on the LED with a resistor, and use a pot to dial in the amount of randomness. Sounds interesting - and easily do-able.
[/quote]

Hi Rick,

Thanks for reacting.

I measured the voltage swing at the CV point on the tonepad schematic and layout. It gives voltages in between 7-8,5 volt. When I use that point to drive a led it is burning. With a trimpot I can control the brightness but as far as I can see it does not pulse.

My guess is that the voltage difference is not big enough to make the led pulse. Or am I missing something.

Marc

[frequencycentral]

I would have expected the voltage swings from the s/h to be quite a bit more than 1.5 volts. Exactly where on the Tonepad schematic are you deriving the CV from? I would take it from the conjuction of Q6 and the 22K resistor. You will need to regulate the CV by having a resistor between it and the LED (maybe start with a 10K trimmer and replace it with a fixed resistor later). The use a pot (to ground) after the LED as your 'randomness' control - maybe 47K or 100K. Have a look at my Vintage Vibe schematic for a graphic representation of what I mean.

Maybe also have a play about with the trimpot after IC3a which defines how much noise is presented to the s/h.

The LED wouldn't have to pulse, it should change subtlely in brightness though, maybe an LDR would detect this better than the naked eye.


EDIT: It occurs to me that maybe the amount of noise seen by the s/h which is desirable for the filter might be different to the amount needed to be desirable to the LED. If so, you may have to have two s/h circuit, both driven by the same LFO and Noise source, one fro the filter, one for the LED. That would just mean duplicating the 10K trimpot, Q5 (and it's 1M resistor), Q6 and the .05 cap.

[Marcvv]

I first took the signal from the Q3 2M7 junction and that was not working, or at least not visible (you were right that does not mean the LDR would not see it.)

Ok, I tried to run the led from the Q6 and the 22K resistor and that works! It is even visible.
But when I run the led I loose the signal controlling the filter. My guess is that I indeed need to create that second s&h circuit as you suggested already Rick.

I will have to get some parts now to try this.


Marc

[frequencycentral]

Ok, I tried to run the led from the Q6 and the 22K resistor and that works! It is even visible.
But when I run the led I loose the signal controlling the filter. My guess is that I indeed need to create that second s&h circuit as you suggested already Rick.

Maybe a unity gain buffer between Q6/22K and the LED would do the trick.

[Marcvv]

Yes, that is what I was thinking as well.

I must have a Jfet buffer around so I will give it a try.

Marc

[Marcvv]

Ok, it has been a while that I had a chance to work on this.

EDIT: It occurs to me that maybe the amount of noise seen by the s/h which is desirable for the filter might be different to the amount needed to be desirable to the LED. If so, you may have to have two s/h circuit, both driven by the same LFO and Noise source, one fro the filter, one for the LED. That would just mean duplicating the 10K trimpot, Q5 (and it's 1M resistor), Q6 and the .05 cap.

I did build the extra s/h circuit driven from the the same LFO and Noise source. It works to control the LED (although I do not know yet if it really gives a random  chance as that is hard to see.
Now I have the LED reacting I get a big ticking noise from the LFO that I do not have without that.

So I am slowly getting closer. Any suggestions?

I have also tried to work with just a buffer but I could not get the LED two work??

Thanks, Marc