Now I have got the safety valve!! Sounds great! …But the bypass is still not working. when set on bypass (input connected directly to output) I have no sound anymore! Only if I disconnect the opamp from the output I have the bypass (but then obviously no more distortion). I don’t understand that!
1 Like
Your bypass should disconnect the output from the op amp and the only signal it should see in that switch position is the input.
In the Description it says it only disconnects the input from the Tube… will check tomorrow, need to sleep now 
Just my opinion, but ideally nothing else touches either input or output when in bypass.
1 Like
I added one thing to the bypass you may want to consider adding as well. I used a double switch which bypasses the input signal to the output AND at the same time switches off the current to the heating filaments of the tube. In this way in bypass mode the load on the power supply goes down quite a bit. This is meant specifically for those cases I’m not using the safety valve(s) (I made 2) at all.
1 Like
Good call on the valve heaters. Unless you’re going to be switching that circuit in and out a lot during a session it’s better to leave the valves cool.
Unless you’re going to be switching that circuit in and out a lot
I have considered that, but the I wouldn’t want to do that via a hand operated switch then but via some CV or gate toggle control.
1 Like
I now did this:
So, when I set the switch to bypass the input goes directly to output of the Module and also the out of the opamp is no longer connected to the output of the Module. If I can get a switch with 3 lines (9 pins) I can still add a LED!
I also added a switch between the input of the second stage of the valve (pin 7) and the input, which gives me a lot more gain in an instance, so I still have the ‘overkill’ setting from the original valve distorting vca! 
(or is it the squelch? that would match the valve vca schematic, but the sound is just more gain and no ‘quirky’ sound… When I short the capacitor between the two valve stages as in the original valve vca schematic, there is just no sound at all anymore, this is what Jos has already pointed out! Still wondering though.)
Might be a lot to ask, but could anyone do an updated Breadboard for the VCA. We are having a lot of troubles with this one the Facebook group.
2 Likes
Hi,
I was wondering if it would be possible to replace the feedback ( squelsh) switch with a potentiometer ( or even a vactrol) to gain more control over the feedback? Any thoughts?
If I read the schematic correctly, what the squelch switch does is bridging point squelch1 and squelch2, effectively feeding the 1st triode’s input to the grid op U1A but also to its own output. This may lead to feedback behaviour which could be sonically interesting. And maybe if indeed you use a potentiometer, you can influence its behaviour. BUT, there is a problem with the diagram as it is. There should be an input capacitor connected to ping 3 of SW?B, so that the input is not dependent on any DC-value from the circuitry connected to input J2 (plus connected to input, minus to the side of U1B’s grid). The same goes for the output, there should be a capacitor between J4 and pin 6 of U2 (plus connected to pin 6, minus on J4 side) so that it does not output any DC value to the following module. 47muF caps should be ok.
Yep, the overkill switch does one thing: it kills the sound. U1A will not function properly anymore because the grid gets the DC value of U1B’s cathode, and then it will stop conducting any current. By killing the sound I mean that it will mute the circuit because U2 will not get a signal anymore. This is probably not what the switch’s initial intention was.
I just noticed that there is also no capacitor at the input of the safety valve (not vca). Is this intentional?
I’d recommend having a look at the matsumin design. This original design has them on input and output of the tubes, as it should be.
Would a Bc547 work in llace of the 107? Didn’t habe 107. Read 547 is equivalent. My signal seens to stop there.
They have similar electrical characteristics, but a single-transistor VCA is pretty crappy as VCAs go, and you want the right kind of crappiness here (note that it’s even mounted backwards in this circuit to do “a better/gnarlier job”). But BC547s are cheap jellybean stuff, so why not try it? Worst thing that can happen is that you don’t like how it sounds.
Yup, I did try it. No sound. Tube is warming up. Nothing blows up. Checked around with an audio probe The signal seems to stop in around the 547. I’ll have to track down a 107.
here’s the schemetic from GMSN for the Safety Valve (a little bit different : with TL072)
that’s with that I built mine :
1 Like
What strikes me is that R6 is 200 kOhm while in the original Matsumin diagram this is 100k. Would 200k result in a higher output voltage swing?