Switching scheme with muting and LEDs

[nickb]

The biggest pop usually comes from relays connected to grids, especially in high gain circuits. This is almost unavoidable that's way they all have mute circuits.

The biggest noise source depends on the ratio of the noise voltage to the signal. I see it like this: In the case of the original tone stack arrangement the worst case noise would be about 100V. For RL3, the signal is about 1/50th of that at the tone stack. The worst case voltage difference comes from the small grid current drop across the 1Meg, may 0.25V i.e. 12.5V at the tone stack therefore the tone stack would seem to be the worst offender. You could argue that once the tone stack drive relay has operated once that the caps are now charged to the same on both sides and that would be true if we ignore leakage. But it's not the worst case.

The other thing that looks very odd is using PNP transistors for switching. I don't recall seeing PNP switching any amp and never used them.

Well now you have

This was the simplest arrangement that I could think of. The attraction of the PNP's is they allow you to keep the one side of the switch at 0V yet still have enough of a voltage swing to allow you to have an LED in the path to the switch. In other words you can put the LED in the footswitch with just a single wire and and switch contact. I see that in the lastest dwg a front panel switch has appeared. Even with that you can still have an LED in the footswitch. The two panel and footswitch LEDS would be in series and so work together, just what you'd want.

[frusciante89]

I put some labels on the components. I was studying how this should work and it's still not clear to me how the switching exactly works: for example, when switch S1 is in position 2, the LED Crunch should be on and all the relays off. But how?

[nickb]

Uh oh! I suspect the reason you're have a problem is that I had the logic wrong Let me try that again



PS: I can think of two ways to do this without any transistors at all. One still allows you to put two of the LEDS in the footswitch but (hint) needs 5V relays, the other uses 12V relays but you lose that feature. See what you can come up with.

[frusciante89]

Thank you nickb, this makes more sense now. Actually, I really don't mind using the transistors! I'm designing my own PCB board and I have quite a lot of space left.
So, right now:
1) I removed Relay 5 and tied together the tonestacks, which are now switched on and off by relay 4 as shown in post #14
2) I added the logic you suggested (thanks again) to turn on and off the three channels, with respective LEDs that I'll mount to the front panel
3) I added the 47nf cap between relay 3 and the grid leak resistor of V3A

Do you think this will be enough to tame switching pop? Are the optocouplers unnecessary at this point? (extra PCB space and extra components are not an issue, my goal is to minimize pop as far as I can without affecting the tone).

Thanks again guys, I'm learning quite a lot from this

[nickb]

47nF not pF

[frusciante89]

sorry, fat fingers Also, that schematic for the switching works by connecting the relay to the 12V supply while keeping the ground always connected. Is there a way to wire it the opposite way? I mean, currently, I have the 12V supply always on, and then I switch on and off the ground according to the channel I'm selecting. It's not a big deal, but I've already wired the pcb that way (see attached)...

[nickb]

I'm not going to say no but I think you are capable of doing this on your own Why not give it a go? You'll get so much more out of it.

I only offered the circuit as it seemed a little costly/ unusual to use opto-isolators where nothing was isolated. But it wasn't wrong.

[frusciante89]

I'm not going to say no but I think you are capable of doing this on your own Why not give it a go? You'll get so much more out of it.

I only offered the circuit as it seemed a little costly/ unusual to use opto-isolators where nothing was isolated. But it wasn't wrong.

Does this look about right?

[nickb]

There is a snag. None of the transistors will ever turn off. Why? Hint: Look at the minimum base to emitter voltage.

[frusciante89]

mmm... should I move the resistors between the base and the ground? (also, I forgot to indicate the 12V supply rail. I've updated the diagram)

[nickb]

The problem was the base to emitter voltage never got low enough (<0.6V) to ever allow the transistors to turn off. I've added a couple of dividers to allow that to happen for T1 and T2, and added a diode on T3 for the same reason. The new resistors also give some protection to T1 & T2 from the big bad outside world. Incidentally, that protection was something that your original opto isolator scheme gave and that was a nice feature.

[Gregg]

I told you to ditch the PNP stuff

[frusciante89]

The problem was the base to emitter voltage never got low enough (<0.6V) to ever allow the transistors to turn off. I've added a couple of dividers to allow that to happen for T1 and T2, and added a diode on T3 for the same reason. The new resistors also give some protection to T1 & T2 from the big bad outside world. Incidentally, that protection was something that your original opto isolator scheme gave and that was a nice feature.
[ATTACH=CONFIG]51313[/ATTACH]

Thanks a lot nick! what NPN should I use? Is there a particular number you suggest? For the diode, is a fast UF4007 good?

[Gregg]

I'm usually using a common type like BC546B or BC550C. No need for fast diodes.
Also check out Mesa amps' switching for more ideas and 'how to'.

[frusciante89]

The problem was the base to emitter voltage never got low enough (<0.6V) to ever allow the transistors to turn off. I've added a couple of dividers to allow that to happen for T1 and T2, and added a diode on T3 for the same reason. The new resistors also give some protection to T1 & T2 from the big bad outside world. Incidentally, that protection was something that your original opto isolator scheme gave and that was a nice feature.
[ATTACH=CONFIG]51313[/ATTACH]

Hey guys. Update on the current situation.
I designed a PCB board with the switching scheme, manufactured it and mounted it on my new amplifier. However, there's is a problem: I have a 3 position switch connected to the front of the amp, which has the middle lug connected to ground, and right and left lugs connected to the "clean" and "lead" pin of the PCB, which correspond to the "clean" and "lead" legs of the scheme proposed by nickb (thanks again btw). The issue is that the "crunch" LED remains always on: if I switch to the clean or to the lead channels, the respective leds turn on correctly, but the crunch led doesn't turn off as it is supposed to, and I get two leds on at the same time (ugggh). Also, the amp remains constantly in the crunch mode, no matter what position is selected on the switch, so I'm assuming the relays are not activated. I've been trying to the debug the circuit but everything seems correct to me. Do you have any clue of what might be happening?

I'm attaching a pic of the actual PCB and a zoomed pic of the switching part