Switching scheme with muting and LEDs

[frusciante89]

Hi guys, I have a problem with designing a channel switch. I usually go for relays with flywheel diodes, with the relay coil ground connected to the switch. However, I implemented this solution in a 3 channel amp and I can not tolerate the popping noise. I looked around the forum and seems that the best way would be to use a muting circuit, for example through an optocoupler. The requirements are:
1) the amps has three channels with separate gains, masters and tonestacks.
2) The are 5 relays. R1, R3, R4 and R5 are tied together, and R2 is separate.
3) when all relays are off, the mode is "Crunch". When R1, R3, R4 and R5 are on, "Lead" mode is on. When R2 is on, "Clean" mode is on.

So far, this is triggered with 3 position switch from the front panel of the amp. It works, but it pops. My idea is to keep the front panel switch, add muting, add LEDs (currently not present), and add a footswitch, possibly using a 5 pin DIN port. My current footswitch works with a stereo jack but has only two footswitches, and to change from CRUNCH to LEAD I have to press one of the buttons twice. I'd ideally like to have three separate buttons for each channel.

Do you guys have any idea on the circuit that is most suitable for this application? Also, is there any amp that currently resembles my architecture? I've attached a sample circuit that I drew... What do you think?
Thanks for the help.

[catalin gramada]

I used with success for switching chanell aplication rotary switches with what is called "make before break" contact type. That means the wiper don.t jump from one position to another breaking the previous circuit then connect the second, but keep it the first circuit connected till the second is connected then break. In this case the popping noise is avoided because it not jump from one voltage potential reference to another but keep it the both channel in parallel for a fraction during the transition . This type of contact switch is pretty common version for any type of switch, I.m pretty sure you.ll find the relays which use it, just Google it

[nickb]

I don't think that will work as shown. Perhaps you intended this:



That said the optos seem like a hard way to do it. Some PNP transistors would seem a cheaper / easier solution.

The noise most likely comes from the circuits that are being switch due to different voltages on the contacts. Can you show us that bit? Often some high value resistors in just the right place will effect a cure.

Mesa often uses a muting pulse generated by their relay control circuit which turn on signal shunting JFETs.

[Enzo]

You get channel switching pops? You didn't show us the most important part - the signal path. Main reason for pops is unterminated caps in the signal path.

You seem to have assumed that the popping is coming from the switching control circuits rather than the signal path.

[frusciante89]

Thanks for the answers guys. I've always built single channel tube amps so far, so this is new territory for me. Forgive me for my noobiness in the area, but this is quite exciting to learn Anyways, this is the signal path:
.
.
Relay 3, 4, 5 are switched together, while Relay 2 takes care of the Clean channel and is separate
.

[Enzo]

V3b is a cathode follower, so I guess it has maybe 100v on the cathode? That cathode is switched back and forth to the two sets of tone stacks. In either case, that 100v gets slammed into empty caps in the stacks. A 1meg resistor across each side of the relay would keep the caps charged.

[nickb]

I expect your biggest pop is coming from the tonestack switching. Method (a) Put a big resistor, say 10meg from pole to each contact of RL5 that drives the tone stack. This will keep the tone caps charged, or (b) add a 100nf cap in the path the the pole of that same relay, or best (c) just loose that half of the relay and have the two stacks connected at all times. The cathode follower has a low impedance and will quite happily drive both stacks.

You could also leave out the other half of RL5 that switches the tone stack outputs to the volume controls. That's not doing much either. Just wire the volume pots to their respective stack along with the 220k resistor ( which should now be 470k).

Hard to tell if RL3 is a problem since the drive to it is missing from the image.

[frusciante89]

I expect your biggest pop is coming from the tonestack switching. Method (a) Put a big resistor, say 10meg from pole to each contact of RL5 that drives the tone stack. This will keep the tone caps charged, or (b) add a 100nf cap in the path the the pole of that same relay, or best (c) just loose that half of the relay and have the two stacks connected at all times. The cathode follower has a low impedance and will quite happily drive both stacks.

You could also leave out the other half of RL5 that switches the tone stack outputs to the volume controls. That's not doing much either. Just wire the volume pots to their respective stack along with the 220k resistor ( which should now be 470k).

Hard to tell if RL3 is a problem since the drive to it is missing from the image.

R3 is in the image, I'm using just one half of the OMRON G5V2, and its purpose is to change between the crunch and lead gains.

So correct me if i'm wrong, you're saying that I could drop Relay 5 entirely? And just swap the 220k with a 470k? Do you see any issues of having both stacks on at all times?

[Enzo]

No, it is done that way often.

[J M Fahey]

Please label all parts shown in the schematic, currently some are, some are not.
Also with multi section relays: you have *two* labelled "relay 5 crunch lead" , you can call them A and B respectively.

[nickb]

R3 is in the image, I'm using just one half of the OMRON G5V2, and its purpose is to change between the crunch and lead gains.

Yes I can see RL3 but what's driving it is not shown. I suggest you add a 47nF between the pole of RL3 and the grid leak resistor to minimize and change in the grid leak voltage due to different wiper positions.

So correct me if i'm wrong, you're saying that I could drop Relay 5 entirely? And just swap the 220k with a 470k? Do you see any issues of having both stacks on at all times?

Correct. I have done this myself. Works fine. You do understand that you'll need two 470k resistors?

[frusciante89]

Yes I can see RL3 but what's driving it is not shown. I suggest you add a 47nF between the pole of RL3 and the grid leak resistor to minimize and change in the grid leak voltage due to different wiper positions.

So something like this?

Correct. I have done this myself. Works fine. You do understand that you'll need two 470k resistors?

I'm not sure where to put the two 470k. Can you explain it? And why? Also, won't the signal passing through the active channel be affected by the setting on the other (off) channel?

Thank you all again

[nickb]

Nope. More this this one that is very similar to yours. Still guessing as to what is driving RL3 (is it a secret? ):



Here's a suggestion of the control cct w/o any opto isolators, Just watch the current in the crunch LED is not so high that it hold the relays in:

[frusciante89]

Nope. More this this one that is very similar to yours. Still guessing as to what is driving RL3 (is it a secret? ):

[ATTACH=CONFIG]51274[/ATTACH]

Here's a suggestion of the control cct w/o any opto isolators, Just watch the current in the crunch LED is not so high that it hold the relays in:

[ATTACH=CONFIG]51275[/ATTACH]

Here it is... I just omitted the power section which is the usual. The parts that i've modified are in shaded red. They should be correct now, do you guys agree? This schem has the relay switching that I'm using right now (On-Off-On switch, no led, stereo jack for footswitch).

I also modified that switching with nickb suggestions, and it looks like this:


Any thoughts?

[Gregg]

I expect your biggest pop is coming from the tonestack switching.

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 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.