Educate me on relays?

[JamesHealey]

Hey guys I've got a custom project for a friend of mine and I've made well over 20 custom amps for people and many many more for Hiwatt some time ago.

But I've never made anything channel switching, anyone able to help with how this is achieved using relays.

Thanks!

[Blind Lemon]

Ted Weber has the switches already to go on a little circuit board, the board is only about 2"X2". The one I have here operates of the 6V filament voltage. They are not hard at all to figure out.

BL

[David Root]

I had problems understanding how they interface with the footswitch until I realized the switch just grounds the signal out.

[tubeswell]

Take a look at PVDB or C30 schematics, they have a simple kind of relay switching.

If you want a bit fancier, you could use optoisolators, in which case you might want to look at the Soldano SLO100

[jaysg]

It appears that the best production/boutique amps provide a separate winding or transformer.

[Cliff Schecht]

I wouldn't see an extra auxiliary winding being necessary unless you couldn't get a decently stable DC voltage off of the taps already provided. I guess it's a nicety more than a necessity.

[ampdoc1]

I had manufactured and sold quite a number of relay board over the last few years, but sold the last of 100 a few months ago. A member here has some for sale that seem to me much the same, but a little different in size. These feature a detachable PS and 3 detachable DPDT relays. No reason to buy special trannys,...I've connected the PS to the (generally unused) 5VAC winding that is on most transformers. This always worked well for me as the 6V relays I used would work between 4 and 9 volts.

ampdoc1

[paulster]

James

Take a look at a Marshall Silver Jubilee schematic. That uses a fairly simple but quite neat way of operating the relay using a regular mono 1/4" footswitch and also supports the LED in it as well as on the front panel. You can use a stereo 1/4" plug with a two-button footswitch with 2 LEDs to switch two relays too using the shield as common and the tip and ring for the two switches.

Using a Cliff jack also means that you can have front panel switches which are overridden when the footswitch is plugged in.

That's my preferred way of doing the actual relay coil energising for relatively simple builds, i.e. with only a couple of relays required.

Now, the channel switching using the relays can be a little more complicated since that's where you're prone to pops, so you might have to add something like 10M resistors to ensure that the blocking caps don't end up with a DC potential on them and cause pops.

Paul

[DonMoose]

Now, the channel switching using the relays can be a little more complicated since that's where you're prone to pops, so you might have to add something like 10M resistors to ensure that the blocking caps don't end up with a DC potential on them and cause pops.

The rule here is that any switching contact must have a DC path to GND and be AC coupled (thru a cap, that is) to any non-GND DC level. 10M is a good value that doesn't really add a noticeable load. The AC coupling cap is usually already there between your switching point and the previous stage (a tone stack is a little more complex, but if you follow the paths, you find it's AC coupled).

Another thing you really want is a diode or cap across the coil (Cathode to +) to provide a path that keeps the coil flyback current from inducing pops on your signal wires.

Hope this helps!

[Cliff Schecht]

The nice thing about cadmium-sulfide based channel switching is the resistive elements can't change very quickly. The typical response time is in the 1-100mS range for most CdS cells which tends to naturally suppress any pops that occur from channel switching.

Some sort of slew limiting circuit on the relay control voltage would kill the pops, but this requires some extra circuitry that not everyone wants to shove in their amp. The 10M pulldown resistor I think works decently to prevent the coupling caps from floating around DC but this doesn't prevent pops that occur when the input to a stage is initially at 0V (grounded perhaps) and a 50V input signal is suddenly applied. Without any slew limiting, you are looking at rise times below the microsecond level. To the stage after the relay, this looks like you are momentarily applying a unit step function which effectively gives you that nice pop. The AC coupling and ever-changing input signal prevent the input from staying high but that initial pop that naturally occurs when a high signal is suddenly applied is an annoying side effect of relay-based switches.

Has anybody played around with time-delayed relays or circuits that could slow down the speed of the relay control voltage?