DIY amp head switcher

[Yoda]

Hello everybody, I have a customer asking me to make him an amp switcher box to switch between four different amplifiers and one speaker cabinet. I did a search for this and not much came up, how feasible is this?

My initial sketch is to have four DPDT relays, selected via 4-position rotary switch, that will switch one pole for the amplifier speaker output from the cabinet out to a dummy load and simultaneously switch the other pole from allowing the guitar input to the amplifier to grounding it to mute the amp.

I imagine a break-before-make switch should be used to keep from connecting any amplifier outputs to each other. Not sure what to expect as far as noise/popping sounds but maybe 1M+ resistors between all the speaker outputs and dummy loads?

Still brainstorming if this is doable but that’s all I have so far, thanks in advance for any help.

[GAStan]

My concern is how to do this without creating a huge ground loop. I'm thinking isolated jacks and 4PDT relays to switch grounds with signals.

To me your reasoning seems good.

[xtian]

And dummy loads for each unused amp?

[Yoda]

And dummy loads for each unused amp?

Correct, one of the poles of the relays will be used to switch the amplifier outputs from speaker cabinet to dummy load when not in use.

[Yoda]

I take it the negative speaker output should not be a common ground between all amplifiers? To avoid ground loops.

[GAStan]

I take it the negative speaker output should not be a common ground between all amplifiers? To avoid ground loops.

That is what I am thinking. Mostly I wanted to bring it into the conversation to see if it is a concern and if so, how to best handle it. My instinct is to start with what I suggested but I'm curious to see what the more experienced techs here have to say.

[Stevem]

For a one of deal he’s far and away better with one of these!

[R.G.]

A four-amp head switcher is a fairly complex device, as you're finding. It's further complicated by there being two types of amps (tube and solid state) that have opposite needs in terms of loading. Tube amps deal very well with their outputs being connected to loads from twice the rated load (i.e. 16 instead of 8 ohms) or even higher all the way down to a short. They have trouble with totally open outputs. Solid state amps are fine with totally open outputs, but struggle and can even die with shorted output loads. A well designed head switcher will never leave any output either shorted or open.

A problem with relays is that "simultaneous switching" is almost never achieved. They're mechanical devices, and the time between one contact opening/closing and the other, even on the same relay, can be up to milliseconds apart. Sounds crazy, but that's long enough for a SS output stage to kill itself, or for a tube output to go into oscillation. For tube amp safety, I'd probably put a 100-220ohm resistor directly across the input jack on all four channels to never leave a possible tube amp open circuited. SS amps won't mind the slight extra loading.

Relays also take different times to make and break; so it's entirely possible that two amps would have outputs paralleled. Tube amps won't mind this at all, as their output impedances will prevent any issues. SS amps might tolerate it or go nuts, depending on the unknowable slight differences between the two. Shorted SS to tube is probably OK, because of the tube amp's impedance. So FAST switching or sequenced switching is a good idea. You might want to do switching as a two-step: switch the going-to-be-off amp to its dummy load first, then switch the going-to-be-on amp to the real speaker to avoid simultaneous connections. I would do this with a $1 PIC microcontroller, but you can do a two-step timer with CMOS or discrete devices too. The two-step would prevent any amp-to-amp connections.

Then there's grounding. Yes, the conservative approach is better, and not ever connecting speaker ground to speaker ground is better. There is a world of two-wire amps that still have not been converted to three-wire, or are not correctly grounded internally, so you have to assume that one or more of the amps will be a doesn't-play-well-with-others type.

Then there are the input issues. A proper amp switcher would switch both the inputs and outputs. You have the same grounding issues at the amp input as the output. A proper head switcher, IMHO should either switch both input and ground and never connect the grounds, or have transformer isolated inputs to keep the grounds apart. You never know what kind of crazy amp will be connected to the inputs, so the grounds should always be kept apart.

It gets complicated.

[solderhead]

I don't think that the OP's customer actually understood the complexity of his request. To design a box like this that will be idiot proof is not a simple task. For the amount of engineering that's designed to go into this type of product I would never consider a one-off design -- to make the project worth your while you'll need to leverage the design through mass production. Then you have to think about the case where an end user is going to try using the device to footswitch heads/cabinets on the fly -- you'll have to build in very careful synchronized switching to assure that a tube amp isn't presented with a high signal/no load condition.

Me? I just use flip switches with the amps on standby.

[solderhead]

oops. double post.

[R.G.]

I don't think that the OP's customer actually understood the complexity of his request.

I agree.

Me? I just use flip switches with the amps on standby.

If I tried that, I would inadvertentlly find ways to make it amp-destructive.

[Yoda]

Thanks for the comments guys.

Yes my next questions were going to be in regards to switching speeds because I was seeing comments elsewhere about microprocessors being used to perform the switching to prevent damage to the amp but I didn’t know how critical this could be. Anything involving microcontrollers and such would be beyond my abilities I’m afraid.

I did look at purpose-made units like the KHE 8x4 and they are impressive and very well thought-out, and I certainly believe you get what you pay for but the price is probably what prompted the customer request.

To DIY this is a little bit nerve-racking because the fate of four amplifiers hang in the balance if it’s not perfect. Imagine the liability if someone hooks up 4 or more vintage amps to one of these things and they all blow.

[danman]

I was looking for a way to switch between my four amp heads and four cabs a few years ago. My situation was much simpler because I didn't need to switch the heads while they were powered up. I simply needed a way to connect different heads to different cabs without pulling them out of the shelving unit that I store them on. I used two heavy duty, 3 deck, 4 position switches...one for the heads and one for the cabs. I paralleled all three decks of the switch to triple the current rating of the contacts and connected each of the four switch contacts to four Cliff jacks. This allows me to use any of my four amp heads with any of the four cabs. I always make sure to do the switching with the power off and triple check myself before powering up the amp that I want to use. I also have a safety resistor across each amp head just in case I make a mistake. It has been working fine for several years now and has definitely saved me time by not having to pull everything out to switch the cables. This isn't the exact switch that I used but it's similar.https://www.ebay.com/itm/326342792970?i ... SwfCVnOpVs

[Yoda]

I’m glad that worked for you. Something like that was doable for me but I still worry too much about it being switched on the fly or by someone unfamiliar with it or maybe without the owner knowing, or by accident in a dark stage/studio, etc. you just never know.

In my case the I believe the customer will be going with the KHE unit in lieu of taking any chances with a DIY deal. Probably best for everybody and especially for the amps.