Anyone know how to build one of these?

[danotron]

https://taweber.powweb.com/store/2head.htm

I suspect its not that hard or expensive...any guesses?

[paulster]

It'll just be a DPDT relay that has two inputs (head 1 and 2) and is wired so that you get two outputs, either input 1 to output 1 and input 2 to output 2 or input 1 to output 2 and input 2 to output 1.

Output 1 goes to the speaker cabinet and output 2 goes to the impedance selector and then to the dummy load.

To be honest it's not a lot of use without selecting the corresponding inputs to the amps at the same time so that the amp that's not connected isn't receiving any signal. At that point you could use a smaller load resistor as you'll only have to deal with a few trailing repeats if you've got delay in the loop.

[ampdoc1]

If you try to make one of these, be sure you use relays with the current/voltage requirements for a speaker level signal. The little relays used in channel switching in most amps won't hold up under these conditions.

a'doc

[paulster]

Very true. Your typical telecom relay won't be up to the job, but it could quite easily have its coil paralleled with a heavier duty relay for the speakers so that the telecom relay switches the input to two heads between the signal and ground.

[ampgeek]

Very timely post! I have been contemplating building one of these as well.

My current vision does, indeed, parallel one of Ampdoc's low level signal (input in this case) relay boards with a larger relay coil for the speakers.

Any thoughts/recommendations on a suitable speaker switching relay product line? Coil voltage will be in the 6 to 12 VDC range. Just not sure about the speaker side specification. 10'ish VAC at 3 to 5'ish amps for a 50 watt output output rating?

Those octal mount relays just seem so big and gawky but are the only thing that I have seen that is close.

Should a switching noise supression approach be integrated into the design as well?

TIA,
Dave O.

[ampdoc1]

If I've done my ohms law calcs correctly, for a 50 W amp into 8 ohms, your voltage will be 20VAC@2.5A, or into 4 ohms, 14VAC@3.6A

a'doc

[paulster]

Your calculation is spot on. Factor in peaks at 100W so use the figures:
4ohm: 20V @ 5A
8ohm: 28V @ 3.5A
16ohm: 40V @ 2.5A

to comfortably accommodate a 50W amp at full tilt.

[Structo]

In addition to the relays, wouldn't you also need something like a resistor shunt or make before break type thing so that when you switch there isn't a point where there isn't a load on the transformer?

Seems like it would want to pop or spike pretty good otherwise.

[paulster]

In addition to the relays, wouldn't you also need something like a resistor shunt or make before break type thing so that when you switch there isn't a point where there isn't a load on the transformer?

I had a discussion about this with an exceedingly knowledgeable and respected amp (and guitar) builder who is about to bring one of these to production and he told me that as long as you picked a relay with nice fast switching contacts there were no issues whatsoever.

A number of high profile artists are using custom-shop versions of the product he's going to bring to market so this theory is holding up on some big stages already.

That was exactly my concern before building one myself and hence the conversation.

[Structo]

Well isn't there already the Headbone for this purpose?

http://www.tonebone.com/tb-headbone-vt.htm

[paulster]

It doesn't like higher gain amps apparently as the line and speaker level signals are too close, and it whistles or squeals. That's been the common complaint.

[danotron]

Is it this simple? Anything else needed? component values?

[paulster]

That's the basic principle. You would probably want to use 4PDT or a pair of DPDT relays for input and output sides though so you can switch ground as well as signal.

Otherwise you're tying the grounds together at the inputs to both amps which is likely to give you a ground loop and also at the outputs which in itself is likely to give you a ground loop.

[alanp]

You'd also want make-before-break switches too, wouldn't you?

[paulster]

Apparently not. Just so long as you use fast switching relays.