The double blind auditions thread

[Jakob2]

Originally Posted by fireworks

I still have an unanswered question: is it ok to switch quickly (using a relay) the load of a power amp from 8 ohms to no load and back a few seconds later ?

As godfrey said, you have to consider what the amplifiers need for safe operation. Another point to think about would be the choice of relays because, depending on load conditions and signal, contact wear is a severe factor.

But the more important question should be, what is the intention behind the switching ?

Ok, to avoid the obvious- if you want to compare to power amplifiers, you have to exchange the loads (aka loudspeakers) and inputs.

But, is rapid switching needed or could you just stop (or fade out) the music and do the switch after that?
Is a switch box a must have or would manual switching already be sufficient?

These questions are connected to the question of practical relevance of any test results and to questions of test strategies and efficiency as well.

[fireworks]

Originally Posted by Jakob2

But the more important question should be, what is the intention behind the switching ?

Ok, to avoid the obvious- if you want to compare to power amplifiers, you have to exchange the loads (aka loudspeakers) and inputs.

But, is rapid switching needed or could you just stop (or fade out) the music and do the switch after that?
Is a switch box a must have or would manual switching already be sufficient?

I want to use relays to switch power amps to the same set of speakers. Switching of the inputs should not be needed: both amps would be connected in parallel to the same low output impedance source.
Yes, rapid switching is a requirement, so switching cables will not do.
Fading in and out the source signal could be envisaged but it would be an additional complexity.

[Jakob2]

Originally Posted by fireworks

I want to use relays to switch power amps to the same set of speakers. Switching of the inputs should not be needed: both amps would be connected in parallel to the same low output impedance source.
Yes, rapid switching is a requirement, so switching cables will not do.
Fading in and out the source signal could be envisaged but it would be an additional complexity.

Wrt to the relays i recommend to use two relays in parallel, one to take the potentially leading to contact wear effects and the other to switch in after the bounce time of the first.
The first relay should have wolfram based contact material; the other should have some nice high quality contacts.

Switching off would be done in reverse order.

I guess that you´ll use the box for listening tests, right?
Then you should better use relays for the input signal as well. This way you will cancel the possibilities for any current loops within the two DUTs.

Even when driven by a low impedance preamplifier stage this will most probably happen within the ground wires. (Symmetrical drives might be different if realized in a proper way)

If listening test results should have practical relevance i´d question the need for rapid switching; if a difference is only detectable with rapid switching (because listeners aural memory is weak) why should it be of relevance for normal listening?
You would simply not notice any improvement the next day.

Otoh rapid switching might lead to a different "awareness" which could prevent participants from "holistic" listening.

[ThorstenL]

Hi,

Originally Posted by fireworks

I want to use relays to switch power amps to the same set of speakers.

A few rules.

1) Funk the EU ROHS regulations and get surplus mercury relays while you can (I am just contemplating to buy a job lot of 1k 4-Way mercury wetted reed relays).

2) Do switch the unused amplifier to a dummy load.

3) Mute during switching, making sure not to un-mute until the relay is reliably closed.

The above makes sure that amplifiers that can take damage without load but with signal (e.g. Vintage Tube Amp's) are protected.

Ciao T

[fireworks]

Originally Posted by Jakob2

Wrt to the relays i recommend to use two relays in parallel, one to take the potentially leading to contact wear effects and the other to switch in after the bounce time of the first.
The first relay should have wolfram based contact material; the other should have some nice high quality contacts.

Switching off would be done in reverse order.

Interesting idea ! This would be the second level, the first being just a simple relay switcher.

Originally Posted by Jakob2

I guess that you´ll use the box for listening tests, right?
Then you should better use relays for the input signal as well. This way you will cancel the possibilities for any current loops within the two DUTs.

Even when driven by a low impedance preamplifier stage this will most probably happen within the ground wires. (Symmetrical drives might be different if realized in a proper way)

Can there be a ground loop if both amps use the same wall socket for supply ?

Originally Posted by Jakob2

If listening test results should have practical relevance i´d question the need for rapid switching; if a difference is only detectable with rapid switching (because listeners aural memory is weak) why should it be of relevance for normal listening?
You would simply not notice any improvement the next day.

Otoh rapid switching might lead to a different "awareness" which could prevent participants from "holistic" listening.

In my opinion, if it's audible when fast switching, then that's a good result !
I'm afraid it will be very difficult to detect any difference between two good amps, even when using fast switching !
I'd rather do an easier test first (try to find differences with fast switching) and if that goes well maybe take on a more difficult one.

In the realm of statistically significant differences when listening it seems that the norm is negative results, at least with modern amps.