How to make a $20 audio switch-over comparator © Alan

[Drdennis]

Alan,

Does the placement near the wall have a significant impact on the sound?
I too have a small room (12x12x9), and I have my C7ES3's about three feet from the back wall.

Dennis Girard

[A.S.]

Well, I think the success of the C7ES3 tells its own story. Yes, of course, the proximity of adjacent walls always lifts the bass somewhat, but that's not necessarily a bad thing if the response remains relatively smooth. If it worked for me in a small bedroom, then it certainly should work for you. What helps is that I have lots of books etc. throughout, so the acoustic is tolerable.

BTW, I have recently re-written the User Manual and I've made more mention of speakers in corners and how to generally improve the room acoustics. We'll be changing to that version shortly, and at that point I'll make the new manual available on the website.

[honmanm]

I must be missing something here, but it hasn't been possible to track down relays with mercury or gold contacts (there are mercury contactors, but those start at 35A and about £40!).

Common contact materials appear to be nickel silver (AgNi), silver tin oxide (AgSnO), and silver cadmium oxide (AgCdO). AgNi is a pale yellow colour, and the relays in the Velleman speaker protection kit are AgCdO. There are relays with gold-plated contacts, with current ratings around the 3A mark.

Here are 3 possible candidates (for comparison purposes, all same manufacturer and supplier, and DPCO type):

• 1A AC current, gold plated AgNi
  http://uk.farnell.com/tyco-electroni...vdc/dp/4219582
• 8A Ac current, AgNi
  http://uk.farnell.com/tyco-electroni...12v/dp/9659293
• 8A current, AgCdO (this one is of similar contruction to the example posted by A.S.)
  http://uk.farnell.com/tyco-electroni...2vdc/dp/388336

(for cable testing experiments the plan is to have remote relays at the speaker ends, hence 2-pole rather than 4 pole... unless once again I'm missing something).

[Takis]

Reading about Mr. A. Shaw’s $20 audio comparator, I decided to build one, so that I can switch between two amps (driven from the same source component) on the fly, driving one pair of speakers, to detect any differences in sound. Very intriguing indeed…
I used 4 double FINDER 40.52 relays. Specs here: http://www.findernet.com/comuni/pdf/S40EN.pdf
Two of them used in connecting simultaneously the two amps (both hot and cold, eight contacts) to the speakers (four contacts) taking good care that left+ of each amp goes to left+ of the speakers, left- to left- and so on.
The other two relays provide a dummy load of 8ohm/60watt for each of the four channels in case that one (or both) of the amps are tube amps and isn’t safe to operate without a load. The dummy loads are inserted in the circuit in pairs independent from one another, when needed, by two twin switches (the red on the right side). They don’t affect in any way the sound, they are totally independent, and connected automatically to the amp (if I want to) which -at the moment- isn’t driving the speakers.
The relays are driven by 6 V DC, they are battery operated too, and I’ve added remote control with a simple on-off switch and 6 meters of cable.

When DC is OFF (relays free), amp 2 is powering the speakers and amp 1 nothing, or the dummy load (if engaged). When DC is ON (relays engaged), amp 1 is powering the speakers and amp 2 nothing, or the dummy load (if engaged). There is also a fan for keeping the power resistors cool (I never used it). In normal listening levels the resistors of the dummy load, if engaged, are becoming only slightly warm. This “toy” can also be used in the opposite direction to feed two pairs of speakers, each at a time, with one amp.

It works perfectly, you can’t even detect the gap in time between connect – disconnect when pushing the magic button! If a relay fails, nothing happens. There is no case that the one amp will be connected to the other one.

I’ve spent less than € 50 and a few hours of my free time on that. Who needs this toy? Perhaps no one. But still, it is fun to watch fellow music lovers, unable to identify their own amp when compared -on the fly- to a “lesser” or “better” one, or finding that differences between good, or decent if you like, amplifiers when playing music at exactly the same volume, are not easily / practically detectable / existing! (Use the words you like). Note that any time I gave a very slightly higher volume to an amp, it got all the votes hands down! (I’ve never used so far, a solid state amp vs a tube one, they were all ss). Now think, what makes a difference in sound reproduction at home.

A great idea Mr. Shaw! Amusing in many ways, and didactic!

IMG_0392.JPGIMG_0393.JPGIMG_0394.JPGIMG_0395.JPG

{Moderator's comment: You see for yourself .. the LOUDNESS difference is the real reason amps "sound different". It is as simple as that.}

[BAS-H]

Excellent, Takis, thank you.

Moderator's comment: You see for yourself .. the LOUDNESS difference is the real reason amps "sound different". It is as simple as that

THE reason? No. The major one, certainly, but there are others. E.g., compare a reference amp with a valve amp with an inadequate output transformer, playing at a given SPL. One can reproduce the whack of a grand piano; the other can't. I'm afraid things aren't so simple!

[darrylm]

Do you happen to have a schematic for the construction of this audio comparator?

[Takis]

Sorry darrylm,

I didn't see your post earlier. I hope this helps:

[coredump]

mechanical relays has this problem of bouncing (mechanical oscillation).

[HUG-1]

mechanical relays has this problem of bouncing (mechanical oscillation).

That is not an issue for switching audio.

One very important point ..... according to the circuit diagram it is possible that the output of the two amplifiers could be momentarily connected to each other and to the speakers. This could damage or destroy the amplifiers or even the speakers.

When selecting the relays to use be sure to buy 'Break before Make' and not 'Make before Break' contact types. This means that the swinging contact must positively disconnect from one contact before swinging to the other. It will ensure that the output from amp A is isolated from the speaker before amp B is connected to the speaker and avoid the two amps being briefly shorted together.

[A.S.]

Sorry - I really should have answered this one ages ago.

Attached you will see a close-up picture of the working parts of a relay. It's about 3cms long, and I've removed the protective case. Can you see that two of the contacts are rigidly fixed in place, but under the influence of a current in the coil, the centre contact can be influenced to move from the rest position. When the current is disconnected, it flops back to rest again.

About mechanical oscillation: as we are only operating a go/no go with a cycle of seconds or minutes, and the contacts are physically quite well separated, this really shouldn't be an issue.

>

[A.S.]

Here are pictures of the internal and external arrangement of my Mk3 switcher. This one has been in use since 2003, and was first used during the development of the Compact 7ES3. At that time, I had two candidate crossover circuits with equally good frequency responses but a radically different circuit complexity. I just had to build this multi-way switcher to actually listen to the crossovers driving real woofers and tweeters with zero time delay to fiddle about changing-over the wires. As relays do not 'know' how they re being used, a switch-over box for switching one amplifier between speakers A and B can equally well be used in reverse to switch one speaker between two amplifiers.

The essence of critical evaluation of audio event A v. B is to eliminate the time gap between these events. A simple tool like the relay is the idea way to eliminate any reliance on audio memory.

>

[Takis]

That is not an issue for switching audio.

One very important point ..... according to the circuit diagram it is possible that the output of the two amplifiers could be momentarily connected to each other and to the speakers. This could damage or destroy the amplifiers or even the speakers.

When selecting the relays to use be sure to buy 'Break before Make' and not 'Make before Break' contact types. This means that the swinging contact must positively disconnect from one contact before swinging to the other. It will ensure that the output from amp A is isolated from the speaker before amp B is connected to the speaker and avoid the two amps being briefly shorted together.

I used C form contact type, DPDT relays: Double Pole Double Throw, actuated by a single coil, "break before make"

The swinging contacts are connected to the speakers, so there is no way that the amps will "see" each other in any way. The amps are connected to the stable contacts. They are, as Mr. Shaw says, physically quite well separated.

Up to now, it operates without any fault.

[A.S.]

I used C form contact type, DPDT relays: Double Pole Double Throw, actuated by a single coil, "brake before make"

The swinging contacts are connected to the speakers, so there is no way that the amps will "see" each other in any way. The amps are connected to the stable contacts. They are, as Mr. Shaw says, physically quite well separated.

Up to now, it operates without any fault.

That's great. Relays are extremely simple devices, normally very reliable even after millions of switch-overs and best of all, you can look inside and see if they have operated or not.

Recently, when I proposed the relay-comparator for comparing amplifiers (or used the other way around, comparing speakers) I admit be being surprised by the extent of personal abuse directed at me for even daring to propose such a simple, cheap and effective means of comparing A with B. But how can you fault the concept? No technical skills required, you can observe the operation of the change-over, it's clear that the contacts are made of good quality material (gold if you are willing to pay for it) and it will work reliably for years or decades.

I couldn't design loudspeakers of the Harbeth type unless I removed my personal emotion from the process and could directly and instantaneously compare A with B. It is the primary tool in my toolkit. It really should be standard issue to anyone serious about critiquing audio equipment - alongside an audio voltmeter and sound level meter. No other skills than basic high school science is required.

[A.S.]

... Relays are extremely simple devices, normally very reliable even after millions of switch-overs and best of all, you can look inside and see if they have operated or not...

We recently invested in two brand new, identical, Danish made, state-of-the-art audio test systems which I have been learning how to use, before training my colleagues and handing over to production staff. The learning curve has been much steeper than I expected. I'm nearly there now and have almost solved the programming of our label printer to print information relating to each drive unit which then is attached to the woofer or tweeter. That will save writing information on their underside by hand. We'll cover this $12,000 investment in an upcoming newsletter.

It's a big jump to invest total confidence in a computerised system. For 25+ years we've completed the computerised measurement test of 100% of drive units by listening to a wide-range audio sweep on every woofer and tweeter before (and after) they are mounted in the cabinet. The new system claims to be able to use its electronic ear to do that, faster and more reliably than a human. That's quite a bold claim! So, until we can convince ourselves (if ever) of that, I've decided to use a hybrid approach to testing with the new system, as we do with the old: let the computer do the measuring of basic parameters, and the human does the listening. If both say OK, then the drive unit is acceptable.

To implement this I have to make a foot-switch operated change-over box with a relay, to send either the computer test signal or the hand-operated swept tone the the woofer or tweeter under test. I have a spare QUAD 303 that'll boost the swept tone to just the right level. I've ordered a little metal box to mount the parts in, bought a foot switch on Ebay (about $3). The circuit arrangement is conceptually the same as my critical listening A-B comparator, and I'll take some pictures as I go along. Then you can see for yourself just how simple it is to make an A-B switchover box for yourself.

[SDRichard]

> But still, it is fun to watch fellow music lovers, unable to identify their own amp when compared -on the fly- to a “lesser” or “better” one

Thank you so much so sharing, Takis. I learned so much by reading this forum.

My big question is, is it safe for me to just walk into a consumer electronic store (like BestBuy here in the US), and buy whatever Sony or Yamaha has to offer? Let's not knock your "better" amp, but what was your "lesser" one?

Now, don't get me wrong. Like many here, we all appreciate nice things. I still like to have a nice, gorgeous looking Italian amp driving my little Harbeth (P3). But these things don't do any of the modern stuff. They don't stream Pandora music, they don't do Apple Airplay.

Life is getting really complicated in terms of audio contents. I wanted to switch to receiver type equipment so I can consolidate all electronics into one box without messing with many boxes and trying to make them all work together. But at the end of the day, sound is still one of the most important factor me to.

Thanks for sharing your thoughts. Takis, if you make another one of your box, I'll buy it from you .

Finally, a big big thank to Alan. You are truly a man of integrity. (And a brave man.) Very few in this business dare to discuss the truth openly.

Richard
San Diego, US