I bought the Moon... and all I got was an Eclipse.

Why is that?

If you are refering to the fact of the amplifier under test not being connected to any speaker then let me tell you something:

Show me an amplifier that performs well in a null test with the above mentioned arrangement and I'll buy it without even blinking.

And if you want to connect a speaker just put it in other room (or use an oscilloscope instead) for hearing only the sound of the distortion from the speaker connected to the "measure" amplifier.

 

Show me an amplifier that performs well in a null test with the above mentioned arrangement and I'll buy it without even blinking.
Erm, Quad 405...

That's why you have to be careful it's actually driving something representative at the time

 

You wrote.

What if both the amplifiers are earthed?

Paul

 

Are you sure about that? Do you null-tested it?

Because if you are right, which I doubt, I'll buy it... no doubt!

 

Then make sure that they aren't.

 

Paul - Beryllium's setup may not work in all cases but that's really beside the point. There are little null tester boxes you can get which make it all a lot simpler than messing about manually as described. They usually include a volume pot to allow you to null test a power amp in isolation (so you can match the input and output voltages).

One thing I haven't quite understood about null testing though is that there has to be a finite delay between the input signal and the amplified output signal which would cause the two to be ever so slightly out of phase with each other which would completely destroy any null test result. In fact, the phase shift would worsen with increasing input frequency.

As a result, you could have a theoretically perfect amp, but if it introduced a uniform 500ms delay across the frequency spectrum it would utterly "fail" any null test. That's not to discredit the value of the null test, any amp having a perfect "null" is still perfect, it just means that there are theoretically perfect amps that would have a terrible null test.

This may be a complete non-issue though. I have no idea of the typical delay between the input and output of a power amp. It may well be so small as to not affect null testing in the audio band.

Michael.

 

So now you're advocating unsafe operation?

Why not just do it properly.

Paul

 

It's not beside the point should someone blindly follow his instructions and kill his amp....

I think the propagation delay through something like a power amp is negligible. If you consider the way the feedback works the output is present at the input, and the phase between the two defines stability. So in the audio band don't worry about it.

A proper digital amp is a different issue, it will have a delay while processing that would require a compensatory delay before the null.

Paul

 

Back to Eclipse

Their bags at the Heathrow show were great. Got one and I still use it!

I think they were the runners at the show.
So many people coming out of the Eclipse room had one. Some even a hand full.

 

BerylliumDust - personaly, i haven't tested a 405 - yet. But null testing was one of the standard tools that Peter Walker did during it's development. He even demonstrated the things daisychained in series as part of a null test. Shame then the amp as sold was so obviously unhappy with many speakers less than a resistive 6 ohms.

The point, though, is that with any such testing the test results are only as good as the assumptions underlying the procedure.

 

Paul,

I use an old Sony for this purpose which is not connected to earth and never was in its stock form.

If that make it unsafe write a letter to Sony Corporation.

 

Ah OK I see it now, Friday night etc...
It still doesn't seem very "real world" to me. But there you are, testing has proven that daisy-chained Quad 405s are the perfect amp

 

Joel,

What doesn't seem very "real world" to you?

We are talking about the signal coming out from your CD player (a musical signal)... this is all what you are intended to hear even if you don't like it.

Daisy-chained Quad 405s fed with static signals like pure sinusoids at fixed frequencies tell you nothing about "real world" music.

 

Michael,

You are wrong. Tube Dude's amp is the first and most probably the only one amplifier that you weren't able to hear.

You can't possible know how it sounds because it doesn't have any sound at all. Now that's revolution from your bed...

All you've heard were his CD player and speakers, which as you said are amongst the best you've ever heard.

An amplifier with no sound, man... that would do a very nice review.

 

He should probably invest in a volume control then, sounds like it could do with being turned up a bit.

-- Ian

 

Made me laugh.

 

Ian,

The funny thing was turning up the volume didn't help either... the most silent amplifier ever (?) it is indeed hard to criticise...

 

BD,
The concept is interesting. I am not sure about your method, though. This requires some thought.
It's a question of confirming in one's own mind if the process described is what it claims to be. It almost never is (especially in audio), so I tend to start from a sceptical POV.
Do you know what signals Peter Walker used in his null testing, and why, in the context of a null test, would these produce less meaningful results than real music?

 

Joel,

Since were both you and Felix who brought about the 405 issue could you please tell me what kind of "real world" test did Peter Walker perform to prove his claim of that being the perfect amplifier.

 

I don't know, but as you went from not knowing that PW did null tests to dismissing his tests on the basis that they weren't real-world, it thought it would be interesting to know why you came to that conclusion.