measurement bollocks.

Good point, well made. As long as everyone admits I'm right and sends me financial tributes to that effect, we can tie this thing up right away.

 

"I'm right." Happy now?

 

Yep. Now pay me

 

That is interesting.

I changed many parts in my old pre-amp (a valve active pre-amp). Many made a difference, some I am not sure.
Taking the alps blue pot out and putting a stepped attenuator in (resistor based passive) made a huge, enormous, massive difference.

Perhaps this is something to do with the soldered connection.

I mention this because since soldering my crossover together instead of crocodile clips - it instantly sounded better. A carbon track I imagine just uses friction for connection, rather like a crocodile clip.

Perhaps also a carbon track is not the best possible material?


All conjecture.

 

I thought the black Noble pots were metal film, the ones I stripped to make a matched unit certainly were.

As long as Rob is happy that his dogma backs up his failing hearing and middle of the road hifi i think we can call it a day.

 

The difference is I'll do such testing in any location, under any conditions, even in your home and stand by the result. In other words, I stand by what I claim 100% and am prepared for any challenge.
I do so regularly and it naturally informs my opinion. If results over a period of time start to indicate something different then I'll re-think. Not until then.

 

You found my humour button.

 

Are there any further details to that Quad test with 50, was it 303's, in a row?

 

The problem is that much of what you are reporting are expected differences and you haven't described any measures taken to ensure consistency and removal of bias.

So for example, it is accepted wisdom that a stepped attenuator is better than a cheap carbon or plastic pot. The only evidence is similarly anecdotal to your own reporting. Add the fact that they look more impressive and cost more. Then add in the time lag taken for you to change the component, and did you ensure that you matched levels at least reasonably with an SPL meter?
Given the highly directive nature of your speakers, where you sitting in precisely the same listening position?

There is no technical reason for a carbon pot to perform less well other than the channel balance usually isn't as good. If the pot is old and dirty that is a different matter but I presume it wasn't.

So you should test your findings by addressing the above factors and listening unsighted IMO. It would address at least some of the potential pitfalls.

On the crossover, the same basic things apply though there is more chance of a real difference emerging, so all the more reason to be sure.
You probably improved the layout within the negative leg of the crossover when you soldered it up. My lash-up at your place was to get the circuit working with the tools and methods on hand.
There is no reason for the solder joints to be audible.

For anyone interested in distortion caused by contacts and lets say less than pristine electrical connections I would recommend Doug Self's series of books on Audio electronics. Listening tests and the science to back them up.

 

Peter Walker never did it, though always wanted to.

Personally I think that would be audible!
I get the gist of his point though - that ears are a lot less discerning than we think and that much passes unnoticed.
I'll do something along those lines soon - perhaps 5 passes through a preamp against nothing in the path, something like that.

Next up is the microphony test lined up for the weekend.

 

I should have spoke in past tense as I left the trade from a sales postion in '88.

The room in five of the shops I continued to do TT setups and misc repairs for were designed for such and ranged from OK to rather good. Still, there's something about a shop environment with people walking in and out or just visible in the hallway through room's glass windows, etc that was a distraction for most folks including those of us that worked there. There's also something to be said about your extreme familiarity with your own room, system and the limited distractions in comparison as well as lower ambient noise.

Again, my opinion only and based on forty years in the hobby.

 

According to Dada / Quad Spot the 303 has rather poor crosstalk figures so I'm pretty sure it would be audible long before you stacked 50 of them up! They recommend using them as monos or bi-amping one per side, though I've not noticed any issue myself. I think it's a great little amp though, it gets a lot of stuff very right IMO assuming you don't push it too hard and I certainly like mine into my Heresys. It's a smooth and easy-going amp with a nice clean mid which is exactly what these normally tube-partnered speakers need.

Peter Walker's comments need context too, at the time they were made the 303 would have been used with a 33 (not the world's greatest preamp) and it's phono stage hooked up to a probably not too well installed turntable, i.e. the 303 and ESL's that would no doubt have been in the picture would have been way stronger than the source. A real mullet system, even with something classic like a 301/401 or 124 with a Decca or SPU up front as the phono stage would have been such a bottle-neck.

Tony.

 

It seems that is a 'no' then - I'll pop back in a few days and see if things have moved on (but I doubt they will have).

 

Microphony. Something on which we can agree. Solid state power amp, nah, pre-amp, maybe just if it's high gain, phonostage, he'll yeh I only have to clunk my SME clamp on the box for mine and my cones jump noticeably.

Recording the output of the phonostage with and without speakers playing is aalso telling.

But in my set up at least pre and power amps are utterly immune.

Are you thinking about shaking some cables?

 

The 303's frequency response reaches -1dB at 35kHz and presumably follows a first order slope, so it would be -3dB at 70kHz and -0.34dB at 20kHz.

Cascade 50 of these and you'll get -50dB at 35kHz, -17dB at 20kHz, -12dB at 10kHz. That is going to be very audible.

Presumably Walker intended this test to be with band-limited signals around 1kHz only?

 

Simon I was going to try testing a SS pre amp.

One containing lots of active op amp based stages, lots of gain and no particular attmept to control microphony.
It will be a pretty extreme test subjecting the unit to high SPLs at close range.
I plan to plot and record the output at full gain and put the results up for view.
Once it has been done, all suggestions at varying the test will be welcome.

Lets see what we get.
Can then try a valve phono stage under the same conditions.
It should establish a few reference points at least.

 

Mark, the topic is a bit of a catch-all and naturally voves on.

We're now looking at microphony and cascading amplifers.

 

Agree entirely.

Using modern wideband kit and limiting to perhaps 5 or 10 passes would probably give a more useful result.

 

Would this be someone presented with 2 glasses of wine not knowing the provenance or price?

What if they could detect a difference but prefered the cheaper?

Is there an objective standard for the "quality" of wine?

Do you buy wine and hi-fi on price alone?

 

Preference and difference tend to get confused during these discussions

Objective testing is only concerned with identifying differences - nothing more.

Preferences are of course entirely subjective.