Neal's 2A3 Amp Measurements at Owston

Andrew · Post by **Andrew** » Tue Jun 18, 2013 6:39 pm

Here's the images of Neal's amp.

Andrew · Post by **Andrew** » Tue Jun 18, 2013 6:44 pm

OK right channel.

Neal · Post by **Neal** » Tue Jun 18, 2013 6:56 pm

Cheers Andrew, I manage to trim the hum down to 0.1mV when I got the amp home and had time to move a few wires about underneath!

Andrew · Post by **Andrew** » Tue Jun 18, 2013 7:17 pm

Excellent news

Andrew · Post by **Andrew** » Tue Jun 18, 2013 7:25 pm

Neal, these are reduced so that they can be loaded on the board, drop me a PM if you want the originals zipped up and sent your way. Thom you're next

Neal · Post by **Neal** » Wed Jun 19, 2013 11:01 am

Ok thanks Andrew.

Measured my WAD K5881 today on the HP 333A

Figures are corrected for sig gen rated distortion:

100Hz
1W: 0.09%
10W: 0.25%

1Khz
1W: 0.03%
10W: 0.06%

10Khz
1W: 0.09%
10W: 0.51%

WAD rated the K5881 MKII at 0.012% THD+N so given my HP333A is out of cal the figures aren't too bad!

The rise in distortion at 10W, 10Khz is due to a slight imbalance creeping in to the phase splitter output, its just starting to be visible on the 'scope.

JamesD · Post by **JamesD** » Wed Jun 19, 2013 11:08 am

Those a damn fine figures for a valve amp! And that phase splitter is doing very well too not show up until 10W at 10KHz...

At 10Watts is the beastie moving towards A2 and is that the extra loading or is something else going on?

James

SimonC · Post by **SimonC** » Wed Jun 19, 2013 12:37 pm

Hi Neal (& Andrew),

If I could have made it on Saturday I would have brought along a virtually std WAD 5881 for comparison. Would have been interesting to have done back to back sound and measurement comparisons so see what effect your mods have had.

The subject of measurement is always an interesting one... why does SE DHT sound so damm good, when the measurements say it shouldn't...

Simon C

JamesD · Post by **JamesD** » Wed Jun 19, 2013 2:30 pm

Flip answer is 'because we're measuring the wrong things!"

More considered answer is that what we measure only minimally correlates with how we hear...

Measuring is an engineering activity and we can use that to design better measuring amplifiers and sometimes that also gives us better sounding amplifiers...

For instance: Looking at the spectrum measured by Andrew of Neals amplifier the thing that stands out for me is the inter-modulation of 50 & 100Hz components with the 1Khz test signal - visible also on the test signal harmonics - now that is 'horrible' and far more worth removing than the direct hum caused by the 50 and 100Hz signals themselves... but which makes the amplifier sound better???

Reducing Intermod, TID, SID, etc. is a much bigger thing than reducing THD once THD is under, say 0.5%... and measuring and testing is the way to do this in a controlled manner...

So its not the measurements saying SE shouldn't sound good - the measurements say nothing about the sound of an amplifier per se but our expectation of what the measurements might mean about the sound of an amplifier that is 'wrong'... We are insufficiently knowledgeable about the sound of measurements

Andrew · Post by **Andrew** » Wed Jun 19, 2013 5:43 pm

Well put James.

Andrew

SimonC · Post by **SimonC** » Thu Jun 20, 2013 12:56 pm

JamesD wrote:
More considered answer is that what we measure only minimally correlates with how we hear...

We are insufficiently knowledgeable about the sound of measurements

I always love this debate

The engineer in me wants to find valid measurement methods, because once you can reduce it to numbers on a page you can then simulate it, and once you have a valid simulation you can iterate the design numerically to achieve whatever result you are aiming for. (Martin King has had a pretty impressive go at this for speaker simulations, and we generally like the results).

The artist in me hopes that I never find a valid measurement method, because as soon as you reduce it to numbers on a page it loses the element of human intuition and mystery. (For me losing this human element would turn it from a hobby into a chore).

Simon C

Post by **Nick** » Thu Jun 20, 2013 8:32 pm

If you can't measure what you value, you end up valuing what you measure.

Paul Barker · Post by **Paul Barker** » Thu Jun 20, 2013 8:40 pm

Maybe the relationship is to an extent measurements guide, but they aren't everything.

JamesD · Post by **JamesD** » Thu Jun 20, 2013 9:15 pm

Nick,

that is the trap - surely the art is learning to stand on quicksand?

Andrew · Post by **Andrew** » Wed Jul 03, 2013 6:39 pm

So I just measured my 300B using the test set I have, not the PC set up used at Owston, they do produce much the same results and have much the same noise floor, but the test set can't do an FFT. The reason for the test set is that its easier to move around and the amp and PC are not in the same room.

Anyway, I got 0.6% THD at 1 watt at 1Khz and get 5% THD at 6 watts, for a work in progress and a SET amp this compares OK, I think, to Neal's excellent sounding PP.

cheers,

Andrew

audio-talk

Neal's 2A3 Amp Measurements at Owston

#1 Neal's 2A3 Amp Measurements at Owston

#2

#3

#4

#5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15