Mr Simon,
how vague you are. Act of diplomacy ?
I wouldn't want to spoil Linear Audio's next issue!
Sorry Ed if I am reading this correctly you have made mistake No. 1 you can change only one thing at a time or this can not be called an experiment. Fun maybe, like my work over of a Hafler kit. You will have a hard time convincing me that increasing the bias to 50W a side was not the only real difference that mattered (pretty slight BTW). All the crap resistors stayed.
I don't have to imagine, I have one. It's a M size (5'-7") built in 1927. My aunt started out in life as a concert pianist. She played Town Hall when she was 16. The piano was a gift to her from my grandfather. It's been in my house for about 6 or 7 years. The piano was rebuilt at the Baldwin factory in the mid 1980s when Steinway was going through bankruptcy or it would have been done there.
On may piano recordings of Steinway pianos, I can come very close to the timbre of the Steinway. It has a very characteristic tonal signiture. More forward and warmer, slightly less brilliant than my Baldwin. What modified AR9 can't duplicate is the presence of the piano. This is because of the gross difference between the way the piano radiates sound into space and the way the speakers radiate sound into space. It's a manifestation of geometric distortion of the field. To correct it, I'd have to redesign the speakers from the gound up. Too much trouble and very expensive. But if I were going to put money and effort into it, that's where it would go. Much more beneficial than replacing a preamplifier IMO.
Scott,
The test is just for the feedback resistor, that is the variable. The preamble is how I got the test amplifier for anyone who wants to repeat the experiment.
I did not compare the before amplifier to the after. That was not the issue. I do have techniques that can be applied to any amplifier and I used some of them on this one, the idea being to eliminate the objection that the amplifier wasn't good enough to show a difference.
The question is can one resistor change the sound of the amplifier when used as the feedback resistor?
So I had an improved amplifier with one switchable variable, the feedback resistor.
One feedback resistor was carbon comp the other a metal film. I trimmed the carbon comp into matching the metal film with parallel carbon film resistors. So both values were 33.25K Of course the carbon comp will change a bit with ambient temperature, but I want to reduce the problem with the gain change ruining the results.
I used all six positions of the switch to select the one in use.
At first I heard clearly heard changes that did not correspond to just changing resistor types, so I took some measurements.
They showed an obvious increase in the noise floor. BUT IT DID NOT TRACK RESISTOR TYPES!
Attachments
Thanks for the tip. Will use it in the future.
But the problem here turned out to be the switch noise!
So as JC mentioned silver switches work best with wiping action. So a bit of cleaning and I ran 19 & 20khz IM distortion plots to compare the two types.
If anyone can see a difference between the two types please point it out to me as I do not on this type of measurement. Note the scale does not show the full input signal.
More after lunch. Converting the plots is slow going.
Attachments
Last edited:
You mean you haven't replaced switching in the signal path with mercury wetted relays? And you people call yourselves high end?
Claims made with plausibility and certainty inversely proportional to evidence, perhaps.
Imprints of others often are more beneficial.
With a lot of money spent on interconnects, little may remain for the real McCoy. (some fancy cheap as sheit off an industrial reel)
Claims made with plausibility and certainty inversely proportional to evidence, perhaps.
I know what I hear and there is nothing I need to prove to anyone.
- Status
- Not open for further replies.