It may even have been referenced in this thread earlier. But I'm afraid my comments are only speculations, and James Boyk's don't extend into the causal (quite properly, IMO). I'm simply struck by the correlation between some topologies or active devices and consistently good reviews for sound quality, even when the active device has other difficult issues. Vacuum valves have no minority carrier complements, require coupling capacitors, etc. for example.
Thanks for any help you might give me,
Chris
+1
I seem obvious that the two ways to decrease distortion in a closed loop system is
- To reduce open loop distortion.
- To increase feedback ratio.
It seem reasonable too, for an audio OPA, to ask its open-loop bandwitch to be at least 20 000 Hz.
Now, for the feedback, its ability to reduce high order distortions is directly connected with the speed of the OPA.
Reason why i FIRST look at slew rates and will never use a <100V/µs high gain OPA. I said that previously, my doctor said it is normal at my age to repeat in circle the same things.
"0.55 pV/Hz1/2, more than
three orders of magnitude lower than would be possible with
the best conventional room-temperature voltage amplifier."
Great, now we gotta do all our recordings in a giant dewar flask.
Pavel,
I am just starting on rail noise out of band vs increased in band noise. As I am 2 years into this give me 6 months to show results.
But here is what I think is an interesting curve. I'll entertain guesses as to what it shows.
But my thanks to the folks who looked at the power supply question. There is no right answer, yet.
Attachments
So you want for me to repeat the test with CCIF 13+14kHz (19+20kHz)? It is not impossible. It will not change much, only OP27 and uA748 will disqualify completely. FFT 1kHz was good for the reason that a single measurement has shown SNR as well, and differences in SNR/Noise Level between the opamps tested.
I think in a way the paper was something of a personal breakthrough for Boyk, as he actually concedes that bipolars might work o.k. in some circumstances
I've probably related the story of a friend who took Jim's class many years ago, and was quite thoroughly indoctrinated into the Way of Tubes. He also developed Fear of Feedback, and worships at the Temple of Maximal Simplicity.
When I sent him the Boyk and Sussman paper I expected some significant reaction, one way or another, and was puzzled when it didn't come. When I asked him about it he said he'd read it, but it seemed to just pertain to IM and he wasn't sure about the conclusions. Meanwhile, I've come to conclude that for him Boyk has fallen from grace, to be replaced with more fanatically subjectivist designers like Peter Qvortrup. Perhaps that's unfair, but a working hypothesis.
To your remark about tubes, one observation is that, without a whole lot of other complications they foster balanced designs to reduce second, and/or the use of transformers*, and along for the ride with such designs we tend to get equality of positive and negative slew rate and settling behavior. As well the thermal mass involved reduces any audio-band shifts with signal history, and the interelectrode capacitances don't change a lot with operating point. Of course they fall apart under sufficiently heavy drive too --- you need that space-charge layer more-or-less intact.
Brad
*which of course have their own set of issues
I used the OPA627 along with a neodymium instrument speaker to modify an old Polytone guitar amp for a jazz guitar player friend. I used the OPA627 to boost the signal coming off of the EQ. This opamp was so clean that it allowed me to insert a beeswax in parallel with mica cap combination in series between the EQ and the OPA627. The compression of the beeswax cap and the hi-frequency "zing" of the mica cap with such clean amplication were true magic. My jazz guitar player friend said that he had not heard as nice of a sound even on boutique amps costing upwards of $5k, tube or solid state.
John and others; a question about IC's. Like a number of readers (rather than participants) in this thread I am out of my depth with the more technical issues but like the idea of gaining insights so as to modify my own system. You have mentioned the AD825 favorably a couple of times. My question concerns "settling time". About 40 yrs ago Gordon King and JLH in the UK were advocating a low settling time as an important criteria for a good subject response from power amps.
I notice the AD825 DOES have a very low settling time compared to a lot of other IC op'amps.
Has anyone got any thoughts on this as an indication of good subjective sound......?
And if so why.........?
I notice the AD825 DOES have a very low settling time compared to a lot of other IC op'amps.
Has anyone got any thoughts on this as an indication of good subjective sound......?
And if so why.........?
Thanks for going into the trouble jn
Still in the dark but it’s not because of you.
Thanks for sharing Pavel.
What is the output level in these measurements and on what load?
Any 5532/5534 on your bench?
George
Not the point, I assume the experimenters were at room temperature. They were able to build an experiment that had <-200dB line contamination, in fact .55pV is almost -246dBV!
JC, how about measuring or letting someone measure Blowtorch to see how much 7th it has under comparable conditions to AD797 etc. I'm sure Pavel will be happy to do so.
Much more convincing than 'pontificating on your 'calculations'. Have you checked your calculations against Baxandall? He actually measures stuff to confirm his measurements of exactly this effect.
Would you like to tell us how with all this supa dupa experience & superior knowledge, you ended up with the original JC2 Mk1 circuit that was caned by the reviewers?
Last edited:
- Status
- Not open for further replies.