Serious question:
Given 2 opamps in similar circuits, where their frequency response measures nearly flat from 20Hz to 100kHz; the noise with no input signal is similar; the load is resistive and light; and where when the input is a musical program, one circuit sounds "bright" and the other one "veiled"; which mechanisms could possibly:
-make one circuit sound "brighter" than the reference
-make one circuit sound more "veiled" than the reference
I cannot think of any that would be producing the second effect, please enlighten me.
Given 2 opamps in similar circuits, where their frequency response measures nearly flat from 20Hz to 100kHz; the noise with no input signal is similar; the load is resistive and light; and where when the input is a musical program, one circuit sounds "bright" and the other one "veiled"; which mechanisms could possibly:
-make one circuit sound "brighter" than the reference
-make one circuit sound more "veiled" than the reference
I cannot think of any that would be producing the second effect, please enlighten me.
There is no reason any more why one whould use the OPA134. OK, it comes in DIL so we can solder it better by hand. I use it for servo when i make hand samples.
The OPA1641 is better i think. Much lower distortion in the treble and very stable input capaciatance. That is important for transfer linearity when inverted circuitry is used.
It should also have less common mode distortion and it has a very linear output stage.
With one of these you can make an awesome MM phono that is hard to beat.
I use a 2 stage topology though out of other reasons.
The OPA1641 is better i think. Much lower distortion in the treble and very stable input capaciatance. That is important for transfer linearity when inverted circuitry is used.
It should also have less common mode distortion and it has a very linear output stage.
With one of these you can make an awesome MM phono that is hard to beat.
I use a 2 stage topology though out of other reasons.
Attachments
Last edited:
let's assume some conditions like the input signal is bandwidth limited to 40kHz, its max amplitude doesn't drive any of the circuits anywhere near saturation / clipping etc. in other words some decently designed circuits where we don't saturate / clip / slew limit etc.
The more you limit the signal in the treble the less slew problems you have. That is a case of " everything sounds they same ". I think the QUAD philosophy is that way.
An extreme case is Quad Germany, the guys that still make the original equipment.
The owner said to me " For me an amplifier is just a case with knobs."
An extreme case is Quad Germany, the guys that still make the original equipment.
The owner said to me " For me an amplifier is just a case with knobs."
But then, the real culprit would be the power supply? which one of the amps manages to mask somehow better due to a higher immunity to its PS shenanigans?
Power supply design is not easy and i am no expert on that. It is not only the rejection of humm and hiss but also a PSU can have distortion.
I think Jan Didden and friends made a subjective comparison where the low distortion types came out first.
Take a cheap CD player and put a cable between the PSU and your amplifier.
You will hear the music extremely distorted.
I think Jan Didden and friends made a subjective comparison where the low distortion types came out first.
Take a cheap CD player and put a cable between the PSU and your amplifier.
You will hear the music extremely distorted.
Last edited:
My current best hypothesis is its down to power supply noise in the main. Power supply noise not originating from the regulators but created by the operation of the output stage being classAB. It follows that the noise is strongly correlated with the signal and has a spectrum more weighted towards HF than the signal itself and dependent on three aspects of the total circuit :
a) the impedance of the power supply
b) the impedance of the load
c) the PSRR of the opamp
Joachim / Abraxalito - thank you for pointing the PS.
I gave it some more thought and I still can't figure how it could "veil" the sound without this also showing in the frequency response. Some intricate form of IM?
OTOH "bright" is pretty straightforward to produce through harmonics.
I gave it some more thought and I still can't figure how it could "veil" the sound without this also showing in the frequency response. Some intricate form of IM?
OTOH "bright" is pretty straightforward to produce through harmonics.
This can be tricky. If it's high gain (like 20db or more) then all passives around the opamp play a larger roll than you think. Also you don't say what two opamps or if they have the same type of power supply. Is one op amp much higher slew rate than the other? If it is a non inverting circuit and a stopper resister is used then was the same kind and value of stopper resistor used on each?
I'm unclear why the IM would need to be 'intricate' ? Simply the addition of HF-weighted correlated noise gives the impression of HF roll-off even with a flat frequency response in my experience.
Would this be "sound masking" like described here?
https://en.m.wikipedia.org/wiki/Sound_masking
I'm afraid a large quantity of such noise needs to be generated. Back to our opamps, if they would be doing it to such extents, it would be trivial to measure?
Let's assume something simple, like line level volume / tone control, opamps work at low gains or as buffer, the supporting circuits are identical and we're comparing apples with apples i.e. both opamps are of same type bipolar or fet (as to not screw everything with a huge difference in input impedance).
I always enjoy trying to explain to SY the goodness of a meat dish, in this case the wonderful pig stomach at dimsum. This visit was particularly enjoyable thanks to all the good company.
I don't want to assume anything. You didn't actually build this? This was all a hypothetical exercise?
No, it was this other gentleman who stated that replacing opamps in his Arcam removed a thick veil.
I don't contest that this may actually have happened.
The question is - how can one know which of the 2 sounds was the correct one?
Oh ok. The one with more top will be the correct one. Then you get to hear the noisy rectifier diodes and cheap resistors. New territory to suss out the gremlins.
- Status
- Not open for further replies.