Could you expand on the BOLD comment above?
And, pls clarify the data recovery between audio and data transports if both used for audio.... which is better and why again? I would think the computer transport could stand No error while audio could take some interpolation if data lost and still sound OK. Also, is using the data transport more reliable or not?
THx-RNMarsh
Esperado should only have mentioned from the start that his expressed experience was from around 1970, the same years that the attachment was printed, yes 741s
George
>Edit. For a unity gain mixer, the problem does not really exist
The popular replacement TL07x was often used also .......
![O2 V11 AC Pwr Gain Stage Op Amps 2 V RMS THD+N vs Frequency BW=22 Khz See Legend_thumb[1].png](https://www.diyaudio.com/community/data/attachments/439/439747-36812ee5a7e07ac0192770716c4b6381.jpg "O2 V11 AC Pwr Gain Stage Op Amps 2 V RMS THD+N vs Frequency BW=22 Khz See Legend_thumb[1].png")
What happens to it at 10KHz with 30 HF tones and/or higher OP level? Could it then reach audibility threshold? Seems very possible.
In my experience with listening and correlating sound with distortions, when distortion makes the sound have less clarity in the treble, it is not unlike describing a freq response roll-off. It is As If it was rolled off. But, its actually, added distortion artifacts --- maybe creating some additional masking, as well.
THx-RNMarsh
Last edited:
I resend that. I did NOT shoot any messenger. I addressed technical details only. There is only one place where he mentioned 'allergy' and I responded with 'religious'. If you think I shot the messenger, you should delete 80% of all posts here.
Are we clear?
Jan
The uA741 was used in many professional studio boards in the 1970's. The 'excuse' was their low gain-bandwidth made them more immune to RFI. That is why we had to PROVE that the 741 was a lousy audio op amp. Yes, there were some better devices like the Harris 911, but each cost an hours pay for an engineer at the time, so only exceptional people like Dick Burwen used them. Mark Levinson made a really nice portable studio board from Burwen modules containing a Harris 911. I used them in my studio board design for the GD in 1970 as well. They were OK, but just OK. Discrete was better, and Mark had me design discrete replacements of all the Burwen modules for this reason.
The 741 was the POPULAR op amp at the time. Wonderful for servo design and such, I used them for that at Ampex in 1969. Wonderfully easy to use.
The 741 was the POPULAR op amp at the time. Wonderful for servo design and such, I used them for that at Ampex in 1969. Wonderfully easy to use.
Muting a slice is not the same than affecting this slice to one or several mixing bus.
Last edited:
Dick,
In a very simple experiment distortion was perceived at a level of -80 db.
The method was simple a 3,000 hertz oscillator was mixed with a 60 hertz one. The high tone was set so that it was clearly perceived above the noise floor in my shop from a commercial loudspeaker. The voltage was measured on the loudspeaker line. It was .003 volts RMS (Fluke 87). The low frequency level was increased until the HF sound was masked the voltage now read 31.3 volts. (Muting the HF on and off made the difference easy to hear.)
Now with a quieter environment and better transducers I suspect you could resolve lower levels.
I will check this with some other ears found around here.
ES
In a very simple experiment distortion was perceived at a level of -80 db.
The method was simple a 3,000 hertz oscillator was mixed with a 60 hertz one. The high tone was set so that it was clearly perceived above the noise floor in my shop from a commercial loudspeaker. The voltage was measured on the loudspeaker line. It was .003 volts RMS (Fluke 87). The low frequency level was increased until the HF sound was masked the voltage now read 31.3 volts. (Muting the HF on and off made the difference easy to hear.)
Now with a quieter environment and better transducers I suspect you could resolve lower levels.
I will check this with some other ears found around here.
ES
That notion of RF immunity is almost hilarious. I think it may have been Jung in an ADI book who wrote that, typically, bipolar input op amps detected RF about 100 times better than BiFET op amps.
But yes, compared to the 109, a lot easier to use. A very clever design.
When I had to incorporate a sort-of shelved crosstalk cancellation approach, invented by John Norris, into a computer audio three-piece system, initially sold by Harman to Dell, I realized that S/N was going to be a challenge. I wound up with a discrete design using some decent general-purpose devices from Toshiba and ran a fair amount of current through them, knowing that the soundcard output impedance was low and thus current noise not a big problem. The wholly-unexpected dividend was an immunity to RFI, including cellphones in close proximity, compared to other cheap systems Harman made---this without common-mode chokes or series-resonant caps across various inputs. The scheme for transitioning smoothly between stereo and Norris quasi-surround was patented within the time window after the product was shipping---although in an example of corporate business shortsightedness, Norris' patent for his basic matrix was abandoned as soon as it required some response to initial USPTO action.
Better in general Faster Slew Rate, Higher bandwidth, less harmonic distortion.
Last edited:
Yes. But i remember a big desk (Freevox), using solid state in a strange way: the active circuits were sealed in a tube like (with the same plugs) opaque enclosure. It was using 741, if i remember, but with an output buffer in discrete components included in the feedback loop. This desk was not so bad t its time.
More connecting dots to solidify the threshold number(s).
Thx-RNMarsh
Repeat with a second set of speakers playing music in the same room.
Of course damn near anything beats the 741 for audio, except perhaps the really sluggish LM124/LM358.
Oddly, TI still has preserved the utter nonsense that the RC4558, certainly no prize either, is essentially a dual 741. I just checked the revised datasheet, updated fairly recently but originally from 1976.
Note that the distortion stated for the TL072 is for an input impedance of <1k. Although it is possible to alleviate the common-mode distortion mostly due to the input devices' gate-to-substrate voltage-dependent capacitance somewhat, by balancing impedances seen by each input terminal, it's only effective to a point. Jung and others have written about substrate bootstrapping (the substrate is the negative power supply input) which is a good deal more effective.
They are cheap at least. For more money, dielectrically isolated parts are a lot better.
well, that is what I did to determine my own delta detection threshold number -- used music. .05% thd+n.
That was added thd in the form of OPS cross-over distortion via bias manipulation.
The TL07x under the conditions talking here about might exceed my detection threshold. Seems to have exceeded Esperado's threshold.
It is also quit likely the replacement CFA had much lower distortion at high freq as that is easy to do with CFA. Not saying changing to a better VFA would not have also been heard as better. Just saying.
THx-RNMarsh
Last edited:
- Status
- Not open for further replies.