Addressing John Curl's concerns on low noise designs

Status
Not open for further replies.
Jump to Latest
#44
It doesn't help that fairly tired old concepts like op-amps that have open-loop gain of 120dB at a 100Hz corner are "slow" are offered as technical criticism. Modern IC's have 100's of MHz unity gain crossover and 1000's of V/us slew rate. it is about time people stop quoting from articles complaining about 741's in audio.
 
#45
scott wurcer said:
It doesn't help that fairly tired old concepts like op-amps that have open-loop gain of 120dB at a 100Hz corner are "slow" are offered as technical criticism. Modern IC's have 100's of MHz unity gain crossover and 1000's of V/us slew rate. it is about time people stop quoting from articles complaining about 741's in audio.
Click to expand...

Very true.

Does the OPA552 meet those criteria?
 
#47
:cop:
Right now I think we need to keep everything on a professional level. Personal comments typically do nothing to advance any discussion. Any attempt to minimize the knowledge or experience of others show me a malicious intent.

Both Syn08 and John have a history of feuding. How about we keep these topics factual with ample evidence to back up any claims? Now that would be professional. I expect other members to consider what is a belief and provable fact before posting please.

John (Curl), one request from you please. It's fine that your design is kept out of the public eye. You have one at home, so when referring to it's performance, use actual measurements to debate a point at the very least. Comments without proof do not do anyone any good. It breaks down into "he said, she said". Useless.

In addition, referring to what you knew 25 years ago or so really doesn't have anything to do with anything I have seen discussed in your thread(s). It's your state of technology now that matters. The only reason I can see to mention what you did earlier is to re-enforce your history in the industry. We get it, it's well understood by all.

-Chris
:cop:
 
#49
scott wurcer said:
It has more than enough of both for this application. Would a discrete 2.4MHz gain of 10 buffer prompt the same comment?
Click to expand...

From me, yes.

I should be more specific.

This particular opamp has a stated DC supply current of about 7 mA. If you assume that *all* of that current is used in the output stage, at just what point does crossover distortion enter into the picture? This is not an AD797 with special circuitry to mitigate that problem. The opamp operates into less than a 20 Ohm load. If a moderately low output moving coil cartridge is used, think about the output from that opamp when a high frequency signal, emphasized by the RIAA equalization inherent in the recording, hits a maximum. Maybe that's not a limitation - I don't know.

That's the origin of the question. I guess it doesn't specifically call into question the slew rate or the open loop bandwidth, so my question may be somewhat badly phrased.

I'd ask the same question about a discrete stage.

I'd also point out that I was asking a question - nothing more.
 
#50
Scott,
>>Would a discrete 2.4MHz gain of 10 buffer prompt the same comment?<<

Not at all, if it had no loop NFB. But if it does, then all bets are off.

The chip in question has a GBP graph showing it starts to drop at around 10Hz. From my 30 years of experience designing and manufacturing high end phono stages, this is a recipe for classic op-amp high NFB sound.

Which some people seem to like, but which I do not! I made my first zero loop NFB designs around 1980, which were like a breath of fresh air to me and my clients. Every so often I try NFB again, and can put up with it for a day of two but then it can't be stood any longer.

My opinions of course, but I'd be very happy to put my RTP3C preamp up against ANYTHING ever made by anyone. Schematic available on my website under "Schematics".

Regards, Allen (Vacuum State)

www.vacuumstate.com
 
#51
Allen, I would like to hear your preamp, sometime. I have often found that well built tube amps and preamps can sound especially wonderful. You know what the tradeoffs are.
I decided to stick with solid state, but I have consulted on tube designs with Audible Illusions and MFA preamp designs.
 
#52
John,
maybe one day, you, me and our preamps can get together for a mutual listening/learning experience. i would find it most enjoyable. Did you look at the schematic?

http://www.vacuumstate.com/index.tpl?rubrik=8&lang=2&a=%0E%252C%1AG%D9%18%250BF&b=733898.4020199500

We have met, at your house around 1976 or so Ed Wodenjak brought me and we compared my (then) NFB SS preamp with your Levi pre. I seem to remember you were quite surprised. I even have a photo of this meeting.

Regards, Allen (Vacuum State)
 
#53
That's a long time ago, Allen, even for me. However, I have just looked at your schematics and am impressed. You have an advantage of only needing N channel jfets. They really are more linear at high frequencies, and quiet too! You know 'where the bear sits' just like Bear does. :nod:
 
#54
John,
we don't use jfets in the differential preamp any more. replacing the 2SK147/170/369's with MAT02 bipolars gave 10dB more gain, and no more noise with no other changes - and it sounds even better!

But other dual bipolars didn't sound good, the LM394 with almost identical specs to the MAT02 had people running from the room...

Regards, Allen (Vacuum State)
 
#60
Never had any problem with RF demod, and it coukldn't have been mexican food, simply because you can't get good mexican outside of the Americas.

The sonic difference between the MAT02 and LM394 was amazing, the 394 sounds harsh and very edgy, the MAT02 smooth yet very high resolution. Higher resolution that the jfets even, IME.

Regards, Allen (Vacuum State)
 
Status
Not open for further replies.
Top Bottom