No one has said that they "don't believe that there could be any differences in this test." Perhaps it might be worthwhile reviewing what was actually written?
Well that was fun -
I said I thought the differences I heard every time I compared was Y - which is a no load condition. This isnt a guess for me. Nor luck. If you have tested/measured IC opamps before and listened to distortion before of loaded stages -- that is what it sounds like. The OPS in IC opamps is their sore spot. JC knows it. We all know it. When current thru OPS goes up, so does distortion. So whats all this dancing around for?
Anyone have a better opamp buffer circuit? We could use one for David's oscillator.
THx-RNMarsh
I said I thought the differences I heard every time I compared was Y - which is a no load condition. This isnt a guess for me. Nor luck. If you have tested/measured IC opamps before and listened to distortion before of loaded stages -- that is what it sounds like. The OPS in IC opamps is their sore spot. JC knows it. We all know it. When current thru OPS goes up, so does distortion. So whats all this dancing around for?
Anyone have a better opamp buffer circuit? We could use one for David's oscillator.
THx-RNMarsh
Last edited:
To what end? It's already trivial to design one that is audibly transparent. The contest for how many zeroes in a THD measurement is one that many find interesting; I am not one of them.
One hundred ohm loading is extreme. I have, however seen 300 ohm loading, and even that is too much, I think. Many other IC's that are more popular here are not designed for 600 ohm loading to the extent that when you load them with 600 ohms, they get very unhappy. 100 ohms would be overwhelming for typical IC's. The 4562 is a notable exception, actually. For example if you want to see what the 2134 does, just look at the Stereophile review of the JC-3.
consumer line level audio drivers are not limited to one op amp circuits
as I repeatedly point out composite op amp circuits: "good", "audio" front end (jfet?) monolithic op amp wrapping feedback around physically separately packaged high bias CFA DSL driver op amps can pretty much remove active device errors from consideration over the audio frequency range
passive components and layout issues are the bigger problems - Class A bias the output, separately sub regulate the input op amp supplies and you've knocked down another couple of implementation issues..
quit whining, drop the drama and just start building composites with best known practice today - if you want low, "unmeasureable" audio frequency distortion you can get there with 2 $5 chips in a loop
as I repeatedly point out composite op amp circuits: "good", "audio" front end (jfet?) monolithic op amp wrapping feedback around physically separately packaged high bias CFA DSL driver op amps can pretty much remove active device errors from consideration over the audio frequency range
passive components and layout issues are the bigger problems - Class A bias the output, separately sub regulate the input op amp supplies and you've knocked down another couple of implementation issues..
quit whining, drop the drama and just start building composites with best known practice today - if you want low, "unmeasureable" audio frequency distortion you can get there with 2 $5 chips in a loop
Last edited:
I am sure you can not do it even if you thought it was interesting.
It is for a lab test instrument. Not an audio amplifier.
But maybe some one else over here could do a suitable job?
JCX - Send to David your circuit schematic for him to try.... he needs the best you can do so that the prototype osc thd isnt compromised.
THx-RNMarsh
Last edited:
Well to be honest, with an opamp loaded almost out of it's specs, with clearly measureable problems, I had expected that everyone, no exception, would shrink back in horror from the X case. That it didn't quite went that way (one even reporting that X sounded most realistic) tells me that even gross problems are often not detected or interpreted wrong.
But of course, changing a resistor brand A for resistor brand B is quite another case - that's so clearly problematic that everybody can hear the difference
jan
But of course, changing a resistor brand A for resistor brand B is quite another case - that's so clearly problematic that everybody can hear the difference
Coincidentally, I did this last night. It was so obvious that my wife (who is not interested in audio) called out from the kitchen, "What did you do? It sounds better?"
Or maybe she said, "Turn that goddam thing down and go do something useful!"
It's sometimes hard to tell with her.
Thanks, I'm giving some thought to the best way to do this
The test has to be as simple as possible for those that just want to listen, and on the other hand we need the technical files for those that want to analyse.I have some ideas......
Dave?
Contact me via email (my initials at syclotron.com) and let's see what the contract would look like.
No accounting for what people like. Lacking experience in correlating what is heard with measured distortion, possibly. Those individuals dont get to be in on the next round of listening.
I would like to know the FFT of the 100 Ohm condition as it will indicate just how sensitive the listening test was. But if I can hear it thru my port pc speakers ... it must be really high, right? Right?
THx-RNMarsh
Dave as in DAVADA in Equip & Tools forum discussions. He needs your help. Contact him right away with your requirments..
My contract is simple. You get publicly honored and the all the glory you can stand. plus parts cost and $1 for being a good sport. Could be more - David decides.
THx-RNMarsh
Last edited:
- Status
- Not open for further replies.