You know what is noise gain divider? If yes, you can imagine the circuit.
I am not going to debate. Everything important is written in the image.
I am not going to debate. Everything important is written in the image.
janneman said:
"Those who cannot learn from history are doomed to repeat it." George Santayana
Perhaps you don't recall (or were too young to be aware of) the scorn and derision poured on those who insisted that differences in passive parts were audible. "One uF is as good as another! Don't bother me with stupid, unfounded claims. You can't prove what you're saying." Yadda, yadda, yadda. Exactly the same crud spouted by so many here when someone brings up a claim for audibility. Only...now that it's been proven, you take parts variability as a given. But if no one had insisted that there was a difference, do you think the differences would have been looked for? Not anytime soon. You ignore the very basis of science--someone notices something unexpected, then forms a hypothesis which may, if all goes well, be upgraded to a theory later on.
To insist on a theory--more particularly, one that is palatable to you--is backwards. Sometimes the proof is counterintuitive or just downright weird. Remember Einstein's rejection of "spooky action through a distance" when he could not accept the implications of his own work? Yet the spooky action is actually the case, whether he liked it or not. Sometimes the world does things that are inconvenient for our personal belief systems.
I've said it before, and I'll say it again, the only intellectually honest options are:
1) I have have/have not heard the item or process in question.
2) I did/did not hear a difference.
3) I thought the difference (if one was heard) was for the better/worse.
Those are the only intellectually honest options. Assertions that the effect is impossible are not valid (who knows, it might be the equivalent of spooky action through a distance). Assertions that the person in question is selling snake oil may or may not be true, but I'll note that I just read an article online that an Oriental mixture of five herbs long regarded as old wives' tales B.S. has turned out to have real benefit for people suffering from eczema. Okay, so how many years have "knowledgeable scientists" contemptuously ignored this remedy? Were they right to do so? Would they not have been better off saying simply, "I have not tested that remedy yet. I have no opinion as to its efficacy."
People in audio who demand a widely accepted theory before they will even listen to things, and who mock (the phrase "snake oil" is a dead giveaway) those presenting the idea, have the reverse effect of the one they intend to evoke. Instead of demonstrating their expertise, knowledge, and ability to see through the thin veneer of sales patter, they demonstrate a complete ignorance of history and proper scientific process. Sometimes you even have to go forward with (oh, the horror!) no working hypothesis at all. Gasp! Ride bareback, as it were, in the hope that a proper hypothesis will arise in due course.
For the record: I have never heard the Bybee gizmos. I have no opinion as to whether they work or not. Period.
I don't care what Bybee's hypothesis/theory is, as I am pragmatic. If something works, it works, regardless of what fancy words are woven around it. If it doesn't work, it doesn't work, regardless of what fancy words are woven around it. I suggest that anyone who has pretensions of approaching audio in a scientific manner adopt a similar attitude.
Grey
P.S.: As a concrete example of something that did not work for me, I offer the VPI Magic Brick. For those who never saw one, it was a heavy wooden block roughly the size of two packs of cigarettes stacked together. The weight caused many bricks to be dropped and the wooden case was thin and easily broken. The 'secret' core was soon revealed to be laminated steel exactly as you would find in an EI transformer. The block was intended to be placed above the power transformer, whereby it would attract stray flux and keep said flux from washing over the active circuitry, causing mischief.
In spite of hearing many Magic Bricks in many systems, I never heard the claimed effects.
Possibilities:
--My hearing is not good enough to hear the difference, or I was listening for the wrong things.
--The systems (including my own) were simply not of enough resolution to reveal the effect.
--The effect was never there in the first place.
I have never, ever said anything beyond, "I did not hear a difference." Period.
And that's in spite of a pretty plausible hypothesis being advanced as to how they worked.
I will note in passing that VPI does not appear to sell the Magic Brick at this time.
CC source
Hi John,
Adcom used the J555 JFET 2 mA current source from op amp output to the -dc rail for the CC in their preamps, for both the AD711 line stage and LT1028C phono stage, if that helps.
Best, Chuck Hansen
john curl said:
Hi John,
Adcom used the J555 JFET 2 mA current source from op amp output to the -dc rail for the CC in their preamps, for both the AD711 line stage and LT1028C phono stage, if that helps.
Best, Chuck Hansen
syn08 said:
Noise gain is a known concept. Using high noise gain in an opamp circuit lets you run the opamp at unity gain for signals but with a very much higher noise gain. This means that if for example you have 80db noise gain all non-linearities are increased by 80dB. So, you measue THD, and then subtract 80dB to get the *real* distortion. Right, Pavel?
Jan Didden
Attachments
GRollins said:
Grey,
I'm afraid it's a waste of time explaining this again, but you are debating the very foundation of the scientific knowledge and approach. Please note that science is not a democracy; you are not innocent/right until you are proven guilty/wrong, but the other way around.
Ok, I'll live, it took me only 10 seconds 🙂
Grey,
I will not quote you verbatim, but apparently you still haven't grasped my philosophy. I never said that something is impossible because it hasn't been explaned yet. I only said that when something is implausible or goes against known/eperienced events, I will not run after it. It would be up to whoever makes the claim to prove that he is right. Time and time again extraordinary claims are made without a shred of proof. And extraordinary claims require extraordinary prove. That I feel is the only rational and intellectually honest option.
What has been proven, then? Where is it documented?
Jan Didden
I will not quote you verbatim, but apparently you still haven't grasped my philosophy. I never said that something is impossible because it hasn't been explaned yet. I only said that when something is implausible or goes against known/eperienced events, I will not run after it. It would be up to whoever makes the claim to prove that he is right. Time and time again extraordinary claims are made without a shred of proof. And extraordinary claims require extraordinary prove. That I feel is the only rational and intellectually honest option.
GRollins said:
What has been proven, then? Where is it documented?
Jan Didden
janneman said:
That's about what I was assuming - but I was thinking of using reactive networks to reach the same effect. However, in my setup, noise is the limiting factor, so why would I want to add so much more noise to my signal? And aren't you making the underlying assumptions of non-linearities being linear with the gain?
syn08 said:
Well, 80dB may indeed be too much in some cases, depending on the circuit noise. You could try it, if there are still clear harmonics above the noise floor it is OK. But even with 40dB noise gain you have effectively made your measurements 40dB more sensitive, so you can measure a -140dB amp with a -100db resolution analyzer. Worthwhile, I would think.
The assumption is indeed that the non-linearities scale with the noise gain. In fact, what happens is that you reduce the available open loop gain for the feedback factor with the difference between signal gain and noise gain: the circuit has 40dB less feedback to reduce the non-linearities. I don't know if the scaling is exactly 100% though. Propably close enough for government work 😉
Jan Didden
Grey, you make sense to me, but what do I know? Let them remain ignorant. It should be easier that way, especially on this thread.
Hi John,
That does not seem like a particularly helpful response. If you want to tag team with Grey, fine. Just remember to show some respect for those who do have an open mind for things with some proof.
I think the other members have been fair in their request for information. A statement without any evidence that can be checked by another in the field will not be accepted. This is perfectly reasonable - otherwise we'd be doing all kinds of very silly things to our audio equipment. Heck - just look at the fringe factor that does exactly that. One phase for you: "Green marker improves the bass response of CD's" This came back in the form of green LEDs in the CD bay. Cold fusion anyone?
I'm not lumping your ideas in with the obvious crazy ones. They were only to illustrate what happens when idea are not examined in an intelligent fashion. If you can't offer proof, then don't expect others to agree blindly with you. Also realize that you are not under attack Per Se. However, your ideas will be if you push an unproven concept without providing what amounts to good reasons for your belief. No amount of rhetoric will pass in a scientific community with out some kind of accepted experimental proof.
-Chris
That does not seem like a particularly helpful response. If you want to tag team with Grey, fine. Just remember to show some respect for those who do have an open mind for things with some proof.
I think the other members have been fair in their request for information. A statement without any evidence that can be checked by another in the field will not be accepted. This is perfectly reasonable - otherwise we'd be doing all kinds of very silly things to our audio equipment. Heck - just look at the fringe factor that does exactly that. One phase for you: "Green marker improves the bass response of CD's" This came back in the form of green LEDs in the CD bay. Cold fusion anyone?
I'm not lumping your ideas in with the obvious crazy ones. They were only to illustrate what happens when idea are not examined in an intelligent fashion. If you can't offer proof, then don't expect others to agree blindly with you. Also realize that you are not under attack Per Se. However, your ideas will be if you push an unproven concept without providing what amounts to good reasons for your belief. No amount of rhetoric will pass in a scientific community with out some kind of accepted experimental proof.
-Chris
janneman said:
Well, regarding explanation, yes, regarding circuits, still different.
In my case, Aol nonlinearity is well seen due to feedback reduction by 80dB. The circuit is inverting, and every input see 10 ohm. Voltage noise is increased by 80dB as well (LT1028 and AD797 both have very low voltage noise, for OPA134 the noise floor is 15 dB higher), AND also sensitivity of input stage to modulation is disclosed, which is interesting. For most of BJT input opamps, you can see 15625Hz Tv line frequency demodulated, never for JFET input opamps. The fast CFB opamps perform well. My favorite is AD844. I do not care about a bit higher THD in this circuit, as far as it is only 2nd and 3rd.
Attachments
This was the LM49710. The test setup is completely equal to that of 797 and 844. What you see are mains line components and detection of USB packets (10Hz approx. and 1kHz). The amplitude strongly depends on distance of the USB cable, but nothing like this you see for 1028, 797 and 844, also nothing like this for JFET input opamps - 134, 627, 827, 825, 744.
The circuit does not oscillate.
Any thoughts?
/The same problem is for uA741, OPA211/
The circuit does not oscillate.
Any thoughts?
/The same problem is for uA741, OPA211/
Attachments
Re: CC source
Also, Linear Technology use 2mA to the negative rail in their RIAA application suggestion in their LT1115 data sheet.
G.
chascode said:
Also, Linear Technology use 2mA to the negative rail in their RIAA application suggestion in their LT1115 data sheet.
G.
PMA said:
Ahhh yes that 15625, there was a time I could *hear* it, I could hear a TV running out of sync by the change in the line frequency. Too long ago....
Anyway, I agree with you that the AD844 is a well-kept secret for audio. I sometimes even use it open loop, with quite good results!
Jan Didden
PMA said:
Hmm, yes, funny. I have no explanation to offer. Maybe some giveaway in the data sheet? I'll look at it.
Jan Didden
Hi, PMA,
Are the impedances of +input and -input the same? (Symmetrical input / same component count and value for +input and -input)?
Are the impedances of +input and -input the same? (Symmetrical input / same component count and value for +input and -input)?
anatech said:
Me too John, you know I've been open minded but just because a garbage electrolytic has measureable and probably audible degradation does not mean there is some 'as yet undiscovered quantum mechanical effect' that makes Caddock and Vishay precision resistors sound 'totally' different in an input attenuator.
It would probably take all the reference cells and Kelvin-Varley dividers at NIST to find even parts per billion difference between the two.
I sit down at lunch several times a week and talk to my physicist friends about far out stuff like the folks at Harvard that have slowed light to a crawl, or the implications of the single photon double slit experiments. Both of these are not very intuitve (at least to me). I just don't think this stuff (or something like it) applies to amplifier design.
- Status
- Not open for further replies.