Well... everyone must have been home today... lot's of distortion and noise on the channel... the CMRR must not be working??
I personally never have used the term "micro-dynamics" but I'd assume that it refers to small variations/sounds/events in the musical soundfield that ride on the "first order envelope of sound"... aka subtleties??
Small deviations in amplitude start to measure like noise riding on the signal, don't they? Noise does ride on the signal and not just remain like mud at the bottom of a clear pond, or does it work some other way?
So - anyone want to raise their hand for feedforward as a possible alternative/augmentation to negative feedback, given that chips are relatively inexpensive We really don't care about doubling the cost of a chip given the potential benefits, do we?
(btw, dogshome, best read back and find out what is being said?)
I personally never have used the term "micro-dynamics" but I'd assume that it refers to small variations/sounds/events in the musical soundfield that ride on the "first order envelope of sound"... aka subtleties??
Small deviations in amplitude start to measure like noise riding on the signal, don't they? Noise does ride on the signal and not just remain like mud at the bottom of a clear pond, or does it work some other way?
So - anyone want to raise their hand for feedforward as a possible alternative/augmentation to negative feedback, given that chips are relatively inexpensive We really don't care about doubling the cost of a chip given the potential benefits, do we?
(btw, dogshome, best read back and find out what is being said?)
The analogy is good only up to a point.
It fails in important ways...
The assumption here is that the speed a car goes is a complete description of "performance". It's not. Braking, cornering, temperature rise, tread life, road noise(!) and other factors all matter - especially depending on the car, the way it is driven and where it is driven.
Within the constraint of "normal" driving speeds - being the limit at highway speeds of 85mph-90mph - the requirements for say motocross tires are at variance with the typical commute.
You are suggesting that the bandwidth limit in effect sets the complete limitation for the performance requirements for amplification in all regards?
Feedforward compensation is already used internally on many op amps, are you suggesting something externally configurable by the user in the same way that negative feedback is controllable by the user?
No one said you did. Micro-dynamics is T&M blasphemy that I learned from Steve Hoffman's forum.
This is what I said,
Compensation?
Talking error correction.
Not sure what ur referring to.
Did you ever read my last PM, btw?
Talking error correction.
Not sure what ur referring to.
Did you ever read my last PM, btw?
Hmm not sure how you inferred that from my post. To explain my previous statement about bandwidth being related to the maximum rate of change of the op amp output voltage: consider that you have an op amp selected because of its awesome 1000 V/us slew rate. But will you ever achieve that slew rate in your circuit? Let's do some back-of-the-envelope calcs:
The bandwidth required for a certain rise time in a single pole system is:
BW = 0.35 / Tr . What's the bandwidth required to take advantage of that 1000 V/us slew rate at regular audio line levels (2Vrms)? Well 2Vrms is 2.828Vp. Theoretically, your op amp could do a 2.828V step change in: 2.828 / 1000 V/us = 2.828ns. Awesome! Micro-transients abound right?!
Well, to achieve this would require the circuit itself to have a bandwidth of: 0.35 / 2.828ns = 123.8 MHz.
I hope you don't have any EMI filtering on the input or outputs otherwise you'll never use all that wonderful slew rate. I also hope the microphone recording the music had 123 MHz of bandwidth, and it was amplified by a really wide bandwidth preamp, and then digitized by an ADC sampling at 250MSPS.
Point being: the audio signal path is always band limited to frequencies far below the 100MHz region, and for good reason. It solves all sorts of problems with noise, EMI/RFI ingress, etc. Hence why an op amp which provides 1000 V/us slew rate is really of little benefit in the signal path.
Notice that the LT1364 itself only has a gain bandwidth of 70 MHz. So you will only achieve 1000 V/us slew for step sizes which cause a rise time greater than 0.35 / 70MHz = 5ns. 5ns * 1000 V/us = 5V. This assumes a gain of 1. But if you put it in gain greater then one (which will limit the bandwidth even further) then it would require an even larger step.
I never said you said!?!
I merely attempted to put some sort of definition on it.
Based upon a presumption, then I added the somewhat important thing about the noise riding on the waveform... does that happen?
And BTW, I have no clue who Steve Hoffman is.
Actually, i've not yet deciphered what "T&M" refers to either.
_-_-
PS. I never used the phrase "enveloping the listener" WRT sound... fyi.
That's your internal interpretation...
T&M - Test and measurement. Steve Hoffman - SteveHoffman.TV – Home of Audiophile Mastering Engineer Steve Hoffman
You are right. Sorry.
Yes, quite so, and this is the "argument" put forth.
Also what I said was on the basis of the conclusion gleaned from your prior post - if you meant something else, sorry.
However, I'm not the one pushing for high slew rate and/or ultra wide bandwidth! Not against it either. No position.
The question may lie more in things like settling time and overshoot? Dunno. Does settling time effect distortion measurement? This I do not know.
What I am suggesting is that possibly the usual metrics either do not test something or perhaps that the combination of metrics is not made in such as way as to yield a (for lack of a better phrase) "figure of merit"??
The problem I have is that they're just not all the same in practice.
Just like some tires "feel good" on your car, and some don't. There's absolutely no tests I am aware of that would let you know that before a given model or brand of tire is tried out on your car, by you.
Perhaps there are no "micro-dynamics" per se. However there well may be the human perception that one might perceive and then use that name to describe it. Being too literal here would lead one to merely dismiss the subjective effect as being non-existent and so ignore it - when something may well be going on.
Separate PS per stage op-amp?
Ok... so it's been said that there is "no benefit".
But maybe there is...
...the other point that was made is that no one had made a 5534 (or anything else I know of) with a "beefed up output stage" (I think that was what was said? And of course there's no commercial market millions, or hundreds of thousands, no no big line mfr would make it.
But why not?
Why not a chip that is everything up to the driver, you put ur own output stage on the end?
Why not a chip with a mosfet output stage?
Why not a chip with pins for adjusting the bias? (think this was just mentioned)
etc...
And of course that would imply bringing out the PS pins for the output stage...
Not terribly practical for production - but in our ivory tower, we're free to blue sky to the day when DIYers will have access to custom fab, like we have access to PCB and front panel makers.
Ok... so it's been said that there is "no benefit".
But maybe there is...
...the other point that was made is that no one had made a 5534 (or anything else I know of) with a "beefed up output stage" (I think that was what was said? And of course there's no commercial market millions, or hundreds of thousands, no no big line mfr would make it.
But why not?
Why not a chip that is everything up to the driver, you put ur own output stage on the end?
Why not a chip with a mosfet output stage?
Why not a chip with pins for adjusting the bias? (think this was just mentioned)
etc...
And of course that would imply bringing out the PS pins for the output stage...
Not terribly practical for production - but in our ivory tower, we're free to blue sky to the day when DIYers will have access to custom fab, like we have access to PCB and front panel makers.
OMG that first example in the BUF datasheet is so easy it's irresistible.... where's my breadboard....
Here is the first time I learned of that term. It's all soul! from the a - - - - hole
to the soulless 🙄"What are Micro Dynamics in music? Beatles Tony Sheridan Polydor era vs. Parlophone..
There is a thread around here about why the Beatles recordings on Polydor before they were famous sound so good. Can someone provide a link to it?
There was a question that I was sent today about Transients and Dynamics in music. Mainly about "Micro Dynamics" and what that phrase means.
Well, it's been around a long, long time and essentially if you want to hear dynamics and especially micro-dynamics in recorded music, listen to the stereo Beatles recordings that came out on "THE BEATLES IN THE BEGINNING" on Polydor.
Those recordings are wonderful and have the micro dynamics that make music sound alive. The EMI recordings by contrast (as much as I love the songs) have NO micro dynamics at all. Not any, dudes.
Compare the two recording styles and you'll understand what micro dynamics are and how important they are to making recorded music sound alive.
Forget the BEAR FAMILY Tony Sheridan version, my copy has been stripped of some dynamics by mini-brickwalling.
Try the $1.99 LP on US Polydor from the 1970s that I mentioned above or the CD that came out in the 1980s...
Think of transients or micro dynamics (little quick varying degrees of soft to loud, etc.) as part of your moving index finger. Hold your hand out in front of you and move your finger up and down, and the little areas in between as quick as you can. Micro Dynamics. Now, stick your hand in a vat of sludge (go on, do it!). Try and move your finger in the little increments like before. Can't do it. Oh, you can still move your finger up and down but the little moves in between are impossible to do, just like the recording system at EMI/Abbey Road.. Heh, not that this is a terrible thing, after all, phonograph records were for playing on cheap systems back then. But when you hear the micro dynamics on a record like "CRY FOR A SHADOW" in stereo you realize what you are missing when you don't hear them. You also realize that your ears are SENSITIVE TO MICRO DYNAMICS, even totally untrained non-audiophile ears..
Of course the more sensitive your stereo system is, the better micro dynamics will resolve but they are clearly audible on average systems as long as you have some resolving power and a neutral playback and the amp is not driven to clipping point.
Over to you.."
It is intriguing, I accidentally discovered on a proto board.
Small signal common source jfet identical 3 stages , with odd/even number test.
Even # two stage much lower thd than one or three I assume from pre distortion.
A more exotic method would be a correction table laser trim during probe.
I sure many ways to accomplish without global feedback.
Probably patent search woud find good expired designs.
fyi GM uses a self correcting lookup table when engine is in open loop mode.
in steady state the table is recalibrated conditionally.
Maybe a set of polynomials for same class of parts?
-
I am aware of these concepts. I have no answer that will please you. Coming from just a tech school training and learning the rest on the job I can only offer you this to ponder: what is the difference between a discrete op amp (basically excepted by the audiophile world) and an integrated op amp (scorned by the audiophile world)? It's not the semiconductors. It's the passives! I'm guessing that the higher speed op amp shifts the sonic signature of the passive part of the IC up in a higher register so that it is not heard so easily. This is, of course, with my sonic impressions favoring video op amps being taken as a given.
I disagree that an NE5534 with a beefed up output stage doesn't make financial sense, since that's kind of what we made with the OPA1622. And it has sold plenty 😀
- Status
- Not open for further replies.