For consumer level devices that is open to interpretation, as in the case of the LM4562 with popcorn noise. Did this issue lead to the dramatic price reduction, or is that a conspiracy theory, and in the quadrillions of devices out there how often has this been repeated? The opamp rollers hear what they hear, but it doesn't seem wise to make judgements unless the parts are qualified after listening, and nobody (few) has the gear to do it.
Last edited:
An interesting thread (if you have not seen it) on the LM4562 and related parts is here: National opamp inflation
Someone who might be able to answer LM4562 questions is Audioman54 but I see the last activity is: "Last Activity: 1st December 2020 07:44 PM"
Someone who might be able to answer LM4562 questions is Audioman54 but I see the last activity is: "Last Activity: 1st December 2020 07:44 PM"
I'm definitely not going to claim it's the best I've found, but for cheap, I'm having good luck using tips I fabricate into a hammerhead shark shape that spans 4 pins. (made by hammering out the end of a copper tip [that has been cutoff square] into a flat "platter" and then cut/file perpendicular to the tip shaft to create the hammerhead shark shape) Then flow enough solder on the hammerhead to ensure good thermal conduction to all 4 pins simultaneously. I use a jeweler's screwdriver ground concave on one of the flat sides to create a sharpened tip as a lifting tool. Then lift a single side via the end of the opamp while heating 4 pins (which slightly bends the pins on the non-heated side in the process). Then ensure no solder bridges from the lifted pins to the pads below. Once clear, apply the hammerhead to the other 4 pins and simply slide the opamp off the pads while grasping opamp with small forceps. I'm usually able to do a pull in 60-120 seconds. I had one near disaster with a pad from not getting good enough thermal contact on one of the pins, but recently modified the tip a little to ensure all 4 are fully contacted in the same portion of the pin bend...plus using a bit more solder on the hammerhead and not worrying about cleaning up some bridged pads after the pull. I've done about 150 pulls of 8 pin SOIC and all the pulls survived and only one pad separation that fortunately did not break from the trace, so was able to solder in a new opamp with no problems. That episode led to tweaking the tip shape for better assurance all 4 pins getting equal heat...
According to Audioman54 (the former National employee involved with LME development that you link in another post) as of 2009 when he posted his comment, 4562/49720/49860 are all the same die/fab. 49860 is tested/binned/labeled for higher voltage (much like "premium" cpu and gpu parts). Fast forward to today, at least my personal experience suggests that the more recent 4562 are 49720 rejects with respect to quiescent current. Mine measure that way anyway... And why else sell them for half of what a 49720 goes for when everyone knows they're "identical"?
As a side note, its interesting to read how a susbstantial part of the LME development/validation was...drum roll please...listening tests.
Last edited:
There are valid business reasons to do this that you may not have though of. Your sample size is too small to be significant, but anyway, they are still identical.
Didn't seem to serve them very well, considering TI killed most of the line. Even if you believe in subjective opinion, the golden ears don't seem to like LME49720 that much.
As many here know, the ear is much more sensitive to some types of distortion than others ...
My thinking is that the best op-amp should have very low distortion and low noise, along with the most inaudible distortion spectrum. This is why I like a high quality tube component and/or zero feedback solid state component in my system, that adds a bit of second harmonic distortion in there
I mean it is really about the spectrum and how pleasant or inaudible it is to our ears. Even very low distortion gear can be intolerable with the best recordings ... just because it has say more 5th harmonic than 2nd or 3rd.
Like, I've bought some digital chip amps that looked very good from an THD and noise perspective and found them just not listenable to me after a period of time. So, I look up the IC's data sheet and find most of the distortion is 5th or 7th, minimal 2nd and 3rd. I just thought, Ahhh ... there you go ... that's why you suck Mr. amp.
It's all about the overtones ...
Sure the ear creates distortion and noise ... but it is distortion and noise we are habituated to, for the most part. We don't 'hear' it. That is how the brain works ...
Seems to me the mechanical ear works somewhat like a cartridge needle following an LP ... something to think about?
My thinking is that the best op-amp should have very low distortion and low noise, along with the most inaudible distortion spectrum. This is why I like a high quality tube component and/or zero feedback solid state component in my system, that adds a bit of second harmonic distortion in there
I mean it is really about the spectrum and how pleasant or inaudible it is to our ears. Even very low distortion gear can be intolerable with the best recordings ... just because it has say more 5th harmonic than 2nd or 3rd.
Like, I've bought some digital chip amps that looked very good from an THD and noise perspective and found them just not listenable to me after a period of time. So, I look up the IC's data sheet and find most of the distortion is 5th or 7th, minimal 2nd and 3rd. I just thought, Ahhh ... there you go ... that's why you suck Mr. amp.
It's all about the overtones ...
Sure the ear creates distortion and noise ... but it is distortion and noise we are habituated to, for the most part. We don't 'hear' it. That is how the brain works ...
Seems to me the mechanical ear works somewhat like a cartridge needle following an LP ... something to think about?
Last edited:
LOL, valid business reasons that apparently you don't know given your lack of a meaningful response. As for sample size, about 150 from 5 different batches across ten years showing clear trends....
It served them very well, how old is 4562 at this point and its only now being supplanted? It cracks me up to see clowns like you droning on and on about how nobody can hear the difference yet you fret incessantly trying to improve measurements! Truly hilarious! But at least you've established that you cannot hear any differences. In fact I'm convinced that 741 vs latest measurement champ is no different to your ears, which begs the question "why measure"?
Last edited:
Skitron, your argument is a non sequitur: all fields of engineering are about making constant improvement -- constant progress. You cannot make progress in audio frequency amplifier design, for example, by relying on one of the most unreliable instruments, the human ear, as an arbiter of quality. Clearly, you are not an engineer, and if you had any formal training in the field, it was gravely wanting.
Last edited:
JBL uses trained listeners to evaluate the SQ of speakers. Of course they measure too. Using a panel of trained listeners can be a valid type of measurement technique. The human ear of one person may not be reliable, but a group of people may have statistical reliability. Particularly if they are trained and qualified.
Not at all: you cannot train anyone not to be fooled by their brain/ears since such mundane things as one's mood, state of nourishment, etc, have a disproportionately profound effect on one's preferences. Double-blind listening tests have shown that non-scientific listening tests are a waste of time and that the ears/brain cannot be relied upon to determine the quality of audio equipment. Please do some truly scientific research on the subject instead of wasting your time with anecdotal old wives' tales.
Last edited:
I thought I recognized the specifications of the LM4562 .
A good number of years ago I built a chip preamp using the LME49710 and following this thread which is quite "lively " I checked up on the LM4562 ---surprise --surprise , the specifications are the same even the circuit provided as an example -passively equalized RIAA phono preamp is identical in component values .
It won product of the year LM4562 ( 2007 ) ==Electronic products.com.
I am not trying to pick sides as I am, willing to listen to both sides of this argument as this is the only thing I can think of that I like to strike a balance in .
I can see the engineering point of view but all this effort in design engineering is aimed at--- organic human beings who have much varied auditory senses and while not relevant on this thread an awful lot of people like "tube sound " with its "melodic" F2 distortion .
Its not easy to "square the circle " and this argument has ranged nearly 50 years and not reached an
amicable agreement it certainly keeps audio websites active.
A good number of years ago I built a chip preamp using the LME49710 and following this thread which is quite "lively " I checked up on the LM4562 ---surprise --surprise , the specifications are the same even the circuit provided as an example -passively equalized RIAA phono preamp is identical in component values .
It won product of the year LM4562 ( 2007 ) ==Electronic products.com.
I am not trying to pick sides as I am, willing to listen to both sides of this argument as this is the only thing I can think of that I like to strike a balance in .
I can see the engineering point of view but all this effort in design engineering is aimed at--- organic human beings who have much varied auditory senses and while not relevant on this thread an awful lot of people like "tube sound " with its "melodic" F2 distortion .
Its not easy to "square the circle " and this argument has ranged nearly 50 years and not reached an
amicable agreement it certainly keeps audio websites active.
