nothing to 'explain away'
very simple heruistic modeling, a little math that that should be expected in 3rd year undergrad Analog EE programs
could probably draw a problem set question or two from the papers
with the National Instruments (Digilent) Analog Discovery module you even make up a Lab duplicating the Hirata work for a undergrad course
but quite crude by modern Nonlinear System Identification standards - try a google search for senior undergrad/intro grad level textbooks from the last decade or so
I only have to stand up to pick Pintelon, Schoukens IEEE Press 2001 "System Identification A Frequency Domain Approach" from my shelf
https://books.google.com/books?id=KhonXGwETWsC&printsec=frontcover#v=onepage&q&f=false
it only does very light Nonlinear Modeling - but quite sufficient for the weak, low order nonlinearites from competent audio electronics
I'm only able to follow the math in the 1st half, but that's the point - if needed I have a resource to learn from
also with many packages for matlab and free equivalents today even a general acquaintance with the math lets you run very sophisticated software
very simple heruistic modeling, a little math that that should be expected in 3rd year undergrad Analog EE programs
could probably draw a problem set question or two from the papers
with the National Instruments (Digilent) Analog Discovery module you even make up a Lab duplicating the Hirata work for a undergrad course
but quite crude by modern Nonlinear System Identification standards - try a google search for senior undergrad/intro grad level textbooks from the last decade or so
I only have to stand up to pick Pintelon, Schoukens IEEE Press 2001 "System Identification A Frequency Domain Approach" from my shelf
https://books.google.com/books?id=KhonXGwETWsC&printsec=frontcover#v=onepage&q&f=false
it only does very light Nonlinear Modeling - but quite sufficient for the weak, low order nonlinearites from competent audio electronics
I'm only able to follow the math in the 1st half, but that's the point - if needed I have a resource to learn from
also with many packages for matlab and free equivalents today even a general acquaintance with the math lets you run very sophisticated software
Last edited:
Away, or in wane?
Even assuming you would understand the math in the paper, you don't need it to be a successful designer, like yourself, in a fashion based industry like High End Audio. As you very well now (and practice here every day), a good story about having a secret sauce for triggering customer ears satisfaction is worth more than a million Fourier transforms. So why bother explaining?
What is the slew rate of the step portion of the signals? I ran through the paper quickly but could not find any indication of that. If I were doing that test, I'd first worry about what the fast edges are going to do to the amplifier.
John
DIM-30, 100 have been AP analyzer macros for a while now
https://www.ap.com/technical-library/dim-30-and-dim-100-measurements-per-iec-60268-3-with-ap2700/
https://www.ap.com/technical-library/dim-30-and-dim-100-measurements-per-iec-60268-3-with-ap2700/
Last edited:
I didn't see any reference to a band limited dim-30 or dim-100 test apparatus within the Hirata papers. His diagrams were simple blocks, no low pass anything.
Did I miss something?
I also find the spectra drawing on fig 2 to be interesting...a cusp at 28Khz then a clear rise in spectra amplitude above 28k. What does that mean? Is the dut being subjected to lots of hf content without regard to what it will do with it?
I would think driving high ultrasonic signals preferentially into either one or the other of the two output polarity sections would certainly cause differences in output, especially when one considers that the output stages probably were not designed for that.
If one really has no understanding of what one is examining with a test, might as well use a scale and consider weight as the quality metric.
Again, if I'm missing something, please let me know...
John
Last edited:
I am not going to do all that... figure the phase angles of that Hirata waveform and decompose it into sign waves? are you serious! You do it. Good luck.
Actually, I think I'll just begin with a train of pos or neg pulses (0-1v) at various rep rates and duty cycle. And see what can be improved on various amplifier circuit designs. Then, maybe an arb gen of the Hirata waveforms. Or other asym waveforms which an average of non zero. First without my dc servo.
THx-RNMarsh
Last edited:
Translation: I have no idea how to do that, and I don't care to figure it out. All I want is to make a point.
Comment: It's actually easier that you could imagine. A good soul (not me, I'm not good enough) could perhaps be bothered to engage in a useless exercise to explain it.
The original one or the improved Yoyodyne version?
Yoyodyne Propulsion Systems Interocitor Mark IV - Lakeside Bicycles Lake Oswego OR 97034 503-699-8665
Hilarious, I mean really hilarious you're not a master of Fourier theory I presume? Really, good luck? It's a nice EE102 homework problem for a sophomore.
100% true story, I had two interviews in one day the first was Raytheon Missile Systems and the second was ADI. I had about 6 people on the first, after the second one I asked a random secretary if there was an exit and I snuck out a back door not to be late for the second, the best decision I ever made. My first interviewing experience at 23 and there was no f'ing way I'm working for .gov.
Last edited:
Give me a clue how to measure it and I'll give it a go once my balanced phono is complete. I still don't think there is an easy way to measure it with the cartridge attached but many here are more skilled in the art.
We agree !!!
write down this date: you are 100% correct.. I don't want to learn it or spend any time learning and doing it that way. and you are not good enough I do want to see how various topologies handle that kind of signal. that's all there is to it. You know .. like increased 2H etc. Including dc servo. The amps are built, the test equipment is here. After I do the multitone, of course. and compare.
Meanwhile, can you compute or measure cmrr of riaa phono with cart attached?
-RNM
Last edited:
- Status
- Not open for further replies.