I was wondering what happened to popcorn noise. It was mentioned in my textbooks 30 years ago, but I haven't seen anything about it recently. Is it still a problem with modern devices?
Last edited:
Popcorn noise disappeared a long time ago about the time employees stopped smoking in fab houses ?
.
Still there, and gettering is still required. SOI processes require top-side gettering, a little tricky. Used to be scratching up the back or implanting argon would do it, but that is blocked by the oxide isolation.
Funny comment, my first boss chain smoked the entire day and one of the fab techs used to line up liittle butts on end all over the lab and they burnt down to the filter slowly for minutes.
Last edited:
I, for one, enjoy your posts about the "history of solid state amps" and hope they continue.
Also, presenting the problems you found and the way you solved them is really educational.
By 1975, we had full 4 quadrant complementary differential designs being developed for home use. The biggest advantage of a balanced bridge amp, in those days, was improved safe area utilization of the output devices. Another approach was the series arrangement of output devices, developed initially by Erno Borbely, for Dyna (I think) that did much the same thing.
The problem as I previously pointed out with solid state power transistors, especially complementary ones, was their useful safe area, that was only a small percentage of its ideal potential. I mean: even 10A, 80V and 150W per device sounds like plenty, UNTIL you try to keep the amp from blowing up, even if you limit the supplies to +/- 40V, current limit 10A peak current, and use a decent heat-sink. The SECOND BREAKDOWN of the output transistors limits you to a small percentage of the actual potential of the output device. Today, it is not so bad, but 35-40 years ago, it was tough. AND, when Motorola (the dominant complementary part maker in that day) would bring out a higher voltage device, they always seriously compromised it in some other way, usually more nonlinear beta, lower F(t), and often, lower peak current. Many of the solid state power amps were made with these (compromised) devices. I never could get one to sound completely right with them, and that is why the JC-3 used older, lower voltage devices, for optimum performance.
If anyone out there really wants to see the problem graphically, then just convert the SAFE AREA GRAPH of a typical bipolar transistor from a log to a linear graph, and plot a typical load line of a resistor, and then something more capacitive or inductive as a load line. You don't have to get extreme, BUT you will find that the effective safe area is almost non-existent outside a limited range, and that is what makes protection circuits often necessary, and why they were annoyingly working a lot of the time. (more later)
The problem as I previously pointed out with solid state power transistors, especially complementary ones, was their useful safe area, that was only a small percentage of its ideal potential. I mean: even 10A, 80V and 150W per device sounds like plenty, UNTIL you try to keep the amp from blowing up, even if you limit the supplies to +/- 40V, current limit 10A peak current, and use a decent heat-sink. The SECOND BREAKDOWN of the output transistors limits you to a small percentage of the actual potential of the output device. Today, it is not so bad, but 35-40 years ago, it was tough. AND, when Motorola (the dominant complementary part maker in that day) would bring out a higher voltage device, they always seriously compromised it in some other way, usually more nonlinear beta, lower F(t), and often, lower peak current. Many of the solid state power amps were made with these (compromised) devices. I never could get one to sound completely right with them, and that is why the JC-3 used older, lower voltage devices, for optimum performance.
If anyone out there really wants to see the problem graphically, then just convert the SAFE AREA GRAPH of a typical bipolar transistor from a log to a linear graph, and plot a typical load line of a resistor, and then something more capacitive or inductive as a load line. You don't have to get extreme, BUT you will find that the effective safe area is almost non-existent outside a limited range, and that is what makes protection circuits often necessary, and why they were annoyingly working a lot of the time. (more later)
Scott, this summary is too brief for my imagination. Plse help me understand the deeper meaning of these two lines,
vac
Noise, there are plenty of places where the noise does not matter like a line amp or inside the loop of a composite amplifier. 14nV/rt-Hz at 2V line level is not very important.
OK, tnx, got you, noise has never been a concern for me with opamps anyways, ever since my first active filter in the beginning of the 80's with LF356's.
For those who are interested, I just put Bruno's feedback article from Linear Audio Vol 1 on-line. It's downloadable from Linear Audio | your tech audio resource under the tab Online resources.
Enjoy!
Enjoy!
Blatent Advertising!
Jan you aren't planning on visiting Texas blindfolded? Do you know what they do to people wearing blindfolds there?
How we hear?
I want to suggest the reading of Frank Leonhard white paper that proposes some interesting discoveries about how we hear; Leonhard explains well what I intuitively found with the pulse test. Look here.
Of the multitude of theory, Leonhard's is the most interesting for the audio people.
I want to suggest the reading of Frank Leonhard white paper that proposes some interesting discoveries about how we hear; Leonhard explains well what I intuitively found with the pulse test. Look here.
Of the multitude of theory, Leonhard's is the most interesting for the audio people.
- Status
- Not open for further replies.