Are hybrid parameters useful for design?

I asked a simple question, the same as you asked, and ask it again:

Do any of the people who post here design electronics using transistor h-parameters (beyond the obvious hFE, static beta)?

Some people sure get angry when I ask that question.

Apparently you fell asleep in civics class too, because India is a paradigm of meritocracy and not its absence. Meritocracy is a political term which describes the selection process for administration positions as being base d on standardized test screening, aka civil service examinations, versus pe rsonal and/or political party affiliation.

I've worked with some Indian engineers who were brilliant theoreticians who did stuff that I can't do, like analyzing complex structures for shock and vibration modes, analytically. I think that sort of seriously difficult math and theory is admired more than the more hands-on work that most EEs do.

Instinct + computer simulation levels the playing field here for the hands-on guys.

The US benefits from having a lot of kids who tinker with cars and tractors and broken radios, and acquire mechanical and electrical instincts before they go to college and learn the theory. At least it used to be that way; we have fewer farmboys now, and too many city kids play video games or write code. Radios aren't as much fun to take apart as they used to be.

I'd think that computer programming/FPGA design would be the ideal venue for places like India to shine. It's clean indoor work, doesn't need a lot of physical support, and importing requirements and exporting code are really easy. It's also non-academic, in that a CS degree has almost nothing to do with coding ability.

Hey, something like half of our Harvard and MIT graduates go into the finance business.

What sort of stuff do you do?

Do you ever use transistors for anything other than non-critical switching, current sources and buffers?

In the early days of transistors, people built class-A audio gain chains and stuff, and gain was expensive. We don't do that much any more, gain is cheap, and one can always do rough math (calculator, or even in your head) estimates, followed by simulation if additional precision is needed. So, not many people do the h-param matrix math stuff that used to be taught. Nobody here has, so far, confessed to doing so.

One place where math rigor is still needed is small-signal RF design, but those guys use s-parameters and different tools. The thing about s-params is that they are provided with RF parts, as a fine-step function of frequency, which can be hundreds or thousands of parameter points for a simple transistor.

Power RF is, I think, usually done with Spice (if you can get the models) or experiment, like semi-automatic load pull analysis.

current sources and buffers?

Some of our current sources are critical, by my standards, like needing to be stable to 0.1% and being very wideband, GHz equivalent bandwidth.

We occasionally do super low-noise stuff with jfets and, very rarely, discrete bipolars. Most gain comes from opamps these days. Really fast stuff often uses PHEMTS.

NMR gradient drivers are current-output amplifiers, from 3 to 120 amps peak, with PPM noise and pulse flatness. We use mosfets in the power stages.

H-params are only useful for bipolar transistors, for small-signal, low-frequency analysis. We don't do much of that any more.

We just did an air flow sensor, based on change in Vbe of a self-heating TO92 transistor. H-params wouldn't help there, either. It might work; if it doesn't, we'll just not include that feature on the product.

meritocracy and not its absence. Meritocracy is a political term which describes the selection process for administration positions as being based on standardized test screening, aka civil service examinations, versus personal and/or political party affiliation.

Administration positions? Civil service? I thought we were talking about electronics design.

Do you think you know more about India than Rajib does?

current sources and buffers?

You're still not getting it through your thick skull that the classic two-port h-parameters are NOT hybrid pi.

the

I read your response as "They are not on transistor data sheets." Is hfe an h-parameter? Can you at least admit that hfe is on most data sheets? All I have said is that you are wrong.

You did not (in earlier posts). If I am wrong, please show me where.

I did not say that h-parameters are useful. I did not say that I use them. I did not say that any other person uses them.

I don't care about whether you like them, or use them, or any of your other excuses to avoid the fact that your statement "They are not on transistor data sheets" is inaccurate. At least hfe is there. There are other, low level bipolar transistor, data sheet which still have h-parameters.

Why can't you accept that?

of meritocracy and not its absence. Meritocracy is a political term which describes the selection process for administration positions as being based on standardized test screening, aka civil service examinations, versus personal and/or political party affiliation.

Do you think *you* do?

And, of course, RF transistors do not show hfe on the data sheet, right? (Don't forget, hfe is an h-parameter)

Please show the post where you asked that question.

What does that have to do with your original statement "They are not on transistor data sheets" ?

Back peddling now, eh?

And you wonder why?

you

laboratories?

the

Static beta, aka hFE (not hfe!) is on basically all bipolar transistor data sheets. The other h-params rarely are.

here:========================================================

look at the datasheet? They're on bunches of others, mostly older, too.

Yeah, mostly older. Of the transistors that we use, most data sheets have hFE, and only one out of the dozen that I checked had any other h-params.

==============================================================

"They" is plural. hFE is singular. We call it "beta." The point of this thread isn't whether beta is on data sheets; the question asked was whether anybody actually uses h-paramater analysis, as is taught in university courses.

Just hFE isn't enough to do the academic matrix h-param stuff. So the answer to the OP is "nobody here seems to use it."

of meritocracy and not its absence. Meritocracy is a political term which describes the selection process for administration positions as being based on standardized test screening, aka civil service examinations, versus personal and/or political party affiliation.

Having never been to India, certainly not. I have worked with, roughly, a dozen Indian engineers.

I doubt that a civil service exam demonstrates any sort of meritocracy in electronics design, anywhere in the world.

It was funny for Fred to insult Rajib for knowing less about India than Fred does. I wonder how long Fred lived in India.

want

with

current sources and buffers?

h-parameters are NOT hybrid pi.

Did I say they were? What I said is that not many people still use h-params to design transistor circuits. Which answers the question that was asked.

I answered your question, too.

It's what Rajib asked.

Nearly all bipolar transistor data sheets include hFE (and, much less often, hfe.) But few include the full set of h-params, what you need to do classic matrix-based h-parameter analysis.

It's shocking how little DC data is on most RF part data sheets.

The hacker/maker movement has brought back the electronics hobby for youngsters, fortunately. It's robotics instead of radios, but why not?

Cheers

Phil Hobbs