Cascode substitutions

Articles like this:

Are dead from the start, because they don't define the terms. What are the definitions of your "axioms" of right and wrong, good and evil?

The problem could be argued to be one of categorization. Essentialism is at the core of many "moral" problems such as abortion, animal rights, and end of life issues.

When does life begin, and when does it end? Nature will never give you a clear cut answer to these questions. Saying a fetus is a "person" after 3 months, and not before, is ridiculous.

Why is a human embryo consisting of a couple cells, with no capacity for pain or fear, afforded more protection than an adult animal, which can clearly suffer? Outside a few fringe circles Americans seem to have no problem slaughtering the latter beings in their millions, every year.

There are cultures and religions, however, that take great offense to that (Jaanism, etc.) Are they "wrong"?

Humans want to box things into neat bins. It's a natural predispositon, since not being able to quickly answer yes or no to whether something is dangerous or not would get us killed real quick 100k years ago.

But science has shown us over and over again that Nature at a more fundamental level doesn't really operate that way.

Check out the "firewall paradox" in quantum cosmology if you want a real trip.

Put briefly, it appears that to truly resolve the black hole information paradox, as observers we can't all even inhabit the same Universe.

Maybe there isn't even any "observer" at all.

It can be done - and I did it when there were no other options with realistic results. The problem comes, for example, when you're dealing with a dispersive structure over a wide bandwidth with frequency dependent losses. With SPICE you have to generate a lumped element equivalent circuit which may consist of 20 sections (to get good numbers over 4 octaves). Each section is a 3 element LP but the L is a set of inductors and resistors to produce a piecewise approximation to skin effect (fractional pole expansion - see

Farkin' tedious, especially when you have 2 balanced lines broken up into 6 subsections that you want to vary length to tune to one frequency, null another, and couple a signal in/out.

So, being able to drop the pieces onto a schematic, plug in dielectric constants, physical dimensions and conductivities, select a well tested model for the dispersive component and go is much nicer.

This is an xml definition for a one section of a coupled microstrip filter:

"Subst1" further expands in a piece of 30 mil thick Rogers Duroid 5880. "Kirschning" is a recognized model for dispersive microstrip. The numbers are entered into a dialog box or pulled in from a library.

Less pain.

Morality is like, is part of, consciousness. If God permeates the universe, He permeates our DNA too. We don't just follow the rules, we are part of the rules.

Pragmatically, societies that play fair do better. In my experience, honest and generous people usually come out ahead (exemping the occasional talented sociopath.)

Good electronic design requires a lot of honesty and a good helping of generosity. It's a team sport.

Which does make you better than your principles, at least part of the time. You might want to inquire--as a matter of some urgency--where that higher moral law (for that is what it is) that judges your intentions comes from.

One of the many problems with that view is the immense agreement between people of all eras and all places on what is and is not OK. There are disagreements at the edges, e.g. whether you have to be unselfish towards everybody or just your own family/tribe/clan/nation, or whether you can have one wife or five.

But I claim that you can't actually imagine a society that sincerely admired people who ran away in battle, or who double-crossed all their friends, for instance.

Cheers

Phil Hobbs

God can't save you. She's out shopping.

Well, yeah, a society made completely of atavistic sociopaths wouldn't last very long. But then that means at some level morality is something that must evolve parallel to humans living in societies at all, which means it must stem from a particular society. i.e. it's not a universal innate quality of humans.

Maybe it might be better to say that certain humans have an innate _capability_ for morality, in the same sense that every human born with certain physical and mental characteristics has the capability to be an astronaut, given the right environment and the right life circumstances.

I'm pretty sure that some people (fortunately a small amount) are born rotten. They come that way from the factory and nothing can change it. I definitely don't belive in any innate "goodness" of humans, or that we're all made in God's image.

A higher moral law giving humans free will, but then essentially revoking it in the same breath by dictating that there will be horrible consequences for not using it in the appropriate fashion, and then giving the players no option whether the wish to play or not (being born) seems like a rather perverse game. You might sometimes question the sanity of whatever moral authority wants to run that one.

But yeah, I'm pretty sure that if there was a Son of God, Christ was a good candidate. Tell people to love each other, get nailed to a cross? Sure. Sounds like just the sort of thing people would do.

What makes you the expert in someone else's moral principles? Sheesh, that arrogance surely knows no bounds.

So therefore there's something extra-physical, mystical, magical, non-material, from which all moral truth descends? That's a total failure of logic.

Could it be that folk just recognize that they need to try to contribute to making and maintaining the kind of world they'd like to live in? That they have a responsibility - even a selfish one - to maintain a moral order?

C'mon, it's not that hard to imagine! No need to invent angels and faeries to define the truth for us, and then try to use logic to prove they're not invented. Hilarious.

Clifford Heath.

It's advantageous for an individual to live in a society of honest and productive and generous people, and it's advantageous for that person to steal all he can. So, we have laws.

Yep, but any society of sufficient size can always stably support a few chronic vampires that will be successful pulling it off.

LT Spice has the ltline element, a classic Heaviside lossy transmission line. It doesn't do higher-order things like skin effect.

At some point, prototype it or get some big-bucks EM modeling software.

I'm not talking about that, I'm talking about the difference between "ought to be" and "is".

But that's contrary to the evidence. You're a victim of your preconceptions here.

Again, morality is about "ought to", not "is".

But how do you define "rotten", if there's no standard? How can you say that a saint or a great humanitarian is "better" than a Mafia don or a drug dealer, if there's no standard?

Cheers

Phil Hobbs

I don't know anybody who advances such a worldview. Do you?

If there were such a person, I certainly would. But there isn't, as far as I know.

Absolutely. But He also exhibited what love looked like in an altogether unique way. Hint: it isn't always about making nice. If your teenage daughter is a heroin addict, you don't say, "That's lovely, dear, have a good time shooting up in the back alley." Sin is like that--harmful from the beginning, and in the end, lethal.

Cheers

Phil Hobbs

Well, including DC basically means replacing inductors with complementary PP, or CCSs. And bandwidth is just the span from "near DC" to whatever the HF cutoff is.

You can tweak a "narrow" amp into a "wide" one by playing with the filter topology and impedances.

Suppose you start with a conventional narrowband amplifier, using parallel tuned resonators, and matched with whatever means apply (L-match, L or C dividers, tapped coils..).

At each resonator, the ratio of resonator Zo to system Z gives the Q factor and therefore width of that stage (give or take stacking resonators for flatter bandpass, sharper skirts, and slightly stretching (peaking) the bandwidth limits).

As you reduce Q (assuming you are able to, by increasing Z and L, and reducing C; normally C has a minimum due to the amplifying device, which trades off with Z as an approximate GBW limitation), the bandwidth spreads out, and the impedance ratios available via reactive matching networks get closer to 1. So you start needing more transformers than reactive networks for matching, and resonators with impedances closer to the device or system impedance. Which tend to get less extreme and therefore easier to construct. Series-parallel transformations can be reversed, giving traditional filter topologies, like series resonant links between parallel resonant tanks, instead of coupled resonators.

As BW rises still further, Q < 1 means R dominates over X for much of the band, and the networks aren't really resonators at all anymore, so you can pull them apart, using large value RFCs to supply bias, and using LPF network design to manage the device capacitance.

At this point, it's usually relatively trivial to convert the remaining RFCs and coupling capacitors into DC-friendly R or CCS parts, at the usual cost of efficiency. Transformers being the one exception, for which you need to change much more (supply voltages, transistor types..). (This being one of the fundamental limitations of vacuum tubes -- high load impedance.)

So along this spectrum, I expect to design an amplifier in one of the latter categories: treating HP and LP behavior separately, using matching if necessary.

"Wideband" means different things to different people. I've seen planar circuits claiming 1GHz "wideband" performance. Sure, that's pretty wide in the usual scheme of things, but out of a 20GHz center frequency, that's a puny 5%... a Q of 20... To me, I mean a circuit where the LP and HP behavior is roughly independent, and the HP is optional (so that it may truly include DC).

Speaking of tubes -- they work just fine at high frequencies. 6C4 is a rather unimpressive type, but an oscillator built with one (running at a modest, say, 100MHz) will show harmonics out to nearly 1GHz! It's the only nonlinear element in the circuit, so all those harmonics are indeed due to changes in electron flow occurring that fast. Of course, those harmonics are in the -80dB range, so it's not like there's much left up there. (It's also relatively unusable, because electrode self-resonances occur in the 300MHz+ range.) But the physics is real. Planar triodes of course went out to about 10GHz, and TWTs are still around with us today.

I don't think most transistors drop off in the same way. Tubes have such low current density (and therefore high Z and low Gm) that, for baseband, electrode capacitance dominates over the fundamental limitations. Most BJTs have R_L* Ccb dropping off around the same range as fT (which is due to Rb * Cbe and diffusion/recombination). I suspect that power switching FETs are similarly Ciss * Rg limited; but RF FETs (in Si) have quite high limitations (small Cin and Cout, low Rg), when made properly.

There was one RF part that you found rather lazy, an HBT I think? That's probably such an example. Relatively high Cout, but fT through the roof?

Coming from your main perspective, I think -- most of your projects have been limited to baseband, and thus your HF limit is pretty much 1 / (2*pi*Cout*R_L). With peaking, a little over double this, but I'm guessing for your applications, dropping in a superlative GaNFET or whatever is more efficacious and economical.

And if this were a job job, I'd do the same, but fortunately it's not, so I can take the learning opportunity instead. :-) Application, just lab use, maybe a little radio when I get there (hey, just add a wideband antenna and you're set, no need for "tuning the finals" :) ).

Tim

Sorry? It's okay with you if I rob you blind because I think it's okay?

I didn't say that. You keep going on about mystical magical stuff, whereas I'm talking about the basic structure of existence.

No, it couldn't. Otherwise nobody would do the right thing when nobody was looking. Some people do.

If anyone actually thought that, it might be. But you're just whistling in the dark.

Cheers

Phil Hobbs

Whoosh. You completely missed the point.

You made the comparison "better" in regard to bitrex's "principles". How do you even know his principles? You merely assume that they are worse than his self-described behavior because you assume they must be worse, because he's a moral relativist. Is it possible that you don't even see the non-sequitur here?

Yes you did. You told bitrex to "inquire where that higher moral law (for that is what it is) that judges your intentions comes from" and then went on to assert that it must exist because every society thinks there must be one.

Guess what? It doesn't come from anywhere but human imagination. We imagine how nice it would be if *everyone else* acted like that, and try to make it so by all means possible.

No, they don't. Everyone acts in their own selfish interest for their own convenience anyway. The only difference is that some people define convenience in terms of an imaginary standard of uprightness, of which if they don't loudly protest their support, the others of their social circle will disapprove of them... and that would actually become inconvenient.

More rational folk admit that this absolutist approach to granting self-approval, in order to gain societal approval, in order to maximize convenience for oneself (but without admitting that's what's happening), is nothing more than a mind trick.

Religion is the only stage-play that relies on the actors denying to themselves they're even acting in a play. How can audience and actor co-exist in one undivided soul? No wonder it creates such perversions.

Clifford Heath.

Some people use "LF" instead of, literally, DC coupled. You know what I mean.

And gain drops as you do it. Lots of RF amps will drop below unity power gain as you try to widen their bandwidth. That's why people are using exotica like GaN to make really wideband amps.

For a given device, you generally have a limiting GBW, so wideband tweaks eventually kill all the gain. RF mosfets and HF mmics are usually internally tuned, to get gain at high frequencies, and there's nothing you can do to make them wideband.

Low noise figures are assiciated with tuned circuits, too.

Lazy? I don't recall that. Phemts and GaN (and, to a lesser extent SiC) parts have way higher Gm/c ratios than silicon, so make much better wideband amps.

We do work in time domain. One way to get outrageous gain-bandwidth is with distributed amplifiers. They are hard to do discretely, so are generally ICs. You can buy a LF-to-40 GHz amp, for example, from Hittite or Triquint or whoever they are these days.

OK, but RF power amps are usually tuned. That is a lot more practical (lower cost, fewer stages, higher efficiency) than going wideband.

Am 01.02.2016 um 01:02 schrieb John Larkin:

I must have lost the beginning of this thread on my server, so...

The biggest RF PNP that I have seen is the NXP BFG31, SOT223. (and I have secured a reel of it :-))

Anything that has a high gain as the product of the gains of two transistors is unfriendly to feedback since its combined gain falls at 12 dB/octave and at gain = 1, Mr. Bode will get you.

The modified Darlington with the diode at the base of the second transistor has a much better behaviour. Beta is only beta1 + beta2, but it keeps it to much higher frequencies as it falls only with

Maybe one can do the same to the Sziklai?

regards, Gerhard