No more modified pdf by file hosters

Kids here get degrees in journalism or geography or fine arts and leave with $150K in student loans and no job prospects.

Kids need to appreciate how much fun electronics can be.

Are they linear in phi_n(f) to that accuracy (n = 1...4 are the outputs), or just phi_n(f)-phi_m(f), i.e. the phase difference between outputs? You only care about the latter when doing SSB voice, but you care about the former for real I(Q).

I'd like to figure out the Weaverish approach sometime.

Cheers

Phil Hobbs

Same here, except the debt is lower, and the repayment is based on the salary you earn above ~80% of the national average wage. So if you don't earn, you don't repay. Drat.

And that is it done by "ordinary people" not the "superbeings" seen in The Glass Teat (Harlan Ellison, RIP).

Up to, say 8, they don't realise some things are difficult, and they just do them. Only in their early teens they start to think that "people like me don't do /that/". Get then past that attitude, and the sky is the limit.

If those are electives then it isn't an EE degree program. Full Stop.

The phase difference of the two outputs is close to 90 degrees, which is all I need to do i/q, with the two outputs having some arbitrary absolute phase shift vs frequency relative to the input. Since we'll work at one frequency at a time, and it *is* an i/q modulator, we can just rotate that out.

"Of course I kill people. I'm a terminator."

Is there a Weaverish way to make a wideband 90 degree phase shifter?

Of course, I'd prefer an ideal Hilbert transformer, preferably in SOT-23.

Think so. Since a Weaver with two copies of the back end gives you USB and LSB with no baseband phase shifter, you can make USB+LSB, which is a frequency-shifted version of I, and USB-LSB, which is a frequency-shifted version of Q.

Plus it unfolds into a giant robot. ;)

Cheers

Phil Hobbs

For all its failings as a newsreader, Thunderbird has the advantage of some very useful filter actions. Simply killing an author's posts is not what I need.

There are folk who I *never* want to read; not them, not any response to them. Skyfuck, I'm looking at you. I feel stupider every time I read you. Don't bother responding. For these authors, I set "ignore subthread" and they and their respondents just vanish.

There are folk who seldom seem to contribute anything worthwhile, but do know stuff and sometimes elicit worthwhile responses. For them, I simply "mark as read" in a filter so "next unread" skips over them. I see responses to anything worthwhile, so if I need to see the original, it's there. Slowman, RickC and others are in this bucket; even though Rick was helpful on GPS timing and I agree with most of what Sloman writes; I just don't need to read it.

Finally, any topic I don't care about, I simply (K)ill the whole thread. If I miss a worthwhile sub-thread, de nada.

These three actions have made reading s.e.d much quick and more productive.

Clifford Heath.

Hobbyists with no higher math education tend to top out as technicians.

Of course some of those go on to great things, e.g. Jim Williams who co-founded LTC, and Errol Dietz who started out as Pease's tech and wound up as CTO of National Semi, but they're the rare exception.

Yours truly, possessing a hobby background and a newly-minted physics and astronomy degree, and having never seen a PLL, lucked into a job doing 2/3 of the time-and-frequency circuitry for the first civilian direct-broadcast satellite system. Drinking from a fire hose doesn't cover it.

I'd never have succeeded without the math background--it's not even that I needed a lot of Bessel functions, though I did, but having some mathematical maturity meant that I wasn't scared of all the unfamiliar signals and systems stuff. (Nice alliteration, eh?)

Cheers

Phil Hobbs

And a great many EE programs are now Electronics and Computer Engineering (ECE).

Cheers

Phil Hobbs

Pointer to the Weaver approach please?

The quadrature hybrid is a very interesting simple approach that can be cascaded for 90-degree shifts at RF. Four stages can cover a decade fairly cleanly.

Each stage is essentially a 1:1 transformer with a capacitor joining the centre taps. All four ports want to see the design impedance, so just feed one port from a buffer amp with series R, ground the other input via R, and the two other outputs produce two signals at 90-degrees to each other at the same impedance. You don't need any active devices to cascade four stages, which can generate 90degree signals from e.g. 3 through 30MHz with only a degree or two of error.

Try it in LTSpice. That was one of my happiest afternoons of simulation. I don't know why the approach isn't more widely known.

Clifford Heath.

Again, that's phi1(f)-phi2(f), which is fine for narrowband signals but not good enough for video-ish ones. (Drat it, we need another word for fast signals with large fractional bandwidths--'video' means something else nowadays.)

It's very well known to anyone who grew up with the ARRL Handbook. ;)

Cheers

Phil Hobbs

We had great value hiring software grads who won our annual software-engineering prize. The academics were happy for us to send someone at the start of the year to offer the class two prizes of $2500 each, one for best SWeng project and one for (I forget), to judge the winners for us, and for us to attend and hand out a prize cheque. The students seemed really motivated by a prize from a desirable local employer.

I think we hired winners and runners-up for several years running, cheapest good hiring we ever did. The same thing could work for EE grads.

Clifford Heath.

Thanks. I knew the technique but not the name.

Right. I wanted to broad-band a short E-field antenna and an M-loop, phase-shifting one wrt the other and adding the signals to get a cardioid directional pattern for direction-finding. It simulates well but I haven't done a full-band test yet.

Right, but I've never seen one in the wild. What applications have you seen it used in?

Clifford Heath.

I haven't been a big over-the-air radio guy since the '80s, though I use a lot of early radio ideas in my optical instruments. Having the image frequency a petahertz away is comforting sometimes. ;)

Cheers

Phil Hobbs

Which doesn't have as much to do with "real" intelligence (whatever that is ) as the test-sellers like to claim. It has a bit more to do with skill at giving educators what they want to hear, but the correlation is still imper fect.

Tests have to be cheap. Chucking out a few good candidates and accepting qu ite a few duds doesn't impact the bottom lie of the test-givers.

If the extended family does not include people who have been to university, working out which degrees might be "worthwhile" can be difficult.

A 1950's Australian study of undergraduate success made the number of books in the parental home the best single indicator of success, closely followe d by having a near relative who had been to university.

In a sense you have to go to through at least one university course to find out what university courses are about, and both Win Hill and I did a lot o f chemistry before we found out that we were better at electronics.

Not my experience. I got my electronics as a graduate student, after having done a sub-major in math (though the second year math course wasn't exactl y exciting or comprehensive).

The math was handy for making sense of a lot of the electronics. I did Theo ry of Computation 1 as a graduate student, which did include some useful ma th (thogu the most useful information came from one of the demonstrators M. Sc. thesis on non-linear multi-parameter curve fitting).

I had a huge Meccano set, and it didn't turn me into a mechanical engineer.

In the sense that solving problems is fun, and math lets you solve more pro blems, or at least gives you some confidence that your solution is close to optimal.

Phil Hobbs wrote in news:S8GdnTvT2Z-aQz snipped-for-privacy@supernews.com:

wow ok tiny:

I built units with 16 of these in each. The synthesizers we used in the units were $36k each.

Take 8 of those each in a rack (they were 4U modules), and then 9 racks of those, all with equal length cabling... and you can put together a stimulator for advanced stealth aircraft.

There are permanent installations of such gear down at the giant Hughes anechoic chamber... I built.

snipped-for-privacy@decadence.org wrote in news:pnuuo1$13m3$1 @gioia.aioe.org:

snip

the

It was actually these...

Can you post the Spice model? I'm not seeing that online.

Thanks

John

On 20/09/18 14:45, John Larkin wrote:

Here's one that shows the adding the E and M field signals with an additional 90 degrees shift from the hybrid. It shows good nulls from 2.7MHz through 4.4MHz, though the direction of the null moves about 20 degrees.

Clifford Heath

Version 4 SHEET 1 1028 680 WIRE 128 64 -80 64 WIRE 160 64 128 64 WIRE 272 64 240 64 WIRE 304 64 272 64 WIRE 400 64 384 64 WIRE 448 64 400 64 WIRE 592 64 448 64 WIRE 448 80 448 64 WIRE -80 96 -80 64 WIRE 272 112 272 64 WIRE 112 224 32 224 WIRE 160 224 112 224 WIRE 272 224 272 176 WIRE 272 224 240 224 WIRE 288 224 272 224 WIRE 400 224 368 224 WIRE 448 224 400 224 WIRE 32 240 32 224 WIRE 448 240 448 224 WIRE -80 336 -80 176 WIRE 32 336 32 320 WIRE 448 336 448 320 FLAG -80 336 0 FLAG 448 336 0 FLAG 128 64 P1 FLAG 400 64 P2 FLAG 400 224 P4 FLAG 32 336 0 FLAG 448 160 0 FLAG 112 224 P3 SYMBOL ind2 256 48 R90 WINDOW 0 5 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName L1 SYMATTR Value {Ls} SYMATTR Type ind SYMBOL ind2 400 48 R90 WINDOW 0 5 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName L2 SYMATTR Value {Ls} SYMATTR Type ind SYMBOL cap 256 112 R0 SYMATTR InstName C2 SYMATTR Value {C2} SYMBOL ind2 144 240 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L3 SYMATTR Value {Ls} SYMATTR Type ind SYMBOL ind2 272 240 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L4 SYMATTR Value {Ls} SYMATTR Type ind SYMBOL res 432 224 R0 SYMATTR InstName R2 SYMATTR Value {Z0} SYMBOL res 432 64 R0 SYMATTR InstName R1 SYMATTR Value {Z0} SYMBOL voltage -80 80 R0 WINDOW 123 24 126 Left 2 WINDOW 39 25 103 Left 2 SYMATTR Value2 AC {Amp} SYMATTR SpiceLine Rser={Z0} SYMATTR InstName V1 SYMATTR Value "" SYMBOL voltage 32 224 R0 WINDOW 123 24 126 Left 2 WINDOW 39 25 99 Left 2 SYMATTR Value2 AC 1v {Phase} SYMATTR SpiceLine Rser={Z0} SYMATTR InstName V2 SYMATTR Value "" TEXT 200 8 Left 2 !K1 L1 L2 L3 L4 1 TEXT 304 520 Left 2 !.param Ls=Z0/(4*2*PI*F0*0.85)\n.measure Ls_ PARAM {Ls+0} TEXT 304 448 Left 2 !.param C2=0.85/(2*PI*F0*Z0)\n.measure C2_ PARAM {C2+0} TEXT -128 448 Left 2 !.param Width=1.4\n.AC lin 300 {F0/Width} {F0*Width} TEXT -128 520 Left 2 !.param Z0=200R TEXT -128 408 Left 2 !; .step param Phase 70 110 10\n.param Phase 90 TEXT -128 544 Left 2 !.param F0=3.6Meg TEXT -120 584 Left 2 !;.param Amp=1v\n.step param Amp 0.8 1.2 0.1