naa, naa, naa poo poo, I know you are but what am I....

Phil, after a while I just stop reading the threads. Sure maybe there's some bit of electronics still going on and I'll miss something, but.....

Have you ever done any coherent population trapping?

Here's the latest news letter. (the copy is written by my boss, who understands it less than I do.)

The signals are big.

George H.

The Dutch uncle approach doesn't work, because it's patronizing, which only increases the irritation level of the participants. A bit of mild and friendly ridicule, OTOH, has been known to help. We'll see.

I've built a fair amount of spectroscopic apparatus, but haven't done much actual spectroscopy myself. This looks like a fun one. As an old friend of mine says, the best gizmos are "simple but devastating."

Cheers

Phil Hobs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203 Briarcliff Manor NY 10510 845-480-2058

hobbs at electrooptical dot net

MOMM-MIE!!! THE LITTLE BOYS ARE BICKERING AGAIN!

In completely unrelated news, I wound a 100kVA transformer the other day. No big deal...

Tim

-- Deep Friar: a very philosophical monk. Website:

Phil Hobbs wrote: : Do carry on--it reminds me to be grateful I'm not in middle : school any more.

My feeling exactly ( tinyurl.com/7dblxyf ).

I have been idly pondering, instead, how to make a Poincare dual out of the T-coil peaking circuit. You know, swapping I U, C L, swapping loops with nodes and so on. The required mutual coupling makes the straighforward conversion difficult. I guess I must write the transfer function and try to synthesize the circuit from scratch.

Regards, Mikko

Real Men do that at 60 Hz.

John

I was doing something like that recently: swapping voltage for current in the traditional H bridge. Hint: left pair becomes top pair, series diodes are required (a MOSFET has an antiparallel diode which conducts in reverse, whereas current more requires open circuit in reverse), and instead of dV/dt or peak voltage snubbing, you use dI/dt or peak current snubbing.

What I haven't quite figured out yet is how to conserve number of snubbers. For example, in the voltage source H bridge, if you indicate total supply inductance as inductance between the nearest bypass capacitor and the bridge supply, this becomes the total supply capacitance attached of the current source (Norton source, current || cap). In the voltage source, the series inductance needs to be snubbed with an RCD in parallel with the bridge supply terminals; it follows that, in the current source, this capacitance needs to be snubbed with an RLD in series with the bridge supply terminals.

That's all well and good, but the challenge appears when you include device parasitics as well. This capacitance resonates with the supply inductance. Analogously, device inductance is distributed around the bridge and resonates with the supply capacitance. In the voltage-fed circuit, you can treat these capacitances with one snubber, because they are in series. In the current-fed circuit, you can't snub the inductances because they aren't in parallel and you have to treat them individually.

The reason I bring all this up is, it works if you mutually couple the inductances. This would be difficult to arrange in practice for my example, but it's a necessary part of yours. Perhaps your answer lies in series-parallel combinations of capacitance as the equivalent to a mutually coupled inductor.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"Okkim Atnarivik"  wrote in message 
news:jerios$ojh$1@epityr.hut.fi...
> Phil Hobbs  wrote:
> : Do carry on--it reminds me to be grateful I'm not in middle
> : school any more.
>
>  My feeling exactly ( tinyurl.com/7dblxyf ).
>
>  I have been idly pondering, instead, how to make a Poincare dual
> out of the T-coil peaking circuit. You know, swapping I  U,
> C  L, swapping loops with nodes and so on. The required
> mutual coupling makes the straighforward conversion difficult.
> I guess I must write the transfer function and try to synthesize
> the circuit from scratch.
>
>  Regards,
>           Mikko

Is oo a big boy now?

Cheers

Phil Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203 Briarcliff Manor NY 10510 845-480-2058

hobbs at electrooptical dot net

100 kVA? That's round-off error where I come from. But seriously, that's a cool thing to wind by hand.

Provide the voltages of the primary and secondary and we'll see.

A quarter crusher coil doesn't count.

--
A hero steps down
and notices not his fall
into the abyss.

A representative transformer for me is a 500 - 230 kV autotransformer good for about 400 MVA or so. Now, to be honest, I don't design them or build them. My speciality is the protection systems that trips them offline when they fault. It was a bit of culture shock moving from university (a transformer can fit in your hand) to a utility environment (climb inside this transformer and take a look).

I've always been impressed by megawatt and gigawatt systems, by the sheer terror that must be associated with bringing them up the first time. Like the first flight of a 747. Big RF things, like megawatt AM stations, are like that too.

It's a nuisance for me to fill a coffee cup with blown mosfets, but blowing up a 400 MW transformer is a different league. Not to mention lights-out for, say, the entire East coast.

John

Like most other things, they don't just 'bring them up the first time'. There are methods of checking compliance and performance without actually producing irreversible damage. Full load becomes basically a thermal issue, as current and voltage stresses are partioned.

Mind you, there's always the first hook-up wiring 'proof' in the field......which is usually when the pretested breakers get their first reality check. You're familiar with those short interruptions in the lights?

RL

Then, it was not, of course, "wound by hand".

Quater crusher coils are huge power wise. The thing is it only lasts for a few milliseconds.

When I was in school, they showed us a film, the "greatest hits" from the LA power department test yard. Some of the transformer and breaker failures looked like the credits from Apocalypse Now. Boom.

John

just for fun, calculate the equivalent power flow for the typical gas station hose when you fill up your car at say 4 gallons of gas per minute......

Mark

OK. 33 KWH/gal, 4 GPM makes 132 KWH per minute, about 8 megawatts. That's a slow pump.

Doesn't sound good for electric cars, does it?

John

My memory is going in my old age. I seem to recall either 10 or 63 MW being the answer.

I had one experience where the electrical designer and the protection dedigner didn't agree on which way the circuit breaker was pointing, with the result that two sets of CTs were swapped and reversed polarity from the specs. When that circuit faulted (before commissioning was complete), there was a lot of head scratching before we could determine the fault location.