The P400 DDG box display (not all that easy to read either) shows the VH and VL open-circuit voltages from the 50-ohm generator. The diode is across the transmission line from the pulse gen to the 50 ohm scope. There's a 20 dB attenuator on the sampling head input.
So the diode sees a 25 ohm source, at half the voltage on the DDG display.
OK. That gives me enough info to "play". ...Jim Thompson
When we did our DSRD pulser, we found a lot of variation in diodes that had nominally the same part number. We made a deal with Mouser to sample us diodes and, if we liked them, sell us the rest of the same reel.
The best DSRD/Grehkov diode turned out the be the C-B junction of an NPN horizontal-output transistor, now obsolete.
...
Tom McEwan patented a circuit which avalanched a 1N4004, producing 100KW pulses in the 100ps region. It required a 1000V step of a few ns (mosfet) followed by a PFN. I mentioned it here a few years back, and it got quite a bit of discussion including multiple posts by Win Hill, and IIRC it was after that that JL actually tried it.
You're probably referring to a pulser that uses a DSRD (opening switch) to drive an avalanche diode (closing switch). I've done the DSRD circuit, to make 2KV at about 3 ns width, but I haven't done the avalanche stage.
I was aware of the McEwan work, but my pulse generator was based on Grehkov's original papers.
McEwan has an enormous number of largely silly patents. He seemed to be a patent addict.
Hopefully he didn't do it during the Super Bowl, or his neighbours would have been after him with pitchforks.
Cheers
Phil Hobbs
See "A Simple Diode Model with Reverse Recovery" Lauritzen and Ma
- this model is implemented in the Simetrix simulator.
It's wa-a-ay past the weekend >:-}
The forward recovery _is_ trivial to model and I've figured a way to keep it from tangling with the reverse recovery... but the reverse recovery mechanism is non-trivial to model... still scratching my... ;-) ...Jim Thompson
[snip]
Does this look to be viable enough...
If so I'll continue and see what it takes to parameterize it to specific diodes.
(100% Behavioral Modeled... 6 lines in the subcircuit >:-) ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
Interesting, but I just told the potential customer to jump in the nearest lake. They played the tired old trick, "we don't want to buy your products, but how would you like to be a consultant?" In other words, they want to own our IP cheap.
Life is such fun ;-) ...Jim Thompson
Current seems about right for a controlled ramp, but I notice dI/dt is infinite (hard switched to a ~dead short). And accordingly, there's no Dirac spike from CJO, or the conduction during t_rr (though t_rr will be quite short under such dramatic conditions).
What's it do with a VPULSE + 25 ohms?
Tim
-- Seven Transistor Labs Electrical Engineering Consultation Website:
"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...
Not so good. I need to more scratching my... >:-} ...Jim Thompson
A new pass at it...
What I need is someone to get a simultaneous scope photo of the above three traces from a real device... all these PhD papers I can find conveniently leave out pertinent data. ...Jim Thompson
-- Seven Transistor Labs Electrical Engineering Consultation Website:
Hey, that's not too bad...
Comically quick lash-up:
The CT is 25 turns on a T37-43 core, or around 200uH. And thus Ch.2 is reading 0.5 A/V. Yes, DC bias is a concern, but the AC waveform seems roughly correct, at least. Reflected resistance is 0.08 ohms, so can be ignored.
Source voltage (as terminated). Warning, the amp is probably not exactly
50 ohms internally, so the Thevenin equivalent is probably not exactly 25 ohms.Results: with the slow vintage diode ("169 001"),
Testing with a diode in series with the CT (so it is able to recover between pulses), turn-on still has a slow current rise but no voltage overshoot. Trying a 1N4007, I see a very slight voltage overshoot, but about the same current rise. (And around 600ns t_rr.) So, characteristic of the circuit I think; maybe the amplifier can't push all that hard, or all that fast. (The response is more-or-less symmetrical with the diode reversed.)
Tim
-- Seven Transistor Labs Electrical Engineering Consultation Website:
OK. I'll toy around with it and see if I can figure how to parameterize the subcircuit. ...Jim Thompson
That's a bit odd; didn't the base spreading resistance get in the way?
Or, did you short B-E and let the emitter be your connection?
What better time, than when there is nothing on TV that's worth watching? TV sports have sucked, since the '70s.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required