Please explain this LtSpice sim of shorted diodes.

Version 4 SHEET 1 880 680 WIRE 176 32 64 32 WIRE 352 32 256 32 WIRE 64 80 64 32 WIRE 352 80 352 32 WIRE 432 144 352 144 WIRE 64 192 64 160 WIRE 352 288 352 272 WIRE 432 288 432 144 WIRE 432 288 352 288 FLAG 64 192 0 FLAG 352 352 0 SYMBOL voltage 64 64 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 10 SYMBOL res 272 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 1k SYMBOL diode 336 80 R0 WINDOW 3 -84 28 Left 2 SYMATTR InstName D1 SYMATTR Value 1N914 SYMBOL diode 336 144 R0 WINDOW 0 44 30 Left 2 WINDOW 3 -85 33 Left 2 SYMATTR InstName D2 SYMATTR Value 1N914 SYMBOL diode 336 208 R0 WINDOW 0 46 31 Left 2 WINDOW 3 -80 31 Left 2 SYMATTR InstName D3 SYMATTR Value 1N914 SYMBOL diode 336 288 R0 WINDOW 0 -35 36 Left 2 WINDOW 3 40 27 Left 2 SYMATTR InstName D4 SYMATTR Value 1N914 TEXT 30 296 Left 2 !.tran 0 0.010 0 startup

its not that I am complaining how ever, I've seen generate noise beyond beleif from 2 diodes in series when placing a jumper wire around them. I've notice it does not seem to do it with a default diode but sure does do different things depending on which diode you selected from the LIB>

Kind of wondering if this is the source of some noise I just can't seem to find when testing real live circuits. The other one is ringing on inductors which does not seem to show up in real time, but we'll save that for later.

Jamie

Reply to
Jamie
Loading thread data ...

I get reasonable simulation results when I include representative parasitics. Stray inductance and junction capacitance are the most important. Dampening is also quite important as SPICE diodes tend to snap and make lovely 50MHz+ squigglies that gobble timesteps.

Real diodes have charge distribution and transition time effects that can't be approximated with one variable capacitor, but as a model of "if you get recovery problems, they'll look something like this", it's somewhat representative.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
 Click to see the full signature
Reply to
Tim Williams

nap

Search on snap-off diodes and step-recovery diodes

formatting link

formatting link

IIRR the charge carriers in the diode can organise themselves to create a very abrupt change of didoe current when the stored charge finally runs out - it shows up on static measurements as a particular form of capacitance dependence on reverse voltage as in hyper-abrupt varactor diodes

formatting link
/hyperabrupt.php

formatting link

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

Although hyperabrupt junctions are the easiest to use (they'll practically do it under constant current conditions, who needs dI/dt or inductors), many diodes will do it under the right conditions.

I read a paper that investigated the properties of "soft recovery" diodes. Under conditions of high dI/dt, high inductance and high voltage, they can exhibit not only snap recovery, but also dynamic avalanching. It's like yanking a cork with a bungee cord and having it bound between smacking your face and slapping the bottle, very painful.

Under rapid recovery conditions, the drift region begins to clear of charge carriers from the P or N side, or both. When dI/dt is especially high, the charge clearance is local, leaving a cloud on the opposite side, or in the middle. As a result, the diode isn't carrying current anymore, because there isn't a complete circuit across it, but the junction is effectively shorter because there's still a cloud of charge in the middle (or to a side) shorting it out. Thus, until that charge clears, suddenly applying high voltage puts way more electric field across some portions of the drift region, leading to true avalanche behavior, at voltages much lower than the static avalanche rating.

Tim

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

y
.
n
e
y
,

e,

ly

f

That's interesting. I'd not heard of it before - it sounds as if it is primarily of interest to people using diodes as fast rectifiers, which I've never been involved in - and I'd guess that it has improved my mental model of the semiconductor diode. Time alone will tell if that's right.

Have you got a reference to the paper?

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

formatting link
$file/copyright_ieee_01mr.pdf

Big IGBT modules tend to be slow and clunky, but they're slowly getting faster. The dI/dt you can pull gets pretty interesting. Discrete IGBTs have been fast enough for a while, and the tests in this paper are at fairly low current (5, 10A/div), so it's not unusual territory by any means.

Tim

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

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.