No reverse breakdown parameter in Spice?

P

Paul Burridge

Guest
Hi all,
Does Spice not model this behaviour? I tried inserting Br=-30 in the
model statement for an N-type JFET but it's come back as not
recongised. Yet diodes at least have it. Is this normal?

p.
--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
Paul,

Does Spice not model this behaviour? I tried
inserting Br=-30 in the model statement for
an N-type JFET but it's come back as not
recongised. Yet diodes at least have it. Is
this normal?
Click to expand...
Most transistor breakdown modes are not included
as part of the base SPICE transistor device. You
have to check the voltage waveforms to make sure
you haven't over-voltaged something.

--Mike
 
In message <4PNZc.10929$QJ3.7847@newssvr21.news.prodigy.com>, Mike
Engelhardt <nospam@spam.org> writes
Paul,

Does Spice not model this behaviour? I tried
inserting Br=-30 in the model statement for
an N-type JFET but it's come back as not
recongised. Yet diodes at least have it. Is
this normal?

Most transistor breakdown modes are not included
as part of the base SPICE transistor device. You
have to check the voltage waveforms to make sure
you haven't over-voltaged something.

--Mike


Breakdown modelling would be valuable for a spice that was primarily for
Click to expand...
switcher circuits. Would reduce the piles of dead ICs .
--
doug
 
Doug,

Most transistor breakdown modes are not included
as part of the base SPICE transistor device. You
have to check the voltage waveforms to make sure
you haven't over-voltaged something.

Breakdown modelling would be valuable for a spice
that was primarily for switcher circuits. Would
reduce the piles of dead ICs .
Click to expand...
Some SMPS's fail so quickly it's hard to see what made
them die. If the parts don't die in simulation, you
get those great storage scope waveforms to see whether
anthing's getting overvoltaged in simulation.
I think more useful if the parts don't die in simulation.

--Mike
 
On Fri, 03 Sep 2004 22:17:14 GMT, "Mike Engelhardt" <nospam@spam.org>
wrote:

Some SMPS's fail so quickly it's hard to see what made
them die. If the parts don't die in simulation, you
get those great storage scope waveforms to see whether
anthing's getting overvoltaged in simulation.
I think more useful if the parts don't die in simulation.
Click to expand...
LOL! Thanks for the laugh, Mike. I blew up several JFETs last Saturday
messin' about trying to measure their maximum parameters. If only
they'd been spice models. :)
That's another great advantage with spice that people often forget to
mention in the list of pros and cons: you get to destroy a part and
next time you hit "Run" it's as good as new!
--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
On Fri, 3 Sep 2004 23:10:39 +0100, doug <dd@ddwyer.demon.co.uk> wrote:

In message <4PNZc.10929$QJ3.7847@newssvr21.news.prodigy.com>, Mike
Engelhardt <nospam@spam.org> writes
Paul,

Does Spice not model this behaviour? I tried
inserting Br=-30 in the model statement for
an N-type JFET but it's come back as not
recongised. Yet diodes at least have it. Is
this normal?

Most transistor breakdown modes are not included
as part of the base SPICE transistor device. You
have to check the voltage waveforms to make sure
you haven't over-voltaged something.

--Mike


Breakdown modelling would be valuable for a spice that was primarily for
switcher circuits. Would reduce the piles of dead ICs .
Click to expand...
In PSpice (and I'm sure also in LTSpice) you can create subcircuits
that include breakdown effects.

I, personally, find it more useful to make up my own macros in Probe
that allow me easily to flag over-voltage, over-current and
safe-operating area violations.

(It may surprise some of you, but safe-operating-area violations can
impact short-channel/high-speed CMOS **LOGIC** (hot electron
effects)).

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
 
"Paul Burridge" <pb@notthisbit.osiris1.co.uk> schrieb im Newsbeitrag
news:93shj05svp1pgpqu89o4d6uggsco080vp3@4ax.com...
On Fri, 03 Sep 2004 22:17:14 GMT, "Mike Engelhardt" <nospam@spam.org
wrote:

Some SMPS's fail so quickly it's hard to see what made
them die. If the parts don't die in simulation, you
get those great storage scope waveforms to see whether
anthing's getting overvoltaged in simulation.
I think more useful if the parts don't die in simulation.

LOL! Thanks for the laugh, Mike. I blew up several JFETs last Saturday
messin' about trying to measure their maximum parameters. If only
they'd been spice models. :)
Click to expand...
Hello Paul,
I assume you have used the JFET it in a small signal application.
A 5$ calculator will be your friend to calculate power dissipation here.
Most people forget to do a few simple calculation of the junction
temperature.

Tj = Tamb + V*I*Rth

Rth is the thermal resistance from the datasheet.

If your calculated Tj reaches 150 degree, then it's very likely that
the JFET will die.


That's another great advantage with spice that people often forget to
mention in the list of pros and cons: you get to destroy a part and
next time you hit "Run" it's as good as new!
Click to expand...
Spice has that flexibility to implement the breakdown or self heating
with subcircuits.

Best regards,
Helmut
 
On Sat, 4 Sep 2004 01:16:29 +0200, "Helmut Sennewald"
<helmutsennewald@t-online.de> wrote:

Hello Paul,
I assume you have used the JFET it in a small signal application.
A 5$ calculator will be your friend to calculate power dissipation here.
Most people forget to do a few simple calculation of the junction
temperature.
Click to expand...
Hi Helmut,
No, I was measuring the spread of Idss in a single batch of J211s I
have. I rigged up a split supply for the purpose, consisting of a 30A
power supply on one side and a small 9V battery on the other. Yes, I
know it sounds dumb, but it would have worked fine if I hadn't
connected the 9V battery the wrong way around (D'oh!) both the FETs
were destroyed by having too much forward bias applied to their G/S
junctions. By the time I smelled burning, it was too late. :-(
Fortunately, however, they're very cheap. :)
--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top