LTSpice, creating a .model for a 2N6286?

2N6286 is an old pnp power darlington transistor,
100V/20A, with a forced hfe of 100, TO-3 package.

I'd like to simulate the (PWM) circuit it is used in,
with LTSpice. An RCA data sheet is available, but
how do I create a .model for it?

Thanks,

PS: I'm thinking of a Hitachi 2SJ217 as a replacement
so need some sort of model for that as well.

Actually 2SJ217 is only a 60V, and I would really
like at least a 100V device. Suggestions welcomed
for a P-type MOSFET, TO247/TO-3P/SOT-93 package,
100V, and as much current capability as the package
will take (to blow a fuse that has a nasty habit of
being replaced by something bigger)....... brick
outhouse time.

 

A Darlington is really two transistors, so you use a subcircuit to
model it.

Study beta versus current and the base-emitter resistors to establish
characteristics of each device.

...Jim Thompson

 

Doubt if it will help much, but FWIW the closest CircuitMaker's
library gets to a power PNP darlington is the 2N6042. Its model is as
follows:

*100V 8A PNP Darlington Transistor
..SUBCKT X2N6042 1 2 3
Q1 1 2 4 QMOD .1
Q2 1 4 3 QMOD
R1 2 4 8K
R2 4 3 120
D1 1 3 DMOD
..MODEL QMOD PNP(IS=9.6E-12 BF=96.6 VAF=180 IKF=4 ISE=6.41E-10 NE=2
+ BR=4 VAR=20 IKR=6 RB=0.2 RE=0.085 RC=0.063 CJE=1.65E-9 VJE=0.74
MJE=0.45
+ TF=1.4E-8 CJC=2.38E-10 VJC=1.1 MJC=0.24 TR=7E-7 VJS=0.75 XTB=1.5 )
..MODEL DMOD D(IS=9.6E-12 RS=0.05 TT=7E-7 CJO=2.38E-10 BV=100 )
..ENDS X2N6042

And the placed part looks like this (after some zooming):
http://www.terrypin.dial.pipex.com/Images/2N6042.gif
(They must have been low on solid dots when they drew it.)

Here are the CM PMOS P-Channel MOSFETs, >= 100V and >= 10A:

IRF9532 100V 10A .25ohm Power MOSFET pkg:TO-220 2,1,3
IRFI9540G 100V 11A .109ohm Power MOSFET pkg:TO-220 2,3,1
IRF9530S 100V 12A .1ohm Power MOSFET pkg2PAK 2,1,3
RF1S9530 100V 12A .25ohm Power MOSFET pkg:TO262AA 2,1,3
RF1S9530SM 100V 12A .25ohm Power MOSFET pkg:TO263AB 4,1,3
MTP12P10 100V 12A .3ohm Power MOSFET pkg:TO-220 2,1,3
IRFIP9140 100V 15A .08ohm Power MOSFET pkg:TO-247 2,3,1
IRF9142 100V 15A .15ohm Power MOSFET pkg:TO-3 3,1,2
IRF9542 100V 15A .15ohm Power MOSFET pkg:TO-220 2,1,3
IRF9540S 100V 19A .0632ohm Power MOSFET pkg2PAK 2,1,3
RF1S9540 100V 19A .15ohm Power MOSFET pkg:TO262AA 2,1,3
RF1S9540SM 100V 19A .15ohm Power MOSFET pkg:TO263AB 4,1,3
IRF9140 100V 19A .15ohm Power MOSFET pkg:TO-3 3,1,2
IRF9540 100V 19A .2 ohms HEXFET pkg:TO-220 2,1,3
IRFP9140 100V 21A .0571ohm Power MOSFET pkg:TO-247 2,3,1
IRFP9243 150V 10A .38ohm Power MOSFET pkg:TO-247 2,1,3
IRF9641 150V 11A .35ohm Power MOSFET pkg:TO-220 2,1,3
IRF9241 150V 11A .35ohm Power MOSFET pkg:TO-3 3,1,2
IRFP9241 150V 12A .38ohm Power MOSFET pkg:TO-247 2,1,3
IRFD9210 200V ?A 3ohm Power MOSFET pkgIP4 1,4,3
IRFP9242 200V 10A .38ohm Power MOSFET pkg:TO-247 2,1,3
IRF9640S 200V 11A .109ohm Power MOSFET pkg2PAK 2,1,3
IRF9240 200V 11A .35ohm Power MOSFET pkg:TO-3 3,1,2
RF1S9640SM 200V 11A .35ohm Power MOSFET pkg:TO263AB 4,1,3
RF1S9640 200V 11A .35ohm Power MOSFET pkg:TO262AA 2,1,3
IRF9640 200V 11A .5 ohms HEXFET pkg:TO-220 2,1,3
IRFP9240 200V 12A .1ohm Power MOSFET pkg:TO-247 2,3,1

 

Have you tried the LTSpice users' group on Yahoo, Tony? That'd be your
best bet if Googling's failed, I reckon.

 

Ah yes, it has to be a subcircuit, not a model.
I didn't think of that.

I can do that ok, but my knowlege of Spice is so pitiful Jim
that I don't understand what most of the little acronyms mean.

 

[snip details]

Thanks for the subcircuit Terry. It looks a good starting
place that I might be able do something with.

I'll visit later today (job to get out urgently).

[snip list]

That one looks a possibility, thanks for that.

 

I'm totally useless at Googling (and Yellow Pages).
My keywords are never their keywords.

 

Absolutely! I just put in the words "at" and "are", and it totally
failed to find your post. What's all that about?

d
Pearce Consulting
http://www.pearce.uk.com

 

Better, here's one I found with Google:

**************************************
* Model Generated by MODPEX *
*Copyright(c) Symmetry Design Systems*
* All Rights Reserved *
* UNPUBLISHED LICENSED SOFTWARE *
* Contains Proprietary Information *
* Which is The Property of *
* SYMMETRY OR ITS LICENSORS *
* Modeling services provided by *
* Interface Technologies www.i-t.com *
**************************************
..SUBCKT 2n6286 1 2 3
* Model generated on Jan 24, 2004
* Model format: PSpice
* Darlington macro model
* External node designations
* Node 1 -> Collect
* Node 2 -> Base
* Node 3 -> Emitter
Q1 1 2 4 qmodel
Q2 1 4 3 q1model 7.81859
D1 1 3 dmodel
R1 2 4 8000
R2 4 3 50
* Default values used in dmodel
* EG=1.11 TT=0 BV=infinite
..MODEL dmodel d
+IS=1e-12 RS=5.85795 N=0.958579 XTI=2.95858
+CJO=0 VJ=0.75 M=0.33 FC=0.5
..MODEL qmodel pnp
+IS=1.69855e-12 BF=221.401 NF=0.961395 VAF=133.261
+IKF=0.262058 ISE=2.84033e-12 NE=1.5434 BR=0.940867
+NR=1.07398 VAR=36.7145 IKR=0.0998515 ISC=1e-13
+NC=1.95857 RB=14.9267 IRB=0.158579 RBM=14.9267
+RE=0.125332 RC=0.62666 XTB=0.384111 XTI=2.95173 EG=1.05
+CJE=4.25427e-10 VJE=0.4 MJE=0.334318 TF=1e-09
+XTF=1 VTF=10 ITF=0.01 CJC=6.15611e-10
+VJC=0.95 MJC=0.289483 XCJC=0.9 FC=0.5
+TR=1e-07 PTF=0 KF=0 AF=1
..MODEL q1model pnp
+IS=1.69855e-12 BF=221.401 NF=0.961395 VAF=133.261
+IKF=0.262058 ISE=2.84033e-12 NE=1.5434 BR=0.940867
+NR=1.07398 VAR=36.7145 IKR=0.0998515 ISC=1e-13
+NC=1.95857 RB=14.9267 IRB=0.158579 RBM=14.9267
+RE=0.125332 RC=0.62666 XTB=0.384111 XTI=2.95173 EG=1.05
+CJE=4.25427e-10 VJE=0.4 MJE=0.334318 TF=1e-09
+XTF=1 VTF=10 ITF=0.01 CJC=0
+VJC=0.95 MJC=0.289483 XCJC=0.9 FC=0.5
+TR=1e-07 PTF=0 KF=0 AF=1
..ENDS

 

If you can't directly find a model for a Darlington, try to determine
what transistor family it's from. Then use that device model as the
basic core and place into Terry's listing:

..SUBCKT 2Nxxxx 1 2 3
* Model format: PSpice
* Darlington macro model
* External node designations
* Node 1 -> Collect
* Node 2 -> Base
* Node 3 -> Emitter
Q1 1 2 4 qmodel 0.1 << vary Q1, Q2 areas to match
Q2 1 4 3 qmodel 1 << data sheet curve trace
D1 1 3 dmodel 1 << vary area of diode for reverse conditions
R1 2 4 8000 << from data sheet
R2 4 3 50 << from data sheet
*
..MODEL dmodel d (....
..MODEL qmodel pnp (....
..ENDS 2Nxxxx

Note that Q1 and Q2 would have the same model name, since the
diffusions are the same. Although there _could_ possibly be
difference in IKF between the two depending on the layout.

...Jim Thompson

 

I'm glad to see that Terry posted a model for you.

And although you are talking about a Darlington and not a simple BJT, I'd like
to add that there's a book on the subject of BJTs that goes into great detail
about how to develop each of the model parameter values from physical
measurements that are achievable with generally available equipment. It
includes all of the theory to apprehend all this, as well. It's Ian Getreu's
"Modeling the Bipolar Transistor." It originally came out in 1976 and the 3rd
printing I have came out in 1979. I don't know of anything quite like it,
before or since. If anyone knows of a book that covers that subject as well
from a more modern perspective, I'd love to hear about it.

Jon

 

Thanks Terry. I also received a model via email
which is very similar. It doesn't look much of
a problem to edit the models into LTSpice syntax.

 

Yes thanks. I've now got two example models to work from
so should be able to produce something reasonable.

 

How do you define "generally available equipment"? And does the text
show how to use this equipment to arrive at accurate values for even
the more obscure and esoteric spice parameters and package paractics?

 

Well, I've not gone entirely through that half of the manual (it is in two main
sections -- theory and then measurement of the theoretical parameters.) But the
experimental setups and the required measurements look to me mostly doable with
meters, power supplies, noise source, and an appropriate oscilloscope. For
example, in the case of r'b, he says:

"Traditionally, r'b has been a difficult parameter to measure, mainly because it
is modeled as a lumped constant resistance although it is actually a
distributed, variable resistance. As a result, the value obtained for r'b
depends strongly on the measurement technique used as well as the transistor's
operating conditions. Several measurement techniques are described here.

"Comparison of measurement techniques for r'b

"r'b should be determined by the method closest to the operating conditions
under study. If r'b is being measured to ascertain its effect on noise
performance, the 'noise measurement technique' should be used. Similarly, if
the transistor is to be used in a switching application, the 'pulse measurement'
technique may provide the most appropriate value. For small-signal analysis
four measurement techniques are described: the 'input impedance circle method,'
the 'phase-cancellation technique,' the 'two-port network method,' and the 'h-y
ratio technique.' A comparison of the advantages and disadvantages of these
four methods is given in Section (c), the small-signal measurement section."

And then he goes on to describe these and then many more techniques for each and
every other parameter of all the models up through the GP model parameters.

Jon

 

On Wed, 27 Oct 2004 18:01:56 GMT, Jonathan Kirwan

[snip]

Thanks, Jonathan. Sounds very interesting indeed. I've done a search
on E-bay but unfortunately it appears there are no copies available
for sale anywhere in the world. I shall just have to check back
regularly and hope!

p.

 

What really, really gripes the heck out of me is that I haven't been able to
find a new book that covers these details for modern semiconductors with the
comprehensive nature of this book. I can find selected *this* and selected
*that*, but nothing that is comprehensive and complete and thorough and detailed
so that one can sit down and do these things for themselves and also have some
idea about how to select from among various options to follow through with.

This book is on the BJT, though some of it applies to any PN junction, I think.
But I've nothing like it for MOSFETs, JFETs, PUJTs, and various other multi-PN
junction things like SCRs, TRIACS, and so on. Some of this book could be
re-examined and re-thought to apply to many of those, I suppose. But I'd like
an expert who did that, not me.

And in all these years, with various modifications and improvements discussed
regarding BJTs, nothing new to replace this book on even this sole subject of
BJTs.

Why? Especially when SPICE is so much more available, now?

As a hobbyist, I'm mad at those who are keeping all these 'secrets' proprietary.
(If that's what is happening, of course.)

Jon