Modeling of BJT Ic vs Vce characteristics in Pspice

[Viren]

Hi,
I have two experimental graphs of Ic vs Vce for two different BJTs.I
need to use Pspice to model the output characteristics of BJT. I would
like to know if there a current controlled current source in Pspice
which I can vary according to base current IB which would also control
Ic current. I would appreciate your help urgently in this matter.
Viren

 

[Tim Wescott]

Hi,
I have two experimental graphs of Ic vs Vce for two different BJTs.I
need to use Pspice to model the output characteristics of BJT. I would
like to know if there a current controlled current source in Pspice
which I can vary according to base current IB which would also control
Ic current. I would appreciate your help urgently in this matter.
Viren

Yes there is, I think a controlled current source is 'G' but that may
only be for voltage-controlled current sources.

SPICE transistor models are pretty obscure, but if I remember correctly
the Early voltage* gets entered directly into the model, so you
shouldn't have to dink with funky sources to model its effect.

* Do a web search if you don't understand what a BJT's Early voltage is.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

 

[Joerg]

Hello Viren,

I have two experimental graphs of Ic vs Vce for two different BJTs.I
need to use Pspice to model the output characteristics of BJT. I would
like to know if there a current controlled current source in Pspice
which I can vary according to base current IB which would also control
Ic current. I would appreciate your help urgently in this matter.

Yes, there is. "F" is the current controlled current source.

Hope this helps,, Joerg

 

[Jim Thompson]

Hello Viren,




Yes, there is. "F" is the current controlled current source.

Hope this helps,, Joerg

But "F" only allows a fixed "gain" to be set.

If you want a behavioral current source use "G" which has allowed
expressions for some time now.

...Jim Thompson

 

[Joerg]

Hello Jim,

But "F" only allows a fixed "gain" to be set.

If you want a behavioral current source use "G" which has allowed
expressions for some time now.

Oops, must have misunderstood, I thought a fixed gain is what he wanted.

Anyway, this looks like one of those homework questions. If true, it
would be a nasty homework, forcing students to model with current
sources. I have seen transistors modeled this way only once and it was
quite confusing. The regular model statements make more sense.

Regards, Joerg

 

[Jim Thompson]

Hello Jim,


Oops, must have misunderstood, I thought a fixed gain is what he wanted.

Anyway, this looks like one of those homework questions. If true, it
would be a nasty homework, forcing students to model with current
sources. I have seen transistors modeled this way only once and it was
quite confusing. The regular model statements make more sense.

Regards, Joerg

Some 20 years ago now my oldest son and I wrote an executable (in
Pascal) that would take IC and IB data versus VBE and VCE and find all
the model DC parameters.

Nasty set of non-linear iteration.

...Jim Thompson

 

[Winfield Hill]

Jim Thompson wrote...

Some 20 years ago now my oldest son and I wrote an executable (in
Pascal) that would take IC and IB data versus VBE and VCE and find
all the model DC parameters. Nasty set of non-linear iteration.

Hey, that sounds like a very useful program, how about posting it
for the gang?

 

[Jim Thompson]

Jim Thompson wrote...

Hey, that sounds like a very useful program, how about posting it
for the gang?

I'll try. It's probably been almost that long that I stopped using
it, since the foundries started modeling devices themselves. But I
vaguely recall seeing it somewhere... probably on the CD where I
backed up all my 5¼" floppies ;-)

I probably also have the mathematical development somewhere on paper.
An employee of mine at GenRad (Phoenix), John Spellman, did all the
cranking.

...Jim Thompson

 

[Winfield Hill]

Jim Thompson wrote...

I'll try. It's probably been almost that long that I stopped using
it, since the foundries started modeling devices themselves. But I
vaguely recall seeing it somewhere... probably on the CD where I
backed up all my 5¼" floppies ;-)

I probably also have the mathematical development somewhere on paper.
An employee of mine at GenRad (Phoenix), John Spellman, did all the
cranking.

Much appreciated.

 

[Jim Thompson]

Jim Thompson wrote...

Much appreciated.

Just as expected, found it on my 5.25" floppy archive CD done in 2000.

Amazing how small stuff was back in the late '80's, early '90's ;-)

The modeling stuff is at...

http://analog-innovations.com/SED/NE.zip

This is Pascal, and I don't remember how it works... but it obviously
runs in DOS.

Since my oldest son wrote it at age 18-19, while he was at U of A, it
is extraordinarily well documented in-line with the code.

I haven't found Spellman's derivations yet. But think about it, if
you plot IC and IB (log axis) versus VBE (linear), you will get
straight lines until you hit crowding (IKF) or, on the low current
end, recombination (NE, and associated terms, from whence the name of
the program ;-)

...Jim Thompson

 

[Winfield Hill]

Jim Thompson wrote...

The modeling stuff is at...
http://analog-innovations.com/SED/NE.zip
Great!

This is Pascal, and I don't remember how it works... but it
obviously runs in DOS.

And wants a Hercules adapter, although perhaps most graphics
cards provide that functionality? The program faults, with a
"Turbo Graphix error #1", "font file missing in InitGraphic".

Since my oldest son wrote it at age 18-19, while he was at U of A,
it is extraordinarily well documented in-line with the code.

Jim, what about the Spanish-language mode in the program?

 

[Jim Thompson]

Jim Thompson wrote...

And wants a Hercules adapter, although perhaps most graphics
cards provide that functionality? The program faults, with a
"Turbo Graphix error #1", "font file missing in InitGraphic".

I didn't try running it... I'm still re-installing taxes on the new
machine :-(

Jim, what about the Spanish-language mode in the program?

Is there really a "Spanish-language mode" there? I can't say that I'm
surprised, Aaron speaks fluent Spanish and Portuguese... my Hispanic
son-in-law doesn't ;-)


...Jim Thompson