Spice models needed

[edvard]

Hey all, I've been all over the 'net looking for spice models for the BC239 and BC309 transistors.
Either that, or I need to know exact replacements that I can use models for.

Also, I've found models for some germanium transistors, can't guarantee accuracy but I bet somebody here could verify.
I'll post them when I get home.

Hey! While we're at it, how about we turn this thread into the spice model request thread?
Any takers?

[alanlan]

I'd just use the standard NPN model or something close and copy it and tweak the model parameters.  You have to remember that a device model only reflects one specific instance of a part whereas a part number like BC309 or whatever is a specification with an amount of variation about a mean or min/max for the given parameters.  For the most part, the standard NPN/PNP models will serve you pretty well.

[edvard]

I kinda figured as much, with gain being the biggest factor.
My problem is, I'm working on a circuit that uses BC239/309 transistors and it sounds radically different with anything else, so I'm curious if a spice simulation might give me a peek at what's going on.

I guess I could always work up something from Analog Service's excel thingummy or find an exact replacement transistor...
Back to Google...

[CynicalMan]

According to this site: http://homepages.which.net/~paul.hills/Circuits/Spice/ModelIndex.html
the BC239 model can be found on the labcenter electronics site and on the cadence orcad page.

[edvard]

Both those pages turned up nothing.
Maybe I should just find a decent replacement.
I'm only using those cause I got a good deal on them a few years back.
If they're discontinued, it's a small wonder I got such a good deal...

[Joe]

Can you post the gain (Hfe) of the transistors, maybe from several parts? That would help find some matches.

[brett]

Hi
I suspect the answer is simple.
There are 3 classes of hFE for the BC239: 120 ~ 220 180 ~ 460 380 ~ 800
Clearly, modern BJTs are usually above the low range.
There's also a big "knee" in the hFE.  All in all, it's perfect for Q1 in a Fuzzface.

Datasheet 1:http://www.datasheetcatalog.org/datasheet/MicroElectronics/mXuvuyv.pdf
Datasheet 2:http://www.datasheetcatalog.org/datasheet/fairchild/BC239.pdf
cheers

[brett]

Hi again
Maybe I jumped to conclusions....
The Ic vs Ft graph indicates (by extrapolation) that the transition frequency falls to hell as Ic falls to a few uA.  At the Ic of some circuits (e.g. Q1 in the fuzzface again), the Ft will probably be limiting the high frequency gain.

Does it sound mellow?
cheers

[edvard]

Well, yes I guess you could say 'mellow'...
I used them in a Percolator mod and they sounded "softer" than other transistors of similar gain.
I think I've tried them in a fuzz face and I got lots of oscillation. 

[George Giblet]

I'm with alanlan on this one.

If you want to model the parts you have to measure the parts you have.
If you create various plots of your parts you will see they can be quite
different from the datasheet in some areas.

You can play with these models - I cannot vouch for them in any way shape or form!

*BC309B MCE
*Ref: Motorola Small-Signal Device databook, Q4/94
*Si 350mW 25V 100mA 360MHz Amp pkg:TO-92B 3,2,1
.MODEL BC309B PNP (IS=10.2F NF=1 BF=260 VAF=90 IKF=60M ISE=5.04P NE=2
+ BR=4 NR=1 VAR=20 IKR=90M RE=0.515 RB=2.06 RC=0.206 XTB=1.5
+ CJE=12.6P VJE=1.1 MJE=0.5 CJC=6.93P VJC=0.3 MJC=0.3 TF=442P TR=307N)

* Siemens
.MODEL qbc309b PNP (
+ IS=17.2542F NF=1 BF=364.356 IKF=83.76M ISE=1.1917P NE=1.973
+ VAF=261.2426 RC=0.55 RB=20 IRB=20.94M RBM=5 RE=13.75M
+ VAR=70.2743 IKR=0.3805 ISC=0.2431P NC=1.9789 NR=0.92 BR=32
+ CJC=8.7P VJC=0.75 MJC=0.4 CJE=9.81P VJE=0.75 MJE=0.4
+ TF=0.53N XTF=0.443 ITF=92.471M VTF=13.0621 TR=43.9282N
+ XCJC=0.893 FC=0.504 XTB=1.6163 XTI=2.643 )

*BC239BP ZETEX Spice model     Last revision  4/90   Low Noise
.MODEL BC239BP NPN(IS=1.8E-14 BF=400 NF=0.9955 VAF=80 IKF=0.14 ISE=5E-14
+ NE=1.46 BR=35.5 NR=1.005 VAR=12.5 IKR=0.03 ISC=1.72E-13 NC=1.27 RB=0.56
+ RE=0.6 RC=0.25 CJE=1.3E-11 TF=6.4E-10 CJC=4E-12 VJC=0.54 TR=5.072E-8 )

* Siemens
.MODEL qbc239b NPN (
+ IS=17.2542F NF=1 BF=364.356 IKF=83.76M ISE=1.1917P NE=1.973
+ VAF=261.2426 RC=0.55 RB=20 IRB=20.94M RBM=5 RE=13.75M
+ VAR=57.2121 IKR=0.3525 ISC=0.3087P NC=2.0105 NR=0.92 BR=32
+ CJC=8.7P VJC=0.75 MJC=0.4 CJE=9.81P VJE=0.75 MJE=0.4
+ TF=0.68N XTF=0.459 ITF=97.0778M VTF=13.0621 TR=46.1166N
+ XCJC=0.659 FC=0.559 XTB=1.6163 XTI=2.959 )

[edvard]

Thank you George!
Yes, I'm quite aware that actual performance may be 'off' what I see in simulation, and many times simulation doesn't let me 'see' what I'm hearing.
I mostly use simulation to get a ballpark view of what should be happening in my circuit before I build it.
When searching for the equivalent of a transistor I don't have, I've found that some alternatives can just drop in any given circuit and some won't even 'wake up' without major tweaking of the original circuit.
I'd rather sort that out before I start freaking that a cap is in the wrong spot or something.
I'm finding that my beloved BC239/309 pieces have some very nice peculiarities that I'm hard pressed to find equivalents for.
No mojo there, just plain 'ol math...

Thanks Brett, I never noticed the Ic/Ft curve before, I'll have to check that when looking for replacements.