Avalanche breakdown

[Alain Portal]

Hi,

I'm a new ngspice user (and a new spice user) and I should want to know if the
avalanche breakdown effect of a bjt is simulated.

I'm trying to simulate a small part of the following circuit:
http://www.holmea.demon.co.uk/Avalanche/Avalanche.htm
(only +130V, R2, Q1, R7, R4, R8, R5)

Here is the description circuit:

* Pulse generator
.include 2N3904.ckt
VCC 1 0 DC 130
R1 1 2 100k
R2 3 0 3.3k
R3 4 5 22
R4 5 0 10
R5 5 6 47
C1 2 0 50p
Q1 2 3 4 2N3904
.tran 100n 1000U
.plot v(5)

2N3904.ckt
* Model generated on Aug 07, 01
* MODEL FORMAT: SPICE3
.MODEL 2n3904 npn
+IS=1.26532e-10 BF=206.302 NF=1.5 VAF=1000
+IKF=0.0272221 ISE=2.30771e-09 NE=3.31052 BR=20.6302
+NR=2.89609 VAR=9.39809 IKR=0.272221 ISC=2.30771e-09
+NC=1.9876 RB=5.8376 IRB=50.3624 RBM=0.634251
+RE=0.0001 RC=2.65711 XTB=0.1 XTI=1
+EG=1.05 CJE=4.64214e-12 VJE=0.4 MJE=0.256227
+TF=4.19578e-10 XTF=0.906167 VTF=8.75418 ITF=0.0105823
+CJC=3.76961e-12 VJC=0.4 MJC=0.238109 XCJC=0.8
+FC=0.512134 CJS=0 VJS=0.75 MJS=0.5
+TR=6.82023e-08 PTF=0 KF=0 AF=1

In the circuit, C1 replace the open-ended 50 ohms coaxial line.

Regards,
Alain

[Dietmar Warning]

Alain,
there is no avalanche breakdown modeled in the standard gummel poon model.
Please look for modern bipolar model documentations like vbic, mextram and hicum.
It is a script under /contrib/vbic which you can use to transform the existing GP model to vbic model as a starting point and the afterwards to play around with vbic model parameters avc1 and avc2 to fit the hopeful available avalanche characteristic data.
Nothing ready for a quick lunch - in other simulators too.
Sorry - Dietmar

[Holger Vogt]

Alan,

attached my simple test on the transistor, prepared according to Dietmars suggestions:

* Pulse generator
*VCC 1 0 DC 130
*R1 1 2 100k
*R2 3 0 3.3k
*R3 4 5 22
*R4 5 0 10
*R5 5 6 47
*C1 2 0 50p

*output characteristics to see if avalanche effect is modelled
IBE 0 3 1u
VEE 4 0 0
VCE 2 0 1
* QXXXXXXX nc nb ne <ns>
Q1 2 3 4 0 2N3904L4

* Model generated on Aug 07, 01
* MODEL FORMAT: SPICE3
* no avalanche effect included in model!!
.MODEL 2n3904L2 npn
+level=2
+IS=1.26532e-10 BF=206.302 NF=1.5 VAF=1000
+IKF=0.0272221 ISE=2.30771e-09 NE=3.31052 BR=20.6302
+NR=2.89609 VAR=9.39809 IKR=0.272221 ISC=2.30771e-09
+NC=1.9876 RB=5.8376 IRB=50.3624 RBM=0.634251
+RE=0.0001 RC=2.65711 XTB=0.1 XTI=1
+EG=1.05 CJE=4.64214e-12 VJE=0.4 MJE=0.256227
+TF=4.19578e-10 XTF=0.906167 VTF=8.75418 ITF=0.0105823
+CJC=3.76961e-12 VJC=0.4 MJC=0.238109 XCJC=0.8
+FC=0.512134 CJS=0 VJS=0.75 MJS=0.5
+TR=6.82023e-08 PTF=0 KF=0 AF=1

.model 2n3904L4 npn
* VBIC model parameters generated from above data by perl script sgp2vbic
* see http://www.designers-guide.org/VBIC/downloads.html
+ level=4
+ rcx  = 2.65711
+ rci  = 0.0
+ rbx  = 0.634251
+ rbi  = 5.203349
+ re   = 0.0001
+ is   = 1.26532e-10
+ nf   = 1.5
+ nr   = 2.89609
+ fc   = 0.512134
+ cje  = 4.64214e-12
+ pe   = 0.4
+ me   = 0.256227
+ cjc  = 3.015688e-12
+ cjep = 7.53922e-13
+ pc   = 0.4
+ mc   = 0.238109
+ cjcp = 0
+ ps   = 0.75
+ ms   = 0.5
+ ibei = 6.13333850374693e-13
+ nei  = 1.5
+ iben = 2.30771e-09
+ nen  = 3.31052
+ ibci = 6.13333850374693e-12
+ nci  = 2.89609
+ ibcn = 2.30771e-09
+ ncn  = 1.9876
+ vef  = 566.483710419377
+ ver  = 8.69028854282359
+ ikf  = 0.0272221
+ ikr  = 0.272221
+ tf   = 4.19578e-10
+ xtf  = 0.906167
+ vtf  = 8.75418
+ itf  = 0.0105823
+ tr   = 6.82023e-08
+ td   = 0
+ ea   = 1.05
+ eaie = 1.05
+ eaic = 1.05
+ eane = 1.05
+ eanc = 1.05
+ xis  = 1
+ xii  = 0.9
+ xin  = 0.668948
+ kfn  = 0
+ afn  = 1
* avalanche parameters added according to
* http://www.designers-guide.org/VBIC/documents/VbicText.pdf, page 13
* You have to play with both parameters to obtain suitable description of real transistor behaviour!
* This is just a guess!!
+ avc1 = 0.02
+ avc2 = 110

* initial condition for transient: V(2) set to 0
*.ic V(2) = 0

.control
*tran 100n 1000U
*plot v(2) v(5)
dc vce 0 150 0.1 ibe 100u 1m 100u
plot vee#branch
.endc

.end

Regards

Holger

[Alain Portal]

Hi,

Thank everybody for pointing me the VBIC model, link is now bookmarked (and some others too).
I succeed to create some others bjt models.

Thank Holger for the circuit file, I learnt a lot with it during this week.

But I failed to find suitable values for avc1 and avc2, circuit simulation isn't as expected.
Is there a way to evaluate them with some characterics form datasheet?

I made a try with a similar circuit using a 2n2369, result is worst…
Circuit is on the Linear Application Notes, page 93:
http://cds.linear.com/docs/Application Note/an47fa.pdf

* Pulse generator
.include 2n2369l4.model

VCC 1 0 DC 100
R1 1 coll 1Meg
R2 base 0 10k
R3 emit 0 50
C1 coll 0 20p

* QXXXXXXX nc nb ne <ns>
Q1 coll base emit 0 Q2N2369L4

* initial condition for transient: V(2) set to 0
.ic V(coll) = 0

.control
tran 1n 1U
plot v(coll) v(emit) v(base)
plot v(emit)
.endc

.end

.model q2n2369l4 npn
+ level=4
+ rcx  = .6
+ rci  = 0.0
+ rbx  = 10
+ rbi  = 0
+ re   = 0
+ is   = 1e-16
+ nf   = 1
+ nr   = 1
+ fc   = .5
+ cje  = 4.5p
+ pe   = .75
+ me   = .2418
+ cjc  = 2.83
+ cjep = 0
+ pc   = .75
+ mc   = 86.19m
+ cjcp = 0
+ ps   = 0.75
+ ms   = 0
+ ibei = 1.27681307456588e-18
+ nei  = 1
+ iben = 9.195e-15
+ nen  = 1.389
+ ibci = 7.88022064617809e-18
+ nci  = 1
+ ibcn = 0
+ ncn  = 2
+ vef  = -1.60440280094598e+16
+ ver  = -0.689700348076935
+ ikf  = .3498
+ ikr  = 0
+ tf   = 227.6p
+ xtf  = 4
+ vtf  = 4
+ itf  = .3
+ tr   = 1.073u
+ td   = 0
+ ea   = 1.11
+ eaie = 1.11
+ eaic = 1.11
+ eane = 1.11
+ eanc = 1.11
+ xis  = 3
+ xii  = 1.5
+ xin  = 0.9165
+ kfn  = 0
+ afn  = 1
+ avc1 = .02
+ avc2 = 100

Regards,
Alain

[Alain Portal]

http://cds.linear.com/docs/Application Note/an47fa.pdf

[Alain Portal]

Sorry, hoping this will work…
http://cds.linear.com/docs/Application%20Note/an47fa.pdf