Hey All,

New user looking to be educated...

I recently installed LTSpice-XVII, mainly just to play (I used a similar program years ago running under linux, maybe the same, I don't remember).

I drew out the schematic of an audio amplifier that I built (many years before circuit simulators) just to see how the simulation compared to the real-world measurements.
And wanting to duplicate the characteristics of the original (mostly obsolete) transistors, I set out to either find the correct parts or define them myself.
This seems that it should be as simple as adding lines to the "standard.bjt" file and plugging in the numbers, but this is where I get confused....

When I open the "standard.bjt" file as a text file, I see lines of transistor numbers followed by its characteristics (I use a common 2N3906 as reference).

But what do these numbers mean?
++++++++
.model 2N3906 PNP(IS=1E-14 VAF=100 BF=200 IKF=0.4 XTB=1.5 BR=4 CJC=4.5E-12 CJE=10E-12 RB=20 RC=0.1 RE=0.1 TR=250E-9 TF=350E-12 ITF=1 VTF=2 XTF=3 Vceo=40 Icrating=200m mfg=Philips)
++++++++

That is, when I compare it to a typical datasheet (as below), I see "Vceo", but not much else that matches up....

Are there new characteristics defined that I don't know about?
(it was recently pointed out to me that we don't use T/S parameters for speakers anymore, we use .frd & .zma files)...

Regards,
JohnR

++++++
2N3906 / MMBT3906 / PZT3906 — PNP General-Purpose Amplifier
© 2010 Fairchild Semiconductor Corporation www.fairchildsemi.com
2N3906 / MMBT3906 / PZT3906 Rev. 1.2.2 3
Electrical Characteristics
Values are at TA = 25°C unless otherwise noted.
Note:
4. Pulse test: pulse width ≤ 300 μs, duty cycle ≤ 2.0%.
Symbol Parameter Conditions Min. Max. Unit
OFF CHARACTERISTICS
V(BR)CEO Collector-Emitter Breakdown
Voltage(4) IC = -1.0 mA, IB = 0 -40 V
V(BR)CBO Collector-Base Breakdown Voltage IC = -10 μA, IE = 0 -40 V
V(BR)EBO Emitter-Base Breakdown Voltage IE = -10 μA, IC = 0 -5.0 V
IBL Base Cut-Off Current VCE = -30 V, VBE = 3.0 V -50 nA
ICEX Collector Cut-Off Current VCE = -30 V, VBE = 3.0 V -50 nA
ON CHARACTERISTICS
hFE DC Current Gain(4)
IC = -0.1 mA, VCE = -1.0 V 60
IC = -1.0 mA, VCE = -1.0 V 80
IC = -10 mA, VCE = -1.0 V 100 300
IC = -50 mA, VCE = -1.0 V 60
IC = -100 mA, VCE = -1.0V 30
VCE(sat) Collector-Emitter Saturation
Voltage
IC = -10 mA, IB = -1.0 mA -0.25 V
IC = -50 mA, IB = -5.0 mA -0.40
VBE(sat) Base-Emitter Saturation Voltage IC = -10 mA, IB = -1.0 mA -0.65 -0.85 V
IC = -50 mA, IB = -5.0 mA -0.95
SMALL SIGNAL CHARACTERISTICS
fT Current Gain - Bandwidth Product IC = -10 mA, VCE = -20 V,
f = 100 MHz 250 MHz
Cobo Output Capacitance VCB = -5.0 V, IE = 0,
f = 100 kHz 4.5 pF
Cibo Input Capacitance VEB = -0.5 V, IC = 0,
f = 100 kHz 10.0 pF
NF Noise Figure
IC = -100 μA, VCE = -5.0 V,
RS = 1.0 kΩ,
f = 10 Hz to 15.7 kHz
4.0 dB
SWITCHING CHARACTERISTICS
td Delay Time VCC = -3.0 V, VBE = -0.5 V
IC = -10 mA, IB1 = -1.0 mA
35 ns
tr Rise Time 35 ns
ts Storage Time VCC = -3.0 V, IC = -10 mA,
IB1 = IB2 = -1.0 mA
225 ns
tf Fall Ti
++++++

 

Hi

The transistor "model" contains "spice" parameters. There is not a direct 1=1 relationship between datasheet specs and the spice parameter names. Rather than try to explain what could be a complicated topic, I will point you a book that can answer most of your questions.

https://www.wiley.com/en-us/The+SPICE+Book-p-9780471609261

 

Hi

The transistor "model" contains "spice" parameters. There is not a direct 1=1 relationship between datasheet specs and the spice parameter names. Rather than try to explain what could be a complicated topic, I will point you a book that can answer most of your questions.

https://www.wiley.com/en-us/The+SPICE+Book-p-9780471609261

Alright, thanks, I'll check that out....
Seems that now that I'm old and retired, I'm finding things completely different than I remember from my youth...

 

The SPICE parameters that go into a ".model" card depend on the particular model being used. One of them is the Gummel-Poon Model. Here is a handy online reference:

https://en.wikipedia.org/wiki/Gummel–Poon_model

The simpler model is called the Ebers-Moll after the fellows who invented it.

https://en.wikipedia.org/wiki/Bipolar_junction_transistor#Ebers–Moll_model

When you're ready to move on to MOSFETS, the number of different models is staggering.

 

Yes, Spice transistor parameters and data sheet transistor parameters are sort of apples and oranges.
Rather than trying to understand and use the Spice transistor parameters, which require specialized measurements to generate and cannot generally be derived from the data sheet, an easier way would be to just look for transistors in the LTspice library that more-or-less match the parameters of the transistor used in your amplifier.
That would likely be close enough to give a good simulation of your amp since the performance of well-designed audio amps is generally not significantly affected by other differences in the BJT transistor parameters if they have voltage, current, and gain ratings that are similar.

 

Rather than trying to understand and use the Spice transistor parameters, which require specialized measurements to generate and cannot generally be derived from the data sheet, an easier way would be to just look for transistors in the LTspice library that more-or-less match the parameters of the transistor used in your amplifier.

Yes, that is what I'm doing now, just thought I'd see what it took to get the correct part.

I did find some spice models in post on this and other forums, and while I am able to add them to the standard.bjt file, when I go to "pick a new transistor", the spice model is there, but lacking some of the description before the spice model.
How do I go about editing that (manuf, Vceo, Ic)??

 

You go to the standard.bjt file in the LTC directory -- \LTC\LTspiceIV\lib\cmp\ standard.bjt.
Then look for that particular transistor definition.

 

Yes, that is what I'm doing now, just thought I'd see what it took to get the correct part.

I did find some spice models in post on this and other forums, and while I am able to add them to the standard.bjt file, when I go to "pick a new transistor", the spice model is there, but lacking some of the description before the spice model.
How do I go about editing that (manuf, Vceo, Ic)??

View attachment 258201

There are descriptive attributes for the model:
Vceo=50 Icrating=.100 mfg=eetech00)

Where format is:
Vceo in Volts
Icrating in Amps
mfg is free text.

These attributes are descriptive only and don't affect the models' behavior.
Be aware this file is replaced with each software update.
I don't recommend editing the standard.bjt file.

Instead, you can create a single file containing each of your model definitions, then store it in a known folder location.
Then place a directive on your schematic:

.inc "known folder path"\mymodels.txt (without quotes)

LTspice will look in "known folder path"\mymodels.txt for the model definition.

If mymodels.txt is in the same folder as the schematic file (.asc) then the folder path is not needed.

 

Be aware this file is replaced with each software update.

That's why I still using LTspice IV with no updates.

 

That's why I still using LTspice IV with no updates.

Sorry...but not a very good reason....

 

Sorry...but not a very good reason....

Well, your welcome to that opinion, but it works well for me.

 

Well, your welcome to that opinion, but it works well for me.

Touche!

 

@eetech00 said:
Be aware this file is replaced with each software update.

In my experience going back to SwitcherCad III. released in 1999, the update process does a merge between the existing file and the updates. All of the changes I have ever done to standard.bjt are still there from decades ago. The same is true of standard.dio
There is literally no reason to be concerned about this.

 

@eetech00 said:
Be aware this file is replaced with each software update.

In my experience going back to SwitcherCad III. released in 1999, the update process does a merge between the existing file and the updates. All of the changes I have ever done to standard.bjt are still there from decades ago. the same is true of standard.dio
There is literally no reason to be concerned about this.

I guess you were one of the lucky ones..
I stand by what I recommended. Avoid editing the standard.bjt

 

I guess you were one of the lucky ones..
I stand by what I recommended. Avoid editing the standard.bjt

I wasn't lucky. Everybody gets the same deal. There are still valid reasons for compartmentalizing things, but you can pretty much do as you will.
It was the case in 1999, it is still the case in 2022, and I suspect it will continue to be the case that updates perform a merge and NOT a replacement.

 

It was the case in 1999, it is still the case in 2022, and I suspect it will continue to be the case that updates perform a merge and NOT a replacement.

Well, that is what was claimed but hasn't been consistent. The files weren't always merged.....why? ask the developers.
I still won't do it and don't recommend it.

 

Well, that is what was claimed but hasn't been consistent. The files weren't always merged.....why? ask the developers.
I still won't do it and don't recommend it.

I never got that memo. Not really sure when the events you are describing happened.

 

There are descriptive attributes for the model:
Vceo=50 Icrating=.100 mfg=eetech00)

Where format is:
Vceo in Volts
Icrating in Amps
mfg is free text.

These attributes are descriptive only and don't affect the models' behavior.

Ok, I found it, I just have to append the line with the info before the last ")",
I.E., ~ MJC=.2 TF=19.8N TR=765N Vceo=140 Icrating=20a mfg=Motorola)

I didn't figure it affected the model, but is a nice reference when scrolling down the list...

Thanks for everyone's input...

JohnR