On 3/26/07, David Gravereaux <[EMAIL PROTECTED]> wrote:
> I'll have to read-up on how to make my own models from the datasheet info.  
> Any
> pointers how to do it?  This is new to me :)

I'm not an expert. What I did was pick the part in the LTSpice library
that had the closest specs. Then, copied the specs from the component
file (e.g. C:\Program Files\LTC\SwCADIII\lib\cmp\standard.bjt) to the
schematic page (using the ".op"
 tool on the menu bar). The part number and specs can then be
modified. In the Help file, under Circuit Elements it explains briefly
what the parameters are.

At the bottom of this post I've included an example of a simple BJT
amplifier. Theoretically you should be able to put that into a file
such as amp.asc and be able to view and run it. If there are problems,
try taking out the newlines and "\n+" in the last two TEXT lines.

The default model for the the 2N3055 transistor doesn't account for
the Earley effect (defaults to infinity). Nevertheless, there is a
slight ripple in the load current corellated to power supply ripple,
perhaps due to C-B capacitance. I added the forward Earley voltage
parameter (Vaf=100 - no idea if that's a realistic value) and the load
current ripple gets a lot larger.

> What I am trying to do is model why my circuit is so sensitive to power supply
> ripple.  It's a 100-watt all FET (mos and j) power amplifier that uses just 10
> devices.  At first, I thought it was matching transconductance of the 
> front-end
> jFETs, but doesn't appear to be the case.   The mixing in the driver stage 
> should
> cancel but isn't for an unknown reason.  And that's why I'm here..  soldering 
> just
> sucks :)
>
> I have the voltage sources set for +/-54V with a 4mV sinewave ripple (should 
> be
> saw, but I'm fine with that), and I do get ripple on my output just like the 
> real
> circuit, so I guess the model is accurate to my needs.  Next is changing 
> around
> parts and values 'till I find the proper cancel.

This may be drifting off topic, but I'd be curious to see your circuit
and learn, along with you, about whatever the problem is with your
circuit. Maybe you need to feed the input stage with a constant
current source rather than a resistor. Or maybe you need some negative
feedback.

One general comment about "designing with SPICE": DON'T DO IT.
Tweaking component values and running countless simulations is very
time-inefficient, and can lead to false "solutions". For example, you
might find some parts values that cancel out your ripple when your
circuit is at 25 degrees C, but not when it warms up a bit. The better
approach is to *understand* the problem and do your calculations by
hand (or a hand calculator or spreadsheet), using the simplest circuit
component models that will work, and designing the circuit to work
over the full range of component parameters (e.g. min to max
transconductance, min to max Vt...).

There is a very active LTSpice Yahoo group that is very good for
getting questions about LTSpice answered. There are also some
newsgroups that discuss electronics.

>
> Is there a way to do batch and compare with LTSpice such that I can modify a 
> part
> value within a range, sim all results, and compare what I got in a graph?

Yup, see the .STEP command.

Cheers,
- Jan

=== following is an LTSpice circuit model .asc file ===

Version 4
SHEET 1 1116 680
WIRE 304 16 -80 16
WIRE 304 48 304 16
WIRE -80 176 -80 16
WIRE 304 192 304 128
WIRE 240 240 144 240
WIRE 144 272 144 240
WIRE 304 304 304 288
WIRE -80 384 -80 256
WIRE 144 384 144 352
WIRE 144 384 -80 384
WIRE 304 384 144 384
WIRE -80 400 -80 384
FLAG -80 400 0
SYMBOL voltage -80 160 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value SINE(20 1 120)
SYMBOL voltage 144 256 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V2
SYMATTR Value 1
SYMBOL npn 240 192 R0
SYMATTR InstName Q1
SYMATTR Value 2N3055x
SYMBOL res 288 32 R0
SYMATTR InstName R1
SYMATTR Value 4
SYMBOL res 288 288 R0
SYMATTR InstName R2
SYMATTR Value .1
TEXT -114 506 Left 0 !.tran 1
TEXT 8 416 Left 0 !.model 2N3055   NPN(Bf=73 Br=2.66 Rb=.81 Rc=.0856
Re=.000856\n+  CJC=1000P PC=.75 MC=.33  Tr=.5703U Is=2.37E-8\n+
CJE=415P PE=.75 ME=.5 TF=99.52N NE=1.26 IK=1 Vceo=60 Icrating=10
mfg=STMicro)
TEXT 8 512 Left 0 !.model 2N3055x   NPN(Bf=73 Br=2.66 Rb=.81 Rc=.0856
Re=.000856 Vaf=100\n+  CJC=1000P PC=.75 MC=.33  Tr=.5703U
Is=2.37E-8\n+  CJE=415P PE=.75 ME=.5 TF=99.52N NE=1.26 IK=1 Vceo=1
Icrating=10 mfg=STMicro)

Reply via email to