Simulation Questions:

Martin Czech martin.czech at intermetall.de
Wed Oct 6 09:20:12 CEST 1999


:::Can someone explain to me what the transistors that are drawn with
:::multiple emitters (on data sheets) actually are???  For example I'm
:::looking at the LM334 constant current source IC from National Semi.
:::There are NPN and PNP drawn like this...

They are what they are ;->
Common collector and base and multiple diffused emitters.
You need the model of such a device. For simple analysis,
just use parallel transistors. But this doesn't work, if
the different emitters have different emitter area (which is
used to partition current between the multiple emitters).

:::I'm playing with temperature sweep analysis and want to see if this
:::behaves like reality...
:::
:::Also... does anyone know how to model tempco resistors ??? My program
:::allows ohms/deg C.   Just make this 3300 ppm ??

Yes, a linear tempco is ok for 1st order.

:::
:::How about zener models... Data sheets suggest that a zero tempco can be
:::had at 4-6 volts for most devices... but the simple models don't seem to
:::work like this... I expect the lower voltages to have a opposite tempco
:::from the high voltage devices...
:::they dont !

There are two effects of diodes in reverse mode operation: Zener
(avalanche) for voltages >5V, diode voltage drops with temp and tunneling
for voltages <5V, diode voltages rises with temp.  Arround 5.3V they
should theoretically cancel. Such diodes suffer from aging, surface
effects etc. Don''t use them for precision circuits.  Better use a
bandgap or other type of "reference diode".

m.c.




More information about the Synth-diy mailing list