Rbe test circuit?

[Martin Czech]

:::After thinking about it, I come to the conclusion that the collector must
:::be open,
:::since the Early effect would change the Vbe too.
:::I think perhaps the way would be to draw the Vbe/Ib courve on log paper.
:::(Did I say paper?!, I mean gnuplot or something!) For an ideal device I'd
:::expect a straight line (just Ebers Moll). In real devices for higher base
:::currents there should be a deviation from the straight line. But this would
:::be difficult to measure, since I think the deviation perhaps is already
:::swamped in the error bar. At some hundred uA of base current the input
:::resistance is maybe still in the range of some hundred Ohm.
:::Btw, Would it be perhaps possible to get the value for R(be) from the
:::h-parameters?
:::Like extrapolating h(ie)vs Ic for Ic -> inf. If Ic (thus Ib) is very large,
:::and therefore the B-E diode's dV/dI very small, the remaining value would
:::be the static Rbe (???).
:::

Ok, I said collector current because that's what you really want.
So perhaps one should do a linear diode plot (because the rbe will be the
linear thing in the plot). And maybe another plot with collector current,
one can varry collector voltage to get an idea of the Early then.

Getting parameters is difficult, you can really just get plots
and try to fit the theoretical values of theoretical devices
afterwards. 

And I guess you need quite accurate measurement gear.


Perhaps this all can be avoided if one uses an oscillator for time
integration of the current, whith low leakage high value cap ( will
eliminate time errors) and known low Ibias buffer (will eliminate
current error).

m.c.