How to achieve an almost constant value of VBE through spice model? What does the VJE parameter do?

 

What qualifies as "almost constant"? Over what range of base currents?

 

How to achieve an almost constant value of VBE through spice model? What does the VJE parameter do?

As I understand the documentation Vje along with Cje, and Mje are used to model non-linear depletion layer capacitances.
You can fix Vbe very tightly by using external components in the simulation. I may be able to find an example of that if you have some patience.

ETA: I looked through my archive and did not find what I thought I remembered.

 

What qualifies as "almost constant"? Over what range of base currents?

I want to fix a constant Vbe = 0.6V and beta = 300 for all three transistors in this circuit. Can you help me, please...

 

Why do you want to do that? If you do, the circuit won't work.

This is a current mirror, which works by having the transistor T1, being diode connected, establish the particular value of Vbe that corresponds to the current set up by V1 and R1.

That voltage is then applied as the Vbe of T2 and T3, which ideally establishes the same current in those transistors as T1 (assuming all three transistors are identical), which then results in twice the current flowing in R2 as is flowing in R1, subject to some deviation due to the Early effect and to T2 and T3 stealing some of the R1 current so that it doesn't feed T1's base.

This only works because changes in the R1 current result in a change in the Vbe. If you fix Vbe, then the left-side of the mirror can't communicate with the right side.

Imagine connecting a 0.6 V battery between the base node and your common. That will give you the fixed Vbe you claim you want. But now you can make all the changes to either side without have any effect at all on the other side.

So you need to carefully consider why you want to fix Vbe to a constant voltage.

 

Thank you so much

 

How to achieve an almost constant value of VBE through spice model? What does the VJE parameter do?

Most constant voltages can be modeled as a constant voltage source for a presumed unilateral current flow, which is probably what you have. I dont know why you would want to do this though maybe you could explain.
If your current can change direction then you could model it as a very very low forward voltage diode and a constant voltage source. That would clamp the voltage to be nearly the same as the voltage source voltage. Again i dont know why you would want to do this though.
In theory if the Vbe does not change the input current does not change unless you have something else that varies also that affects the transistor.