The collector current of BJT

On what parameters does the Ic collector current of a transistor depend upon, base emitter voltage or base current?

I know that BJT is a current controlled device. But still we have graphs that show that Ic changes when Vbe changes.

Which one should I consider?

• Have you read: en.wikipedia.org/wiki/Bipolar_junction_transistor? When Vbe changes, will Ib stay constant or change as well? What does that mean for Ic? It is not a matter of considering one or the other, they're all related. Remember that the BE junction is a diode. – Bimpelrekkie Jan 27 '19 at 13:05
• Yes - they are related. However - the question is legitimate: Which quantity is the physical cause of Ic control? My answer: The voltage Vbe only. Or do you think that - for example - two charged carriers (base current change) can cause the release of 1000 additional charged carriers from the emitter (assuming a beta value of 500)?. – LvW Jan 27 '19 at 13:12
• Indeed, few charged carriers in base can lead on average to (beta +1) times as much carriers to be released from the emitter, beta times of which are swooped up by the collector. Why? Because when you introduce a carrier of one type in the base, the emitter will have to supply one carrier of opposite type to neutralize it. But this carrier does not have a map to reach the offending carrier - it will cross the forward biased junction and hope for the best. The probability it will (contribute to) neutralize the carrier is small, because most carriers diffuse rapidly and end up in the collector – Sredni Vashtar Sep 25 '19 at 14:13
• You can find a description of this role of the base current, among the others, on Streetman, "Solid State Electronic Devices", and in Levinshtein & Simin, "Transistors: from crystals to integrated circuit". Note: this does not imply that the BJT is exclusively current controlled, since as many other have said, voltage and current are concomitant and not one the cause of the other. – Sredni Vashtar Sep 25 '19 at 14:18

• @LvW I actually like the parsimonous way Spehro wrote this out. For small-signal behavior, the BJT is operated in active mode, it's collector acts more like a current source/sink, and $g_m$ matters. For switch behavior, the BJT is operated in saturated mode, it's collector acts more like a voltage source, and sufficient recombination current is the main concern. – jonk Jan 28 '19 at 4:16