>>Vin ------ +ve relative to ground >> | >> R >> R R load >> R >> | >> _______ >> | | >> Meter R >> | R Re >> E R >> B--------| NPN >> C R >> | R Rc >> | R >>-----.|------------| >> >..... >>However, when an emitter-base resistor is added (Re) operation becomes >>"unusual". >>Note that an b-e resistor will usually hold a transistor >>What appears to happen is that the junction breakdown voltage is reduced. >> > >Russell, I am working thru your expanation, and here is a partial answer. > >Not really "unusual", as adding Re [Rc open] is exactly the factor >that takes the NPN from acting as passive back-to-back diodes to >operation in "active" inverse mode. I don't believe this has any effect >whatsoever on "junction breakdown voltage [is reduced]", but rather >the transistor is amplifying Yes & no. Many people have proposed the inverse transistor model which of course does apply to this situation. However, when Rc is added and Re is not the result is very similar (slightly lower current). In this case the resistor is a B-E resistor for the effective inverted transistor and 'should" act to turn it off. The best (proof" that something unusual is happening is this - "If the original transistor had the c & e connections physically swapped, would it behave in a similar manner?" If it does then the inverse transistor model is an adequate explanation. If it doesn't, then something else is also happening. Now, if the transistor physically had c & e connections swapped and Rc added as above then the resistor would now be between true base and true emitter and the transistor would be normally biased across c-e and the transistor would be turned off and no current would flow. But, with the inverse transistor, current does flow. Put this way it's clear to me NOW (not then) how to demonstrate whether something odd is in fact happening. I'll try to perform the above very simple test which should definitively prove or disprove whether there is a fundamental difference between the inverse transistor and a normal one. (Non fundamental differences such as beta which have already been discussed, are accepted). RM