In simple words, why is Darlington pair slow ? The author of "The art of electronics" states " the combination tends to act like a rather slow transistor because Q1 cannot turn off Q2 quickly".
And why does adding a resistor between the base and emitter of the second transistor improve performance?

 

There is no path from the second transistor’s base to ground. There is capacitance in the transistor that cannot be discharged quickly, so it stays on for a bit. A resistor to ground gives you the discharge path.

Bb

 

There is no path from the second transistor’s base to ground. There is capacitance in the transistor that cannot be discharged quickly, so it stays on for a bit. A resistor to ground gives you the discharge path.

Thank you, now it is clear.

 

Many darlington transistors already have the "speedup" resistors built-in:

 

Many darlington transistors already have the "speedup" resistors built-in:

What's the diode do? Flyback discharge path for junction capacitance?

 

The diode in a darlington transistor is usually never used. It is probably part of how it is made.

 

The diode in a darlington transistor is usually never used. It is probably part of how it is made.

Can you think of any purpose it might serve? I racked my brain and did a quick search but came up short.

 

Its not mentioned in the datasheet so almost certainly a parasitic from its construction and can't be used. Its probably shown on the schematic to indicate that it will appear in reverse-biassed Vce testing and so doesn't mean the device is faulty.

 

Its not mentioned in the datasheet so almost certainly a parasitic from its construction and can't be used. Its probably shown on the schematic to indicate that it will appear in reverse-biassed Vce testing and so doesn't mean the device is faulty.

By parasitic do you mean as they manufactured the part, a PN junction was formed that they are unable to eliminate? Ultimately what I'm getting at is, if there is no information on its (the diode) characteristic, how can I be sure it won't cause a conflict?

I have some cheap TIP120s so I'll be doing a few potentially destructive tests

 

By parasitic do you mean as they manufactured the part, a PN junction was formed that they are unable to eliminate? Ultimately what I'm getting at is, if there is no information on its (the diode) characteristic, how can I be sure it won't cause a conflict?

I have some cheap TIP120s so I'll be doing a few potentially destructive tests

Yes, that's not uncommon. In normal use its reverse-biassed so is invisible - not sure what conflict you think it could cause...

 

Yes, that's not uncommon. In normal use its reverse-biassed so is invisible - not sure what conflict you think it could cause...

No conflicts in particular, just 'what if' scenarios similar to the body diode in a MOSFET.

 

if there is no information on its (the diode) characteristic, how can I be sure it won't cause a conflict?

You eliminate any conflict by not reverse biasing the collector-emitter terminals (which you should never do in practice anyway, even for a non-Darlington BJT).
You are talking about a pathological condition which is of no interest in normal circuit operation.