Go Down

Topic: advanced transistor operation (Read 3862 times) previous topic - next topic

ralphd

I'm trying to find some information on advanced transistor operation, in particular reverse polarity behavior of NPNs.
I've recently learned that for some transistors, reverse breakdown between the emitter and base is non-destructive.
http://jlnlabs.online.fr/cnr/negosc.htm
However my (untested) belief is Vceo is a destructive limit.  For example if I connect the emitter of a 2n3904 to 100v and the collector to ground, I believe I'll fry it, since Vceo is 40V.
But for Vcbo is it non-destructive like Vebo?
Lastly I've been trying to find out more about reverse-polarized operation where the base voltage is higher than emitter, and with the collector voltage higher than the base.  According to transistor theory all transistors will work in this mode, and if I understand it correctly gain is always <1, since collector current is always the sum of base and emitter current.
I did some tests with a 2n3904 and the reverse polarized voltage drop (Vec-sat) was ~1v, compared to ~0.1v for Vce-sat forward polarized.  The datasheet lists 0.2V as the maximum for Vce-sat with 10mA of collector current, but I don't see anything in the datasheet for reverse polarized.  Does anyone know where I can find that information (short of actually testing the transistors)?

runaway_pancake

The article doesn't state that back-biasing a base-emitter junction is absolutely "non-destructive."
Like any diode there definitely is a limit, a reverse breakdown voltage.
The article does state that there is a region, a range, within which a certain effect can be appreciated.
"Who is like unto the beast? who is able to make war with him?"
When all else fails, check your wiring!

JChristensen

P-N junction breakdown doesn't mean anything actually breaks, or that magic smoke is released, etc.

From http://en.wikipedia.org/wiki/P%E2%80%93n_junction#Reverse_bias:

Quote
The strength of the depletion zone electric field increases as the reverse-bias voltage increases. Once the electric field intensity increases beyond a critical level, the p-n junction depletion zone breaks down and current begins to flow, usually by either the Zener or the avalanche breakdown processes. Both of these breakdown processes are non-destructive and are reversible, as long as the amount of current flowing does not reach levels that cause the semiconductor material to overheat and cause thermal damage.

ralphd

OK, but when I look at diode i-v curves in avalance breakdown, the reverse voltage does not decrease with an increase of current - unlike the so-called negative resistance of the collector-base junction.  Is there some diodes that behave this way, or is it for some reason unique to transistors?

MarkT

If you hit reverse breakdown with enough current, the transistor is toast.  Limit the current
enough and its basically just a zener diode (because just like a zener the base-emitter
junction is heavily doped and a very narrow depletion zone).   The base metalization and
bond wires are not designed for large currents like those of the emitter and collector, note.

I've taken out a 15A power NPN transistor with negative Vbe, just a short pulse was enough...
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

westfw

Quote
Is there some diodes that behave this way

Sure.  Pretty much all diodes can operate "backwards."  Zener diodes are a particular type that is DESIGNED to operate that way.
You usually have to have some other mechanism for limiting current (and thus power dissipation) to a safe amount.
Reverse-biased base/emitter junctions are sometimes used as random noise generators: http://holdenc.altervista.org/avalanche/

ralphd


Quote
Is there some diodes that behave this way

Sure.  Pretty much all diodes can operate "backwards."  Zener diodes are a particular type that is DESIGNED to operate that way.

But a zener stops conducting when the voltage drops below the breakdown voltage, but with the reverse emitter-base junction it will continue to conduct if the voltage drops below Vebo.  Conduction only stops when the current drops to near zero.
Is that the difference between zener breakdown and avalanche breakdown?

Grumpy_Mike

Quote
But a zener stops conducting when the voltage drops below the breakdown voltage

No, look at the I/V curve of a real device.

MarkT

Low voltage zeners (genuine zeners) have a very soft knee.  Higher voltage devices
(which are just avalanche-breakdown devices disguised as zener-effect diodes) have
sharper knees.

If you want a decent sharp response then use a zener buffered with a BJT - resistor R from
emitter to base, zener from base to collector.  The zener current is approx 0.6/R, so you
can choose which part of the curve, and once the collector is more than Vz+0.6V the
BJT turns on and takes most of the current, buffering the zener voltage.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

westfw

Quote
but with the reverse emitter-base junction it will continue to conduct if the voltage drops below Vebo

Are you sure?  There are 4-layer devices that behave that way (tunnel diodes, "sidactors", SCRs in breakdown), but normal reverse-biased single junctions behave more like zeners.  The original article you linked is reverse-biasing the CE connection (three layers, I guess...)

ralphd


Quote
But a zener stops conducting when the voltage drops below the breakdown voltage

No, look at the I/V curve of a real device.

Are you being pedantic and referring to the normal near-zero reverse leakage of all diodes, or something material that I'm missing?

ralphd


Quote
but with the reverse emitter-base junction it will continue to conduct if the voltage drops below Vebo

Are you sure?  There are 4-layer devices that behave that way (tunnel diodes, "sidactors", SCRs in breakdown), but normal reverse-biased single junctions behave more like zeners.  The original article you linked is reverse-biasing the CE connection (three layers, I guess...)


I wasn't paying close enough attention to that  (reversing CE not BE) - thanks for pointing that out.  So the reverse emitter base junction on a 2n3904 should act like a 6V zener.  The reverse CE, however, should not.
Today I found another article that says basically the same thing, and shows some nice scope shots of the zener vs reverse transistor knee.
http://cr4.globalspec.com/thread/72501

westfw

Sidactors are cool, BTW.  You can convert a disposable camera flash to a repeating strobe with one component.
Somewhat similar components are used for overvoltage protection of telco circuits (and lots of other things), but they have someone different IV curves...

Go Up