As in pulse generation avalanche?
The only guaranteed parts that I know of are the Zetex's, FMMT417 and such.
Imagine making a 300 volt, 60 amp, 18 kilowatt pulse from a SOT23.
If you mean just protective high-voltage zenering behavior, there are lots of those.
Know of any NPNs with guaranteed avalanche characteristics? (bad joke: do they fly if they have a good SOAR?)
Might that exist?
Well, put it as adjustable high-voltage zenering behavior where the base drive will change the avalanche curve. I have wasted almost an hour trying to find datasheet curves; i guess one must DIY as they do not exist. Thanks.
Have run into dozens that do that in the dim past. Curves roughly like:
1) Ib=0 curve Ic=0 to avalanche, then slow switchback as source voltage is increased (yes, negative resistance then almost vertical line to a point. 2) As Ib drive is increased, the Ic increases as in a normal transistor; the switchback voltage decreases, the negative slope decreases (ie more vertical). 3) At sufficient drive, there is no switchback and all slope are positive. 4) Curve set has, "in the limit", an "intercept" point at some large voltage and current. As long as the transistor is not over-driven, and does not get hot, this is repeatable and reliable for a given transistor. Too long ago to say if a given 2N and manufacturer could be selected and used (fake example: 2N2222 by TyltTex Semi always works).
Are you describing avalanche, or mistaking normal BVceo behavior for avalanche? ...Jim Thompson
The Zetex parts are commonly base-triggered, which sorta corresponds to the above. The cool thing about the Zetex's is that, once they fire, they basically short themselves to near zero volts for a while, even at 10s of amps, whereas most other avalanche transistors only blip down a fraction of Vcc.
I was told that the FMMT parts are fabbed in Russia, on an ancient diffusion line. Good avalanche transistors seem to be low Ft diffused parts, not modern fast epi stuff.
Old Tek sampling scopes used a selected 2Nsomething as a fast avalanche pulse generator, to drive the sampling diodes. They got to about 2 GHz bandwidth that way, in the 7S14. Their faster samplers used an avalanche transistor to drive an SRD.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required