[time-nuts] TAPR TICC boxed (input protection)

David davidwhess at gmail.com
Wed Apr 5 13:18:18 EDT 2017


On Wed, 5 Apr 2017 02:40:13 -0400, you wrote:

>David wrote:
>
>> So collector-base junctions make good low leakage high voltage diodes
>> although they are slow
>
>I guess it depends on what one means by "slow" and "fast."

I was referring to within the same transistor; emitter-base junctions
are much faster than collector-base junctions.

>The B-C junction of an MPSH10/MMBTH10 or 2N/PN/MMBT5179 switches on in 
><1nS and off in <2nS, which is comparable with Schottky microwave mixer 
>diodes such as the Agilent HSMS282x series and better than "ultra-fast" 
>silicon switching diodes such as the FD700 and 1S1585.  (I did my 
>switching tests at 20mA.)  (Note that the silicon and Schottky switching 
>diodes have reverse leakage currents from several hundred to tens of 
>thousands of times higher than the B-C junction of an MPSH10/MMBTH10.)

I have never actually tried this with RF transistors.  I know one
thing to watch out for if you are looking for low leakage is gold
doping and some less that reputable manufacturers "cheat" in this
respect so transistors used as low leakage diodes should be at least
qualified by manufacturer which is a problem with counterfeits and
unscrupulous purchasing managers.

Out of curiosity, and I tried to look this up years ago, what doping
is used for PNP RF transistors and saturated switches if it is not
gold?  Does it also increase leakage?

And I have another question if you know.  How is rb'Cc measured?
Tektronix at some point was grading 2N3906s for rb'Cc < 50ps.

>The gate junction of a 2N/PN/MMBF4117A JFET switches on in <2nS and off 
>in <4nS.
>
>> The cheapest guaranteed low leakage diode is probably some variety of
>> 4117/4118/4119 n-channel JFET.
>
>If the 5pA reverse leakage current of the MPSH10/MMBTH10 is too much and 
>one must, must, must get leakage down to 1pA, the 2N/PN/MMBF4117A is the 
>best inexpensive choice that I'm aware of.

The advantage of the 4117/4118/4119 is that the leakage is already
tested to a given specification so no qualification or testing is
necessary.


More information about the time-nuts mailing list