[time-nuts] TAPR TICC boxed (input protection)
davidwhess at gmail.com
Tue Apr 4 20:02:38 EDT 2017
On Wed, 5 Apr 2017 09:13:34 +1200 (NZST), you wrote:
>A protection diode needs to also have a fast turn on with little or no overshoot of the forward voltage.
That would be ideal but forward turn on time is rarely specified and
usually assumed to be fast and some fast diodes have appallingly slow
turn on. This is one of those things that needs to be qualified or
selected for if it is important.
I suspect there is some obscure processing issue with diodes that
causes slow turn-on that does not show up in transistors.
>Reverse recovery time can be an issue if one is relying on the clamp for protection against a periodic overload such as when an input is overdriven by a sinewave input and one wishes to make useful measurements whilst this occurs.
>The internal protection diodes of HCMOS devices can severely degrade the device propagation delay jitter when they conduct.
They sure can but isn't this because of minority carrier injection? I
wonder if this is only a problem with junction isolated integrated
circuit processes. I probably knew at one point but forgot.
Dual and quad analog ICs can suffer from a different problem where
exceeding the common mode input voltage range screws up common bias
circuits causing other elements to malfunction.
More information about the time-nuts