[Elecraft] Help! Rescue me from stupid error!

[Ian White]

>
>George: Not to put too fine a point on it but ZNRs are large and slow
moving
>so they do have mechanical aspects but more to the point they do NOT
have
>capacitance, save lead capacitance (a few pf), 

Not true: MOVs have a self-capacitance on the order of 1000pF, which is
sometimes an advantage.  The surge limiting components with a very low
capacitance are spark gaps, but those have quite different
characteristics.

MOVs are bidirectional devices which are "soft limiting", so they are
not as effective as transzorbs in limiting any incoming voltage spikes.
On the other hand, MOVs are bulk-effect devices which are capable of
absorbing much more surge energy than the semiconductor junctions in the
transzorbs. 

The main function of MOVs is to "take a bullet" for the more sensitive
(and valuable) components downstream. However, this also involves some
cumulative damage to the crystal structure and eventually an MOV will
either become ineffective or fail (failure could involve either an
upward or a downward shift in the limiting voltage, ending in either an
open or a short circuit). These failure modes can be an acceptable price
to pay for the  energy absorbing capabilities, so long as both the
designer and the user are aware of the possibilities.
 
The most effective protection strategies will combine two or more
different types of device to exploit each of their best features. One
example is to install a big MOV first in line, followed by a series
impedance (usually a mains filter) and then a fast device such as a
transzorb to clip any residual spikes. 

Another very simple tip is to install all of these supply protection
components AFTER the switch, so they are not  exposed to surges when
equipment is not in use.



73 from Ian GM3SEK