[time-nuts] Why are 1PPS signals so skinny?

[Mike S]

On 5/14/2012 8:21 PM, Tom Van Baak wrote:
> one day during an experiment where I was
> comparing a large set of clocks I noticed my lab's digital AC power
> meter was jumping by tens of watts every second.
>
> The last thing you want
> in a precision timing lab is to load your AC line down exactly once a
> second.

How does a short pulse help? It's still "tens of watts every second," 
but instead of lasting 0.5 seconds, it lasts 0.00005 seconds. Less power 
used overall, but still the same sudden change on the second.