Re: question: how can input ac phase be delayed

```Some more ideas...

If you hook a variable transformer between two phases, the phase of the
voltage on the tap varies between the two source phases.  This technique is
used with three phase supplies and three variacs to generate a three phase
source with arbitrary phase relative to the supply.

So, some nifty combination of capacitors and a variac would probably do the
trick.

>Yes, but John also wrote that Lou doesnīt have access to anything but
>single phase power sources. Plus, the real problem I see with multi-
>phase driving, in this case, is the fact that the phase to phase
>relationship would always be 120°. Iīm not quite sure if this would
>solve the problem (the problem of RSG imperfection is still there.) It
>is just shifted an additional 120°. I also donīt know if Lou wants to
>have that much delay.
>
>For a few degrees of phase delay, I think a variable reactor or using a
>cap bank in series (maybe with several taps for different "delays") with
>one of the PSUs is an easy method. As Lou isnīt running gobs of power,
>I think a cap-delay-circuit would be a transportable solution. If you go
>up in input VA you would start needing giant caps as they have to let
>the required power pass through them. Ideally they would only "move"
>the phase, but in reality they will cause some power losses (dissipated
>in the form of heat).
>
>Another idea that comes to mind (armchair analysis!!) is to use two
>different breakrates in the RSGs. However, I am not quite sure if
>they would have to be "exact" multiples of 2x Fmains (i.e. 240bps,
>360 bps, etc). This would be the same as running it on different
>phases (i.e: ~120° separation). If the xformers are NOT NSTs or
>other fragile xformers, I think you could also use totally different
>breakrates (i.e: non-sync) in the two RSGs (esp. to find out what
>RSG to RSG "breakrate ratios" give the best results).
>
>A last idea: How about using different sized caps in the two tank
>circuits. I.e: one is a resonant sized (or almost) cap and the other
>is a (slightly) LTR cap. Or even two LTR caps of different size
>(easy to do with an MMC containing "taps"). That way you could
>phase both (maybe you would need a variable motor mounting to
>change the rotor position) so that you get optimum spark length
>and then "reverse engineer" the reasoning behind the "why is it
>getting a better spark length"
>

```