[Delphinus]

Dez, he either has to make a modification, or he has to upgrade his pump if he doesn't want to touch the skimmer itself. I'm sure a swimming pool pump would run that skimmer just fine as-is. If it was me, I'd be looking at both pump and design: a mag18 draws 145W and a Mak4 draws 110W, so the Mak4 is cheaper to run, AND I think will perform a little better on a larger beckett skimmer (although it is not going to solve the problem with the skimmer as it is now). So I stand by my comments, although, for sure, it's not the complete solution.

Sean, I think one possible key thing to focus on here is the length of the injector assembly. If you really want to avoid shortening the reaction chamber any further then look at least at shortening that (should your next run of it, without the 90 degree bends in the bottom, turns out to not fix it). I know that your original idea was to increase contact time but it seems to me that you're now faced with a "too long" contact time? Dez's skimmer shoots water right into the reaction chamber, but also significant, I think, is that this means his injection tube ends up being shorter.

Just a question though, how long is the feed pipe from the pump to the injector? What diameter pipe is it (it looks like 1")? Is there anything that could be restricting the pump at all. How old is it? If it's second-hand .. what shape is the impeller in? You'd be surprised at how little of a restriction on the pump (or a particle of crud in the becket itself) can adversely impact foam production.