Yeniraki, your skimmers do not look much like Volcanoes w/ respect to dimensions/pumps. FWIW, Nanook's isnt much the same as well besides the diamter of the neck and body. Nanook, your neck is quite a bit taller, which seems like it helps your situation as well, not to mention, you most likely are running a much better impeller than the 'Volcano' Dart impeller, so you can run 3900lph of air or more and not worry about the mixing pump 'burping' up larger bubbles. One of the biggest problems with the Volcano needlewheel is that it cant handle/generate as much water flow on its own (needs more needles, mesh, etc), so it cant draw as much air, nor handle as much if force fed. Thats where most any solution for the Volcano starts, otherwise it seems that even 3900lph (alita 40) can be too much for the pump to blend enough (passes some large bubbles).
I do have a correction to make though too:
"Yeniraki noted that by turning down the gate valve you increase air aspiration via the venturi effect."
Im not sure where Yeniraki said this... I didnt see it (but this thread is getting a little nuts so I most likely missed it or it was phrased different)... but that statement is actually wrong.
FWIW, it could be seen as a major flaw in the design for some of the earlier 'venturis' that I have seen... varying the valve has nothing to do with the venturi, simply varying the restriction of water flow going into the pump. The ONLY way to adjust a venturi is through varying the venturi's diameter (its point of narrowest restriction). A venturi works not as a valve (otherwise, we would all just have a T'ed inlet for our needlewheels with one valve for air, and one for water... assuming that less of one would mean more of another or something like that)... it works by increasing the velocity of the water (by narrowing the pipe diameter but trying to keep the volume of flow as much as possible)... when water increases in velocity, its pressure (at a perpendicular) drops... sucking in air. So unless a 'valve then venturi' assembly can actually increase the water velocity by restricting/opening the valve before the venturi (which it doesnt)... you arent doing anything. Think about it... if you restrict the water at the valve... all you are doing is lowering the water flow rate going through the skimmer, so if anything, you are dropping the water velocity, and actually hurting the venturi.
So in short, while having/adjusting the valve in front of your venturi might SEEM like its changing something in a good way, all it can do in reality is harm it and performance wise it does nothing. The best idea is to leave the valve 100% open.
Now if you are force feeding... its a different story... then you can adjust the relative throughputs of water and air with valves in the air and water inlets. A venturi does very little then. It might help blend the air into the water before the needlewheel a little better so the needlewheel has more 'bite', but a good needlewheel will make short order of large bubbles anyways.
The best solutions are still to:
1. just add a linear air pump: the 'problem skimmer' here seems to be the 18" diameter, 4' tall ones that aren't force fed. That body diameter and 6-8" neck should easily handle 4000lph at the neck, and up to 2x that in the body... but going over 4000lph means you are going to have to drop the waterline below the reducer, and this can lead to other problems (or, it can be just fine).
2. modify the pump: add mesh, more pins, or a whole new impeller all together to the dart motor. Now it will draw more air on its own from more 'bite' into the water (yet its not likely to be enough), and be able to handle more air if force fed.
3. add airstones: Leave the pump alone sucking its 1200-1400lph of air, and add an array of fine ceramic airstones being fed by a linear air pump.
4. change the pump... a hammerhead needlewheel would work on its own... drawing over 4000lph (most likely 7000lph or more) without needing an air pump. Thats a high wattage motor though.
The one remaining problem is the reducer... an 18" body can handle much more than 4000lph, but an 8" neck cant... well... not unless you have a reducer that is very steep (which allows bubbles to rise above the 'waterline' without getting trapped under it). I can see that Nanook's skimmer uses this (along with a taller neck in proportion which does help). This way... you can get away with adding alot more air than a given neck diameter can normally handle... because the 'waterline' is down in the reducer somewhere, but the 'resistance' of the steeper 'funnel' doesnt disrupt the foam as it rises... only gently condensing it. Something that might work (besides a cone body swap out or a larger neck being added) is to make the neck taller... by 6" about. This isn't ideal, but it would be a way to compensate because it would move the waterline up higher in the skimmer.