Thanks for your post, Sergei!
I have indeed been exhausting myself trying to learn all about these systems but as you point out it is hard to go from zero knowledge to a point where you can make optimal decisions on what you are learning about. I figure no matter what, there is always something I might do differently had I known. But I'm working hard to minimize the chances of saying "wish I had" in big ways on this.
Perhaps I have not been explaining myself very well as I am actually doing much of what you suggest - routing the waste plumbing and switches for dump valves and macerator to that compartment. I was originally going to out the pump in there too but now I think that I will put it right beside the compartment as there is a perfect sliver of space net to it inside the closet where it encroaches.
As for gravity feed, that is what I am trying to optimize for at the moment. But we may disagree on where the gravity is needed! I have concluded that at least this jabsco pump model I bought should be able to prime itself, as per the documentation (which says 4ft dry and 5ft wet) and the testing I did today. Like a straw down into to a cup, the pump had ZERO trouble sucking the water up and spitting it out. It was VERY powerful. So now I'm trying to optimize gravity to the "straw" suction point so EFFLUENT IS THERE as long as possible.... And actually sucking down like a drain before sucking up (by creating a P trap).
What I was saying about the pump speed, I did indeed mean that it was very fast when WET. In the test I did today, the pump was outperforming the garden hose I was testing with. Well, wet after 1 to 2 feet of priming behind air. That first air was like it wasn't even there. Water flowed nearly instantaneously, no straining whatsoever.
But as you probably know, it does not like trying to suck in air and water at the same time. My conclusion was that even if I were to mount the pump sideways directly into the 3 inch main dump outlet, there would be both air and water in the line as soon as the tank level dropped below maybe 1/2 to 3/4 full. My 3 inch pipes are essentially at the same level as the tanks so the amount of liquid in them should generally match the tank level.
BUT by adding a P trap type setup, I am 75% sure this will overcome the problem. The P trap becomes the lowest point in the sewage system and the pump should operate just fine as long as the tanks keep the p trap filled. The only thing better would be to mount the pump lower than the 3 inch pipe like I think you did. But I believe the P trap will work almost as well. Even for solids.... Well, as long as they are not TOO solid. I have gotten to know sewage more than I would care too and it seems like things get a little slurrious by dump time so hopefully there won't be huge blocks.
Unfortunately (for purposes of a macerator install) our Airstreams have a pretty flat waste water system where those clumsy design SOB's out there I suspect have an easier time for this kind of install and a bigger drop on that main 3 inch line toward the final outlet (as the example diagram at the poopsheet link you provided shows).
I had looked at the sanicon system and others and concluded that they are all basically a pump and then a bunch of accessories. Some of them are critical to the functionality such as the hose. I have been looking for a great hose and had missed that the sanicon hose was collapsing somehow. I may look into trying to buy that as a part.
The other main reason I decided against the sanicon was that the plumbing they provide for the "input" side of the pump isn't really a great fit for the permanent installation I want. So that would leave the output hose as the only reason to have bought it. Had I understood the coiling and retraction I might have actually gone that way.
Finally, the cycling of the pump toward the end of dumping was speculation that might be necessary to avoid running dry. The last part of my current dumps the regular way, I get a good trickle coming out of the tanks for a couple of minutes. And if all macerator pumps are similar to this jabsco, I believe they will all be working faster than the trickle. So I am guessing that I will need to cycle the pump at the end so the trickle can catch up unless I just ignore that last little bit and move on.
I hope I am making a little more sense now but please continue to chime in and please challenge my assumptions! I would much rather have my theories disproven here before I learn something the hard way.
Texas Gulf Coast Unit, Texas Airstream Harbor
Angelina National Forest, Zavalla, TX