I think there are a lot of reasons to use a y valve. Forespar also makes on in the North Americam markets.
It visually allows you to see which tank is on line
I didn't "get" this until now. I assumed when I first looked that it was a waste Y-valve. But I see you can use then for a variety of purposes. Thanks.
I think the issue is manifold (pun intended). First, the level gauge apparently only measures one tank. Then, I think when one tank runs dry, presumably the bow tank, the pump sucks air.
The only real solution without adding a second tank gauge is to live with a water level gauge with a weird, non-straight-line curve for remaining water versus gauge indication.
The gauge is on the port tank. I
think the bow tank is mostly above the level of the port tank. This means that if both tank valves are open, the bow tank will keep the port tank full actually, over-full, to the point where the vent line will fill up to the water level in the bow tank. The gauge will read full until the bow tank is empty, and the level starts to come down in the port tank.
To interested readers, thank you. This is a puzzle to me, perhaps you can solve it.
I'm attaching a diagram, partial, of the water system as I know it. Note that there's a mistake in it, as installed, in that the pipe from the port tank is not connected to the manifold port marked "PORT." I think this may have occurred when the yard winterized and then recommissioned the system, and carelessly connected the port line to the wrong manifold port. I assume it's marked because it's important! My suspicion is that when the bow tank empties, it allows the pump to suck air from the bow tank, because the communication between the tanks is only at the manifold, and the flow between can't keep up with the pump's suction. I think if I rectify the connection mistake, it will work since the tank with water will be closer to the pump. Does that make sense? What I've done for now is to close the bow tank ball valve.