Note: groups.io will be down for maintenance on Monday, September 26th, starting at 9AM Pacific Time (4PM Monday September 26, 2022 UTC), for approximately one hour.
Turning off the freshwater pump may not be enough.
If you (for example) are on the hard, living on the boat with the seachest thru-hull closed, the leak is small and incremental. If you turn off the water at 2PM and go into town, and then turn it back on at 7PM it was off only for 5 out of 24 hours. The other 19 hours there is a trickle leaking past the flush valve ball and pressurizing the seachest. While that's happening water is being pushed from the seachest to the generator. Even though you shut off the water pump, the accumulator tank has a volume of water under pressure, and it will push into the system, basically supplying water for the leak until the pressure drops down far enough the valve no longer leaks. So shutting it off for 5 hours, or overnight won't solve the problem. Of course the problem only exists if the seachest thru-hull is closed with the lid on the seachest.
Here's the takeaway: if you ever close the seacock, take off the lid to the seachest. If you don't you're just gambling.
Apollo valves had a warning about their three way valves that follows (note paragraph 2):
When ports “A” and “B” are the inlets, and port “C” is the outlet, the
valve becomes a mixing valve. With minor variations in position 2
the percentage of components at “A” and “B” can be varied to the
outlet “C”. This has been successfully applied to hydronic systems.
It may not be possible to isolate the ports from one another in any
position. If the valve is in position 1, and the pressure at port “B” is
significantly higher than port “A”, the ball may be forced off the seat
allowing mixing from all ports. Whether or not this is a problem
depends on the application and its sensitivity to unwanted mixing.
The situation where the pressure at port "B" is significantly higher than Port "A" is exactly what we have in the flush valve at the watermaker. What happens is that the freshwater side "unseats" the ball and water leaks to the pre-filter and thus the seawater system. Whether this is unique to Apollo's valves, or is typical of ball valves I don't know, but now I am very very suspicious.
You can test yours by shutting the seacock with the water pump on, noting the level of water in it, and observing for 12 hours. Of course you must not use any seawater (no airconditioning, no toilets, etc.). On mine it would fill about 1/3 full in 8 hours.
This represents a lot of water over the course of a month as well.
I'm planning on putting a shut-off valve on the fresh water prior to the flush valve. That way I'll sleep better and I won't worry about the new 3-way valve leaking.