Our two most recent zinc anodes haven't lasted 6 months.

They used to.

What changed?

(1) Solar panels. (2) GFCI outlets. (3) A few small things.

The solar panels mean that we now generate power constantly. Constantly.

I added a Camp zinc grouper. This is 6 pounds of zinc, clipped to a grounded chainplate.

You can sort of see it hanging from its wire in this picture. It's a little grey blob down in the water.

I measured 37.5 mA of current flowing at 275 mV. This is about 0.01 W. This was in broad daylight, with batteries being charged. Current flow at night was comparable. It's not merely the solar panels, then. Also, it's DC, so it has nothing to do with the inverter or GFCI outlets.

There are some possible bonding/grounding issues. The shaft may not be electrically isolated from the engine; this is sometimes seen as a problem.

It could be stray current at the marina. That's a known (and common) problem.

I'm unsure on what I should be doing next. I'm starting by reading this:

http://www.bristol27.com/projects/bonding-grounding-system

Wow. Everything from Nigel Calder, Don Casey, and Beth Leonard in one fabulous summary.

Perhaps I need to be absolutely sure that my bonding between through-hulls and electrical system are cleanly separated. Or. Perhaps this:

• Bonding – the original idea behind bonding was to put all underwater fittings at the same potential to stop galvanic corrosion. Unfortunately, this type of bonding invites more-destructive stray-current corrosion. Bonding is still intended to put fittings at the same potential, but today the purpose is to prevent side flashes from voltage differences in the event of a lightning strike. The rules for bonding are simple: bond all sizable metal components within 1.8m (6′) of mast or rigging to the mast ground, but do not bond any submerged metal (ground plate excepted). (Sailboat Electrics Simplified, p. 164)

From this it looks like the 1981 wiring plan was to bond all the bronze through-hulls and mast to a Dynaplate. The chainplates are part of this, even though they're not submerged.

The modern idea appears to be to disconnect all the bronze through-hulls from any dynaplate bonding.

The recent survey said that all of the bronze looked great: no signs of galvanic or stray-current corrosion. Since "it ain't broke" I hate to try and fix it by rewiring the grounding needlessly. The zinc was properly sacrificed to preserve the bronze, so that's working. It's just working at a higher rate than it used to.

Update 5/25

At anchor — away from the marina — we measured the current flow just to be absolutely sure it wasn't anything to do with stray current at the marina.

I checked at just about sunset, panels producing almost no power. We had some interior lights on, the anchor light on, chart-plotter and radio. We were pulling about 1.4 A. This shows about 50 mA at 400 mV. More current. Higher voltage. Twice the power: 0.02 W.

Clearly, this is the DC electrical system ground that is flowing through the zinc fish. Perhaps the fish is a more efficient ground than the dynaplate? Perhaps there's some disconnection from the dynaplate?