Galvanic intrigue

DDW

Super Anarchist
6,240
972
I brought a small dehumidifier to the boat today, thinking I would plug the shore cord in for the winter. Slowly converting to the mindset of the daysailer after years as the commuter cruiser. A little concerned about galvanic action, grounds, etc. (as the boat has never been left plugged in for the first 11 years), I thought I should check. This Volvo saildrive is intentionally insulated from the engine, there should be no continuity between engine and transmission. To my surprise when I measured, I got about 15 ohms between them, reversing polarity I got negative 8 ohms. That suggested a potential and sure enough, the engine is about 3.7 mV above the drive. I removed the shore cord and tested again same result, so I began testing everything. The boat green ground is 0.48V below the shore ground. The keel is 0.48 V above the DC ground bus. DC ground and green are connected in one place. 

The keel is not intentionally grounded, but there is a lightening protection cable between the sailtrack and keel, and it occurred to me that there must be an errant ground path there: VHF antenna mounted via the shield to SS bracket, mounted to carbon mast, mounted to Al sailtrack. I disconnected the lightening cable but it made no difference. 

I am mystified, probably due to my weak understanding of galvanic potentials. The keel is floating with no wires to it. The engine is connected to the grounding system, but the transmission and drive is not. The drive has two zincs, one on the drive leg and one on the prop. There are no other metal parts in the water. The carbon rudder is conductive, but it is isolated from the boat by the teflon lined rod end in the drag link (I verified this). The keel should be floating, the drive should be floating, and the boat DC should be floating compared to shore ground. Except of course they are all sitting in salt water. Even the engine I suppose, if we consider the raw water system. I did not check the transmission shift cable, though it has insulators in it. 

What's going on here? I have in the past tested the continuity between transmission and engine and it was open. But I think that may have been hauled out, don't remember if I have done this before in (salt) water. Why would there be a consistent 0.48V between dock ground and floating boat?

For the sake of this discussion, let's assume I know the boat and its wiring well, there is no possibility of the PO having hidden a wire somewhere. 

 
Start at the beginning, anything under 20 ohms is pretty serious continuity.  What's going on with the engine drive isolation.  That was your initial concern, I don't think you would see a random thru water connection at 15ohms.  Odds are it's something between drive and transmission.

 

DDW

Super Anarchist
6,240
972
The 15 ohms is likely an artifact. There is a potential of 3.7 mV between them. That's why I get NEGATIVE 8 ohms when the polarity of the DVM is reversed. Negative resistance usually means a voltage source. A little hard to get stable measurements that low, have to scrape through the oxides and such. 

Coming from a nearby boat - that is a suspicion. I'm not quite sure what normal would be. Why would the isolated boat green be at 1/2 volt less than the dock. I tried measuring a current in the green wire between dock and boat and got nothing. Same thing between engine and tranny. But I'm not sure of the sensitivity of the meter. Fluke says resolution of 0.001A but that isn't quite the same thing as sensitivity. Why would the isolated keel be 0.7V above the isolated DC ground? How much continuity can I get from the raw water system for the (grounded) engine or the (grounded) refrigerator?

 
Last edited by a moderator:
It's a little over my head but mv potential is what we used to check our boat for zincs and if anything was hot.  Half cell in the water and other probe to all the metal bits.  There is a chart of acceptable values to see if you are over or under.  Possibly a ground connection in the battery charger would explain a discrepancy in transmission vs eng. Not the continuity though. Or some other ac device, if ac neutral and ground are connected in the boat  panel and there is leakage between neutral and hot anywhere would also account for it.  I would trust the fluke but you are getting more in the process meter level.  Might be worth getting a good marine electrical person with experience to take a look.  It's why I got our boat tested.  I wired the whole boat but really wanted a unbias set of eyes to check things.

 

IStream

Super Anarchist
10,715
2,921
Is your mast connected to your keel for lighting protection? If so, and if your VHF antenna doesn't have a plastic isolator to insulate it from the mounting bracket on your mast, that might be the source of a keel ground.

Regarding the saildrive to engine potential, 3.7mV seems inconsequential. Even if the housing of the drive is nominally isolated, there's still metallic splined shafts and gearing all the way from then engine to the prop. There are a lot of places along the way to get a high resistance galvanic connection. It would be interesting to try to measure the current flow.

 

DDW

Super Anarchist
6,240
972
As explained in the OP, I suspected the lightening cable might be contributing and disconnected it, with no change. There are antennae on the mizzen (also carbon) but the mizzen is not grounded otherwise, nor connected to the keel. 

The 3.7 mV is small, but a stable consistent measurement. The meter (Fluke 117) has 0.1 mv +/- 2 counts resolution, so well above that. Where would that potential come from? The Volvo saildrive (actually built by ZF) is electrically isolated by design, including the moving parts. I did try to measure the current flow through the meter. Assuming this is some sort of galvanic battery, shorting it with an ammeter should show current flow. It did not (0.001A +/- 3 counts). 

It is hard to know what reference to use. I have a silver-silver chloride half cell, I think I will try referencing all of the measuring points to that in seawater. Perhaps the most confusing is why there should be a potential between the things that are in the sea and the DC ground system which should be isolated, but is perhaps connected in some way by the salt water columns into the raw water system (though they have been flushed with fresh). I'd also expect the dock green wire to be a seawater potential in a perfect world. 

At the least, I intend to install a galvanic isolator before leaving it plugged in. 

The marina is the same one involved in  the Rapidly Disintagrating Saildrive thread. Don't want mine to go the same way. 

 
Last edited by a moderator:
Check the transmission cable is not providing a current path between engine and transmission.  That, and autopilot sensors on the rudder, are common sources of continuity between bits that aren’t supposed to be connected.

 
Last edited by a moderator:

El Borracho

Sam’s friend
6,255
2,330
Pacific Rim
I have no idea...but in keeping with SA policy here is my opinion: The OHM measurement with a typical meter does not measure Ohms directly but rather measures current and does some arithmetic. So it is useless in a live circuit like you have. Second, voltage differences are interesting evidence, however voltage does not cause galvanic issues. Current does. But current is much harder to measure because stuff needs to be disconnected...ugh. Also a chunk of metal that is unconnected to the other thing can be at any arbitrary voltage with respect to any other thing. Especially so when measured with a sensitive meter (Fluke).

Direct evidence might be obtained by keeping a very close watch for corrosion on the saildrive, thruhulls, anodes and such.

 

IStream

Super Anarchist
10,715
2,921
Sorry, not enough coffee this morning. I re-read your OP and note that you disconnected the lightning cable with no difference. Not to get too pedantic, but did you re-check for continuity between the mast and the keel after disconnecting or did you just check potentials again? I'm thinking you need to question assumptions about continuity.

Regarding the lack of current flow between the engine and saildrive, it's a lot easier to get a high impedance bridge across the isolation than a low impedance bridge. Contamination of non-conductive components, ionic contaminants in the lube, and other things can all give you a measurable voltage but can't pass a lot of current. If you're not getting current flow then you're not getting galvanic corrosion so between the lack of current and the low magnitude of the voltage difference, I'm less concerned about this observation.

However, the 0.5V deltas you're seeing between grounds is a big concern. I'm not intimately familiar with Anomaly's construction but here are the things I'd do/check in your situation:

1. Make sure there's no saltwater in the bilge that could be electrically connecting the keel bolts to other conductive components in or near the bilge (e.g. bilge pumps)

2. Drain the raw water system enough to create an air gap in the strainer between the raw water seacock and the engine. I wouldn't rely on a freshwater fill to break conductivity here because there's so much saltwater in the engine that can contaminate the fresh.

3. Yank your shore power cable and mast cable (as you've done), any other cables leading to the dock, and turn off your main battery switch to kill the charger, inverter, and all other electronics so there are no active electronics that could be affecting your ground potentials.

4. Measure continuity between your grounds again

5. Use your half cell or an anode on a wire to measure the potential between the seawater around the boat

With these data, you'll be a better position to take stock. 

 

DDW

Super Anarchist
6,240
972
The transmission shift cable was noted, it is designed to be isolated but I did not retest it. Since it is a Morse single lever, it has potential continuity between engine and transmission, but again the transmission end it intended to be isolated. Though the engine is clean, I do wonder about ionic contaminants outside or inside the bell housing. The autopilot install is on the steering gearbox, isolated from the rudder by the teflon lined rod ends in the drag link - I checked that. Nothing else nearby to check its potential.

I'm aware of how the DVM works, but generally even a high impedance one like this will drain capacitance over a bit of time, and shorting them through the amps function did not change anything. 

There are no other wires or continuity path to the keel. I can see all the keel bolts, the bilge is dry. The readings all persist, even with the shore cable, all battery switches, and solar controller off. The raw water system for engine and frig were fresh water flushed by running only fresh thru them for about 5 minutes - but could still be somewhat conductive. The strainers are above the waterline so easy to air gap - I intended to try that but ran out of daylight.

Next time I'm there I will try to measure everything relative to the half cell, including the dock green. Probably not till next weekend, goddamn non-retired people clogging the roads too much on weekdays. 

 

DDW

Super Anarchist
6,240
972
I'll be checking the Morse cable. Did yours have a fault, or was it never isolated to begin with?

 

El Borracho

Sam’s friend
6,255
2,330
Pacific Rim
Modern DVM's are so fabulously sensitive. You did not measure negative resistance. You had the meter on the Ohm scale and measured a current. That is why it is different in each 'polarity'. To measure the resistance between Saildrive and engine you must remove it all from boat, hang from a single point hoist, and then measure. They are connected thru the seawater, if nothing else.

The half-cell measurement seems like the only way. But keep in mind that even that measurement is suspect for things that are truly isolated. It is not due to capacitance but just whatever electrical field the object is in. Usually it is an AC voltage (that's how radios work) but some random DC voltage as compared to the half-cell is possible too. For sensitive meters and tiny voltage measurements one must eliminate the difference between the probe metal and the metal of the tested object (that's how batteries work).

 

Max Rockatansky

holy fuckfarts!
3,543
816
It was never isolated. I was doing some continuity testing, found that the saildrive was grounded, then took shit off the saildrive until I found the problem. Actually the shift was the first thing.

The throttle Morse cable housing was grounded on the engine. That caused the saildrive to be grounded thru its shift linkage cable via the shift mechanism at the helm. The fix was to just isolate the throttle linkage end on the engine.

Volvo has since changed to plastic parts to isolate, but my stuff is from 1992.

 
Last edited by a moderator:

DDW

Super Anarchist
6,240
972
I've claimed I understand how a DVM works. As I said in the OP, a negative resistance reading indicates a voltage source. Modern DVM FET front end actually measure voltage, they are unlike old galvanometer type meters in that respect. But the leakage current will still drain small capacitance values, especially on the ohms setting which is after all trying to induce a small current. I think I'd be unlikely to see 0.6V DC induced by far away AC fields. I will try to isolate the engine from the seawater and see if it makes a difference.

Max, this engine is from about 2006, the Morse cable was isolated at one time, I'm going to have to check it again to make sure it still is. That would not explain the engine and transmission creating a voltage source though - almost has to be the zinc on the saildrive, combined with the engine being in contact with seawater. 

 

Ishmael

≈≈≈≈≈≈≈≈≈≈≈≈
48,316
9,501
Fuctifino
I had many issues like this which cost me two years of time and money and then my neighbour sold his boat and all my problems went away. I suggest you either sink your slip neighbour or sell his boat when he isn't looking. Alternatively, you could unplug him from dock power and see if your numbers improve.

 
Top