Electrolysis injected from Windlass.

[Screen Saver]

I have been chasing for some years a relatively minor electrolysis problem. It has been slowly (over 15 years) eating my prop and its turning from gold to pink and pitting quite badly at the tips.. The other day we were raising the anchor at night, something we don’t do often. But I observed small sparks coming off the chain as it went across the front chainplate assembly base that has the bow roller in, clearly suggest it has a voltage on it. With the windlass not doing anything but enabled if I measure between the windlass/gypsy and say the pull pit (which is ground negative)I get about 0.013volts. My understanding is this is not enough to be a problem. However, when the winch is working there is in excess of 5Vdc when the anchor is going up or down. Im guessing maybe more just cant hold the button down for a long time while at anchor.

So my question is could someone measure the same points say between the Windlass metal part and ground or the pull pit if its grounded or a suitable ground and tell me what you see at idle and when you activate the winch. I’m doubtful that the duration of its operation is enough to cause my problem, but equally I don’t believe this is right.

 

[Davidasailor26]

It doesn’t surprise me that there’s some amount of difference between the windlass ground and the pulpit when the motor is running. We talk about “voltage drop” from wires carrying large amperage, but that “drop” is really in 2 forms - the positive cable sagging lower and closer to ground, and the negative cable getting pushed up above ground. The pulpit isn’t carrying the load so it should still be at the real ground potential. I presume the 5V you saw was that the motor’s ground was higher potential than the pulpit, not the other way around? If the pulpit is higher voltage than the motor ground that would indeed be strange. Beyond that my only question would be whether 5V is a reasonable amount for the motor’s ground to be elevated. That does sound pretty high to me. If the ground were up 5V it would mean the motor is only getting 7V minus whatever drop is on the positive side.

 

[American_Mainsail]

I'm not 100% following what the issue related to a DC short to the windlass surface itself is. Seeing minimal voltage potential between the windlass and the ground seems reasonable, then seeing 5V between windlass and ground when on suggests a lot of drop in the windlass circuit, but that doesn't mean that something is grounding internally to the windlass surface.

I definitely feel like this is an interesting thought. If there's a short to the windlass surface, this gives a path for current to short from the boat, through the chain, and into the seawater.

If you take the DC- off the stud and touch it to the case for an instant, will the windlass run when you press the up or down switch?

Have you gone through a full electrical audit with an ammeter checking the draw when off of all electrical loads, starting with your bilge pump?

Do you have one of these? They are the amazing for diagnosing electrical issues. They have super long leads so you can connect to ground and power remotely and work all over the boat. With one of these, you could leave the circuit in tact, and (just for an instant) inject power onto the surface of the windlass to see if it spins. That will tell you if there is an internal short somewhere on the windlass.
Circuit Tester Kit

You could also do a reference electrode test to see if current is going in the water when you raise/lower the chain.

My boat is out of the water for maintenance this year and I won't be there for at least a couple of weeks. I can check it out for you if no one else can get there first.

 

[dlochner]

Why is the pulpit grounded? I would not expect it to be part of any electrical circuit, unless the wire for the running lights is making contact with the pulpit.

Small currents over long periods of time can be just as damaging as high currents over short periods. If you have access to a silver chloride reference anode you can measure the potential between the metal and anode. It should be around -900 mv. A positive value is definitely problematic.

To be a bit pedantic, what you appear to have is stray current corrosion, not electrolysis.

 

[Screen Saver]

Just to clear up some points and update further investigation:

1. There is no actual ground on the windlass as such there are 3 wires 2 red for up and down and one yellow negative.
2. Yesterday I could not run the winch for long while measuring. Today I could and there is a full 12.8v leakage when the winch is active where yesterday I said is about 5v.(ie between winch/chain and pulpit.
3. Today I measured from negative (yellow) on the windlass to the windlass case and got a full +12.8v when the winch is going either up or down. This there for means that a full +12.8v will go from case, to the gypsy to the chain. This explains why I saw sparks as the chain went across the metal pad part of the bow roller assembly which is part of the pulpit and is wired to the boats earth(negative)
4. Been a while since I have played inside DC motors but as I understand it of the four coils(EM’s) if one of these was some how connected to the case because the insulation broke down for example, then I would expect it to be only when the winch was going in one direction when that pair was enabled. Im assuming only 2 of the four are activated at any one time. The same analogy would hold true for some breakdown in insulation on the commutator going through the shaft and bearings. I would expect it only when that segment is active. So if that’s correct it means for me two breakdowns in insulation seems unlikely.
5. The windlass circuit is not dropping that amount of power. If it were anywhere near that I doubt it would still work
6. If not been game to take the DC (red) off and touch it to the case as mentioned, given when I put my DVM on continuity beep test etc, it says the case of the windlass is connected to the pulpit.
7. Have been through full electric audit as was asked. Some time ago yes I spent about 4 days going through checking grounds current, starting gens sets, disconnecting Solar and HF etc. Its abig job on this boat. But I haven’t done it again for a while. That said this problem to me is very much local to the windlass somehow.
8. Whys the Pullpit grounded. Its part of the earth leakage and lightning protection. Everything, ie pull pit, pushpit, rigging, mast, solar mounting system, davits, everything metal outside is grounded.

I don’t think I need anyone to do the test I asked for now as Im convinced what Im seeing can not be right. It may not be related to my prop slowly being eaten, but it still can't be right. Im now looking for a single point of failure or problem

 

[Davidasailor26]

1. Yesterday I could not run the winch for long while measuring. Today I could and there is a full 12.8v leakage when the winch is active where yesterday I said is about 5v.(ie between winch/chain and pulpit.
2. Today I measured from negative (yellow) on the windlass to the windlass case and got a full +12.8v when the winch is going either up or down. This there for means that a full +12.8v will go from case, to the gypsy to the chain. This explains why I saw sparks as the chain went across the metal pad part of the bow roller assembly which is part of the pulpit and is wired to the boats earth(negative)

Indeed it does sound like you have problems beyond just voltage drop. A full 12V + on the windlass case does explain the sparking. In theory if the chain is at +12V it would be the part that corrodes, rather than the prop on the negative side of the circuit, but who knows what other strangeness might be going on there.

 

[dlochner]

Whys the Pullpit grounded. Its part of the earth leakage and lightning protection. Everything, ie pull pit, pushpit, rigging, mast, solar mounting system, davits, everything metal outside is grounded.

I think this is a big part of the problem in addition to the windlass problem. The lightning protection system should be independent of the boat's DC negative system. This article by Stan Honey delves into the issue of bonding and grounding, it may be helpful in tracking down your problem.

Try disconnecting the pulpit from the bonding system to see if there is voltage between the pulpit and winch.

In your spare time, shopping for a new prop is in order. The pinking and pitting indicates the zinc in the bronze is leaking out making the metal more brittle.

 

[Davidasailor26]

I think this is a big part of the problem in addition to the windlass problem. The lightning protection system should be independent of the boat's DC negative system. This article by Stan Honey delves into the issue of bonding and grounding, it may be helpful in tracking down your problem.

Try disconnecting the pulpit from the bonding system to see if there is voltage between the pulpit and winch.

That’s an interesting article by Honey. Calder on the other hand talks pretty extensively about tying the lightning/grounding bus and DC ground bus together. His full discussion is too long to fully repeat here, but it’s in the “Grounded versus Grounding” and “The Common Ground Point” sections of chapter 5 in his Boatowners Mechanical and Electrical Manual. One excerpt from that section -

A grounding point or bus should be established as the boat’s common ground point. All the non-current-carrying grounding circuits are fastened to this bus, including:
• A connection to the main DC negative bus (this will not be current carrying since it is not part of the circuit to the batteries).
• The grounding wire (green or bare) from the AC distribution panel (never the neutral wire) if the AC grounding system is to be connected to the boat’s grounding system.
• The lightning/bonding circuits.
• A connection to radio ground plates.
• The external ground plate or strip (for lightning dissipation).
• Any sacrificial anodes for cathodic protection.

The function of this common ground point or bus bar is to hold the DC negative circuit and the various grounding circuits to a common (earth) potential (Figure 5-35).

 

[American_Mainsail]

There's two approaches to bonding. I know that Nigel tends to lean toward the "bond everything together" camp. Hunter (at least my Hunter - built in '03) took the opposite approach, which is also considered acceptable, and is probably what Honey is advocating for. My thru-hulls are not bonded together to a common DC ground. I'm honestly not sure if the lightning protection system is bonded to a common ground, I haven't chased my DC grounds through enough to figure that out.

I'm leaning toward thinking that there might be some benefit to leaving the lightning grounding system isolated. Those wires are going to carry a tremendous amount of voltage, anything electrical that they're tied to in the event of a strike is probably going to fry. Nigel is a pretty sharp guy though, I'm sure he's coming from an informed opinion. He might be thinking that the more surface area of wire you can spread that voltage spike out to, the better.

I believe Nigel tends to lean toward the "bond everything together" side because it can be slightly safer when an AC load is shorting to ground, tending to spread out the voltage leak across all underwater metals, instead of one, very hot metal.

I'm not sure that bringing the grounding system debate up really helps this conversation though. It seems like OP has an issue, that it looks like he's got a handle on, with the windlass shorting directly to the case.

P.S. @dlochner - I bit on the Honey article. He did a great job on it. Thanks for posting.

 

[Screen Saver]

Agree with the above, bonding grounds is not really got anything to do with my problem. As yet I can't get the "beep.... beep" Windlass out, the shaft is stuck and gentle persuasion and WD40 is not working. More significant persuasion bigger hammer will likely expand the shaft end making it worse, so im looking to make some sort of puller arrangement and fit to the top of the shaft. I also have a new bronze bearing for the top, so I need to get this apart somehow. That all said I can only think that the coils inside the motor must somehow, somewhere be touching the case, or insulation on the armature is failing. I have looked for a Lewmar motor only, but in true form Lewmars's smallest part is motor and gear box, or some third-party made in China thing that looks like plastic/nylon drive shaft, rather not go there.