Minor Boat Project Odyssey

[Rick D]

This is way too typical for boat stuff. What seems simple turns into a mad, Charlie Sheen-ish, odyssey.

So, my 40.5 had only a direct reading fuel gauge on the tank. For my year and gear, it involved:

A. Remove pins on both bimini struts.
B. Fold bimini forward to clear wing cockpit locker doors.
C. Remove staysail from locker.
D. Remove access door from locker bottom.
E. Fill fuel tank while hanging over gauge to not over fill.
F. Pay the man & put it all back together.

Anyhow, so we boaters are nothing if not adaptable. But, this got tiring after 11 years. So, I decided to put in a remote gauge this year. I did some research and selected Moeller because of a decent website and since they do a lot of OEM work and sell to relabelers.

So, I had a discount coupon from WM and bought the gauge and sending unit for a 12" to 24" tank (the 40.5 is 18.5"). However, that wasn't smart since the raw retail cost is just about double what it is on this site and then I pay about 10% sales tax, so $100+, less a $15 discount = no deal but convenient.

Well, the first thing I discover is the sending unit has to be modified very precisely for the tank dimensions. First, you shorten the float arm based upon a chart and then the support arm by a factor of 50% tank depth plus some fraction of inches. OK, so that makes sense as to dropping the pivot point half way down the tank and they obviously have a lot of experience I don't, so I dutifully cut everything to spec.

I drilled a 2 1/8" hole (hole saw $13) for the gauge and ran a hot wire from the oil pressure gauge, then hooked it all up, and it worked (sorta). The gauge showed 1/8 tank when the mechanical gauge showed 3/8. I measured the fuel with a tape, and the mechanical gauge was on the money. So, this is not good; I analyze the situation and decide to install another sending unit but not cut the float arm at all and see where we wind up.

I install the new sending unit and it reads just about a quarter tank. Close enough, I figure. However, when I go to fill up (three weeks ago at $4.16/gal) the gauge isn't reading anything. When I get to Catalina after a round trip, I check the gauge connections and it comes to life, reading about 3/4 full. I use the boat, and the gauge fails to work. I pull it out, and figure it is the gauge connection since it worked when the hot lead was wiggled and now nothing. So, I order a new gauge from this site for half of WM $$ and presto, it works.

So, new trip. Tsunami day (another story), we take a trip up the coast for the YC's Opening Days. Fill up before returning home ($5.18/gal). Gauge works fine but off 1/8 at full (short of full indicated). Once again find out no gauge reading again when half way home. So, I pull everything apart and check connections. Good hot lead. Hooked up a battery to the sending unit terminals and got good volt readings at the gauge, so those connections are good. So, I figure it has to be the gauge since the system worked fine after I replaced the gauge last time. So maybe a two-failure of gauges withing a day of operation? Seems strange, but everything checks out, so probably so.

In frustration, I went to WM and purchased a Telefex gauge I liked much better and it was cheaper by 25%. Came back, hooked it up and.... nothing :cussing:

The Telefex gauge came with a decent set of instructions that included a fail-test mode. I checked it out, and this gauge as well as the Moeller were good. Since I know the wiring is good, the only logical culprit is the sending unit.

OK, so here is my plan: replace the sending unit (ordered from here). Leave the float arm alone and DO NOT cut the sending unit mounting arm, but locate the pivot point lower since it reads 1/8 tank low at both full and low volume.

So, three sending units and three gauges, and maybe I'll have this right.

In retrospect, this stuff was all within warranty, but I didn't have the time to play send-it-back and didn't want to stick WM with my (possible) incompetence. So, this is how a $60 one-day job becomes a $240 month-long marathon. For those of you who do this in the future, there is a module you can add on top of the direct-reading sending unit to convert it to electronic. Whew, that is the way to go!

 

[Ed A]

And thats just how it is in the boating world. I surely know how you feel, ive been lmao.
Ed

 

[Rick D]

OK, Need Some Electrical Tech Advice Here

So today, I installed another sending unit. Still no reading on the gauge. I double checked the gauge; it's good. I figured it's a ground problem since there is 12.8 volts going into the gauge. I grounded the gauge to the fuel tank directly. No change. I reinstalled all terminals; no change. I installed a complete jumper wire to the gauge; no change. Finally, I checked the sending unit ohm range; it was on the mark. So, just before shutting everything down, I tried grounding the tank.

Eureka! That was it. So, I ran a ground wire from the house battery terminal to the lug on the tank and all is well with the gauge.

Here's my uneasiness; why did the gauge work on two occasions before, at least for a while? I could find no grounding wire or strap to the aluminum tank anywhere; remember that it had a direct reading mechanical gauge. What I grounded it to is probably a lifting lug. So, why did it work before (for a while). And, more importantly, have I done anything nasty, like electrolysis by grounding the tank to the negative house battery ground???? I don't think so, but don't feel secure in that. Opinions?

 

[Bill Roosa]

Yes there are some issues. A DC grounded fuel tank is different than a bonded one and also different from a tank grounded to protect it from lightning. Which is why they DO NOT PUT ELECTRICAL TANK GAGES ON FUEL TANKS in at the factory.
You should have asked first dude and we could have saved you the trouble. Fear not you can install (yep more work) a nylon reducer between the tank and sending unit and just ground the sending unit with a wire to reduce the galvanic and lightning issues (mostly). There is still a connection between the sending unit and the tank through the fuel. A high resistance one but rust never sleeps.
I would also bond the tank proper to the lightning ground for good measure.

Now if your boat was mounted on rubber wheels……..

 

[Bill Roosa]

I think it would be easier to just cut a window in the access port to the mechanical gage and push the stuff out of the way while you eyeball it with a flashlight.
To answer your specific question, why the intermittent working? You probably have a connection (inadvertent type) between the tank and the DC ground via a metal object or bear wire etc. What you are describing is an intermittent ground problem which you figured out. Now the question becomes what is coming into contact with the tank that also is making a ground connection with the DC circuits? This may be an indication of a future problem or just a "left the boat hook lying on the tank" type of issue. I'm thinking it would be more the former than the latter as the tank is mostly enclosed. Can you remember any one thing in the locker that correlates with the gage working or not working?
I know that is not much help but some place to start.

 

[robertsapp]

So if you put an electrical sending unit into a fuel tank, you're introducing the possibility for electrolysis?

 

[Bill Roosa]

Yep, the circuit is battery + to meter to sender to tank to battery ground. The obivous places are the connections at the sender and tank and the tank and ground. The not so obivous ones are sea water tracks to the bilge from someplace on the tank. If the tank to ground connection developes corrosion (becomes resistive) then the prefered way back to ground for the sender current is "some other lower resistance route." Since the battery ground is tied to the engine and it has that prop thing sticking into the water this could be a salt water track or even moist dirt along the hull. The point where the current leaves the tank will develop corrosion.
With all that said you are probably not going to get into any trouble if you keep the tank ground clean and tight however you do apparently have something grounding the tank in an unknown fashion. that is what I am concerned about.

 

[Rick D]

Eureka Moment and Follow Up Question for Bill

[Since the battery ground is tied to the engine and it has that prop thing sticking into the water this could be a salt water track or even moist dirt along the hull. The point where the current leaves the tank will develop corrosion.
With all that said you are probably not going to get into any trouble if you keep the tank ground clean and tight however you do apparently have something grounding the tank in an unknown fashion. that is what I am concerned about.[/quote]

Bill, the last post wasn't me. You said a better way to do this would be to isolate the sender completely by way of a plastic insulator. So gasket & screw the nylon to the tank and then the sending unit to the nylon. Did I get that right?

I too wonder how I was getting a partial ground connection. To answer directly, I know of nothing touching the tank, which as you note, is pretty much enclosed, and I could see no bonding wire. Whoops! Not so, the Single Side Band copper was bonded to the tank! And, I found a bad connection at the keel, so I replaced it with a KISS counterpoise. Eureka ll! That was the ground path, and is why it was intermittent. So, am I any better bonding the tank to a through-hull for ground instead of isolating the sender and providing a battery ground?

 

[Bill Roosa]

It depends Don

West marine has a great document on grounding. DC, AC, corrosion, radios, and lightning all have to be addressed in my humble opinion. The skinning of the cat is different for every boat but as a general rule the ABYC does not recommend using and fuel tank as a return ground for the DC system (all circuits shall be 2 wire circuits) as it ties it to the entire boat (bow nav light to mast head to stern nav light and everything in between) and makes lightning potentially mix with fuel. Clearly a bad idea.
You got it right with the sender unit. The nylon screws into the tank and the sender into the nylon. Ground wire comes from the sender (may have to make a terminal here) and not the tank. That leaves only the float in contact with both the DC system and the fuel and it is probaly not metal (or not). You can use an ohm meter to check for a circuit.
After reading the WM article I think you will see that your SSB ground is not done properly (allows DC to pass) and have a better grasp of what you are up against.
I wish I could tell you this or that is what you need to do but all the grounds interact so I'd have to be aware of everything and how it is laid out and what all the particulars are before I could do that.

Here is the WM link:
http://www.westmarine.com/webapp/wc...catalogId=10001&page=Marine-Grounding-Systems

 

[Bill Roosa]

One more circuit that is commonly overlooked, all fuel tanks are tied to the engine through the fuel line. Even if you have rubber lines the fuel itself will conduct. Not that there is much we can do about that!