Fuel Contamination with Water

[fno]

Recently, the 225-HP [Yamaha-made Mercury-branded] four-stroke-power-cycle engine failed to start on my 2005 210 Outrage after the boat had not been used for more than a year.

I removed the cowling and observed the red ring that indicates water in the fuel filter was floating at the top. I removed the filter bowl and dumped the contents into a jar. The liquid smelled more like water than gasoline so I pumped some more with the bulb and filled the jar. It was not encouraging.

My main concern was for the engine. I don't want and don’t need and can’t afford a new engine.

I bought a new hose and primer bulb assembly, and I connected new, clean fuel from a 5-gallon can to the engine, but the engine did not start.

I opened the drain on the fuel-vapor separating filter. The fuel that flowed out was white and creamy. I continued to force more clean fuel into the engine and finally got it to run smoothly on a hose adaptor in the driveway. See Figure 1 below.

Fig. 1. Fuel taken from the engine showing phase separation between gasoline (upper layer) and an emulsion of water and gasoline.

gas from engine (1).JPG (31.94 KiB) Viewed 2641 times

The next day I bought a plastic barrel from a feed store and began the process of draining the fuel tank. I extended the fuel line from the tank though the engine splash well drain and filled a Mason jar. See Figure 2 below.

Fig. 2. Fuel taken from the fuel tank. Again, phase separation has occurred, but now with three layers: gasoline, gasoline-water emulsion with sludge, and water.

gas from tank.JPG (31.07 KiB) Viewed 2641 times

I then decided to completely drain all fuel--about 30 to 40-gallons--from the fuel tank. Gravity-draining was not possible. I used a small AG-sprayer 12-Volt pump that specifically said to NOT use for flammable liquids. It was a plastic pump so the dangers were sparks and the plastic disintegrating while pumping. Neither happened, and I drained the tank efficiently. I suspect that waterlogged gasoline vapors would not pose the same threat as fresh gasoline, and it was a breezy hot day.

While I was watching the tank empty I decided to remove the RACOR fuel-water separator filter and check it out. I did not take any photos of the filter but it was rusty as can be. I did drain the fluid inside the filter and I use the term fluid lightly. See Figure 3 below.

Fig. 3. Fuel taken from the fuel filter canister. This liquid is contaminants that were blocked by the filter media from passing through to the engine.

gas from filter.JPG (24.42 KiB) Viewed 2641 times

I cleaned the filter assembly, replaced the canister and put it back in the boat.

I suspected that the water in gasoline might have come from several places.

The fuel (as delivered by the fuel vendor) may have contained water; many on-water fuel vendors can pump fuel with some water. I do not think this is my source, as the boat fuel tank developed water without being re-fueled or being used much.

The other possible sources might be the fuel fill fitting on the gunwale, the hoses for fill and feed, and the seal of the tank level sensor.

A worst case cause of water in the fuel wouldbe is a crack in the plastic fuel tank.

I am ordering a new fill fitting, new hoses, and will replace them all. The boat is 20-years-old and due for some new fuel lines.

CAUTION
Boston Whaler recommends marine-grade fuel hose and double stainless steel clamps everywhere. The hose is either rated type A1 (rated for continuous contact with fuel) or A2 (fuel vent or vapor hose). Ordinary automotive hose is not a good solution and should only be used as a temporary solution. I will continue the narrative describing the process of replacing the hoses in my boat. It should be entertaining for all but myself.

[Moderator's note: I added figure numbers and captions to the illustrations. These can be emended by the author if necessary.]

[jimh]

You are going to faint when you see the retail price of new marine fuel hose certified A1-15 Alcohol Resistant. Perhaps you can locate a more affordable source than your local West Marine store, where it is $10-per-foot.

[fno]

[The total cost for the new hoses and clamps I will need comes to] just under $400 for

• 8-feet of 1-1/2-hose
• 8-feet of 5/8-nch hose.
• new fill cap ($66)
• good-quality 316-stainless steel hose clamps ($5 to $7 each)

And, yes, I was sitting down with a much needed beverage in my hand.

[fno]

An update to this project.

I decided against a try-this-try-that approach to find the water intrusion. Having water in your fuel is both a cause for damage to a fuel injected motor especially one that is 20-years-old. It is also extremely expensive to draw down the contaminated fuel to get as much of the water as possible. I now have about 35 gallons of compromised fuel in a plastic barrel that needs disposal. I am haggling with the local fire department to use it as a training exercise.

My course of action was to replace or at least seal any intrusion points. This meant a new:

• fuel fill port;
• hoses;
• double clamps;
• gasoline-friendly sealant for the level sender; and,
• new deck plates.

The fuel fill and vent hose at first seemed a daunting task.

The fill is located on the gunwate, seen below in Figure 4.

Fig. 4. A close up view of the fill fittings and hoses. The space is tight.

gunnel fittings close up.JPG (68.63 KiB) Viewed 2426 times

.

The tank inlets are located in the console below a 5-1/2-inch deck plate, shown in Figure 5.

Fig. 5. A view of the tank inlets and hoses inside the console.

console tank fittings close up.JPG (85.78 KiB) Viewed 2426 times

The first lesson was to remove both deck plates, as 6-1/2-inches of opening made the process 100-times easier. Plus, I got to re-seal the deck plates and replace the cheapo rivets with 316 stainless screws and locknuts.

My first attempt was to try to send the two hoses from the gunwale port to the console. I hit a brick wall even with an electrician's wire-pulling snake on the end. Pulling the fill and vent hoses "up" from the tank port went much easier, and within 15 minutes I had my hoses in place and a beer in hand.

The next day I planned to install the 3/8-inch hoses from the fuel tank to the fuel filter. The tank outlet is located sort of under a deck plate in the stern of the boat.

Fig. 6. View of the outlet fitting and hose.

fuel-engineFittings.JPG (67.49 KiB) Viewed 2417 times

Workspace is much better there so I thought this would be easy. Not close. The fitting is located to the side of the deck plate, it was corroded and the hose eventually was cut off the fitting. Routing the hose was fairly easy if I put my whole arm through the deck plate hole. It should be noted that removing the deck plates is optional but highly recommended as any extra inch of hole space is proportional to the reduction in damage to the skin on your arms.

I also purchased and used a Milwaukee flexible inspection camera. Liberal amounts of RTV and deck sealant were used on this project also. Since completion the tank was filled, engine run and tested on the water for about 75 miles of operation.

[MarkCz]

Very interesting write up. Do you live in an area where fuel contains ethanol? The amount of water shown on the lower part of your jars is huge. Nice job on the diagnosis of water in the fuel.

[fno]

Mark, me and the boat reside in inland Florida and most all of my gasoline is non-ethanol that comes from gasoline retail stations landlocked.

The same situation presented itself several years ago. At the time, I was not sure if the water in the gasoline was pumped up from the multiple stations storage tanks or found its way into the boat via rain or washing. At the time, I did remove the fuel filter, and added a filter-water separator.

Fast forward to the last month, the water I found was definitely from intrusion as the boat had not been used for about two years and was sitting in my driveway. Rather than changing out piece by piece of the puzzle and waste a bunch more fuel, I decided to replace all of the components that could potentially allow water into the fuel tank.

One step I did not take but considered was to pressure test the tank itself. If there is a next event, that will be on the list of things to do.

[MarkCz]

Your humidity levels in Florida can get pretty high so I wonder if that may be part of the problem. I remember watching a YouTube video on a channel called "Born again boating" where the person was showing how much water can be absorbed into gasoline from the air. It was a surprisingly large amount. I don't recall if he was showing the results with non ethanol fuel though.

[Mambo Minnow]

FNO - nice reference article that I will save for future reference. Currently working in Italy and the boat is is long term lay-up. I hope that I don't have to go through all of this on my return. As a precaution, I do intend to completely drain the tank with whatever minimal fuel is still there.

If you could share the fuel pump kit information you used, I would be interested as I have been scared to use anything electrical around fuel.

Hope all is well in Fort White. I will be retiring at the end of the calendar year and returning home.

[jimh]

I know a pressure test of the fuel tank is often mentioned, but actually getting a valid test seems like it will involve a bit of preparation. The tank has several inlets and outlets that could be sources of leaking off pressure in the tank:

• fuel inlet hose (very large diameter)
• fuel vent hose
• fuel pick-up hose or hoses
• fuel tank level indicator fitting

The fuel filler hose is connected to the fuel fill fitting, and typically those fittings use an O-ring seal. On the other hand, a pressure test could be useful if it detects that the fuel-fill fitting is not an air-tight seal. Perhaps adding some additional seal to the fuel inlet cap, such as electrical tape, might be useful.

The vent hose is connected to an open vent, but you could perhaps seal that vent using some more electrical tape.

The fuel pick-up hoses run through filters then to the engine. If the engine has a fuel line disconnect fitting, you could just disconnect the fuel line from the engine during the test. Usually higher-power engines do not have a disconnect fitting. You might have to temporarily disconnect the fuel hose from the engine and seal the end with a tapered plug.

If there is no check-valve in the fuel pick-up line, you could use that hose as the means of introducing a positive pressure into the tank. Or you could use the vent hose as the route to introduce pressure into the tank.

Usually the fuel tank will have a label that will explicitly designate a maximum testing pressure, and that pressure will be quite low, maybe just 2-PSI.

Any better ideas for how to actually pressure test the fuel tank would be very welcome.

[fno]

My thought process was to make hose plugs from aluminum stock and insert them into the hoses. A 1-1/2" for the fill, a 5/8" for the vent and use the outlet hose to introduce pressure. One caveat, I have a lathe and can make any plugs I want or need. For others, it might not be so easy. I do not think there is a check valve in this system because it would be in the tank and that is not a good place to put a check valve. As for pressure, I agree it should be quite low with a quality gauge that can read well in that range. The idea is to put some pressure in the tank and see how long it can maintain that pressure. As I have a polyethylene tank I am not worried about corrosion like I would with an aluminum tank but a crack or split seam would be my undoing. In my case, I would be removing the T-top, the console, and the center deck insert to access my fuel tank.

[jimh]

Regarding a check valve in the fuel outlet from the fuel tank, I think the usual place for this is right at the tank fitting. The purpose is to prevent back-flow of fuel into the tank once it has been sucked out of the tank and up the pick-up hose. On my c.1990 boat I don't see a check valve on either of the two fuel pick-up ports and attached hoses.

If there were a check valve, the engine lift pump would have to exert enough suction to get the check valve to open against its spring, and that might result in a reduced fuel flow rate at high engine speeds when fuel consumption was the greatest.