I remember the NPR radio show “Car Talk”, and how each week Tom and Ray would have a segment called “the puzzler” in which they tried to diagnose some weird car problem. Well, I have a boat version of the puzzler. As I mentioned in a previous post, when preparing to depart from Tenacatita the engine suddenly shut down after idling along for a few minutes. This was not good. At all. Diesel engines are simple and reliable… and almost all issues are related to fuel delivery. We have been battling fuel delivery demons on and off since departing in October, and believe me, you do not want to have a fuel delivery issue when traveling offshore.
My immediate suspect was the Racor 900 duplex filter assembly. This fancy setup has two fuel filters connected by a selector valve. You can run the fuel through one filter or the other, or both. The value of this setup is redundancy. If, for some reason, your fuel filter gets clogged (say by some bad fuel), you can simply select the other filter, keep the engine running, and then change the clogged filter. Sounds good right? Yes, if it actually works.
You may recall that we actually bought a brand new filter assembly back in Sidney before we started on the journey down the coast. This was part of solving the air bubbles in fuel line problem that was causing very disturbing RPM variation (post). We determined that the old filter assembly was leaking air, and as a matter of expediency, we simply had Philbrooks install a new one. We were dismayed to observe that the new filter assembly also leaked, so the guys tightened up the bolts on the fuel selector valve, and all (seemed) good. In retrospect I believe that was a mistake. Anyway, we took off, ran 2700 NM down the Pacific Coast of North America, and had no problems…. Until I changed the fuel filter. I did what I always do in this situation – I turned the selector to the unused filter and replaced the used filter. The next time we started the boat, the engine died. It was clear that there was something wrong with the selector valve, at least in the position of the forward, or looking at the assembly, the left filter. I also noted that the selector lever was extremely tight, and it was very difficult to feel the “detent” indicating the selection of that filter. Long story short(er), we had the selector valve rebuilt, did a sea trial, tested all positions, seemingly successfully, and thought we were good to go.
We left from La Cruz down to Bahia Chamela, and later to Bahia Tenacatita, a total of about 130 NM underway. All good. Until the morning in Tenacatita. When the engine shut down, I checked the filter assembly. The level of fuel in the active filter was quite low. I refilled the filter with fresh fuel and then went through the process of priming the fuel system and bleeding the injectors. It seemed obvious when working the manual priming pump that there was air getting into the fuel line regardless which filter was selected. My experience has been that when you are working the manual pump, it becomes stiffer as the air is replaced with fuel when bleeding at the secondary, or engine, filter. This was not happening. I could not get the engine primed with fuel. Also, I noticed that the selector valve was very tight, like before, even after it was rebuilt.
Because I was very suspicious of the selector valve, I decided to disassemble the manifold and plumb together a single filter module. I got it done and went through the priming routine again, and this felt a bit better – the pump was offering some resistance. But, bleeding the injectors was not successful. I managed to get engine started, but it shut down again, and again, it seemed clear that there was air coming in somehow. And again, the fuel level in the filter module was low, even though I refilled it completely when I reassembled it. Listening carefully, I could actually hear the sound of some air leaks around the body of the filter module. The supply line from the tank via the manifold had some old black electrical tape at the joint between the hose and fitting. I wondered if it had been suspected as a leak previously, so I replaced the electrical tape with a good wrap of rescue tape. That wasn’t it. There was a black plastic nut beside the fuel input port, and I was able to tighten that a little bit. Also, it seemed that there may have been a leak between the upper and lower parts of the filter assembly, so I tried tightening the four retaining bolts and was able to get a bit of a turn on three of them. Repeating the priming process again, still no start.
In desperation I made another call to my man Lance at Diesel Premier (he had been taking my calls and offering advice all day – even though it was Super Bowl Sunday). His suspicion was the supply from the tank. We had been drawing and returning to the starboard tank, but have regularly alternated between port and starboard. It didn’t make much sense to me… but it was an easy thing to try. I though there must be something else, so I put a wrench on all of the fittings on the Racor filter and on the engine side. I was able to get a bit of a turn on each of them, including those to and from the fuel pump assembly on the engine. After one last round of priming and injector bleeding, I was finally able to get the engine started and running. We ran it up to 1700 RPM for a few minutes and left it to idle for at least 30 minutes. No problems. I put an old filter top vacuum gauge on and it was recording good, low, but non-zero vacuum. After running, I checked the fuel level in the housing, and disturbingly, it was low. I estimated that I needed to add about 24 oz of diesel to bring it back up to the top. I did see a little bit of fuel between the bowl and housing when I checked the level, but it is hard to tell whether that is a real leak or the result of small drips when topping off with diesel. I suspect a leak in the filter housing itself.
I refilled it and we decided to take the chance the next day on the 12 NM run to Barra de Navidad where we could be at a marina to make repairs. We made it with the wing engine idling the whole time, just in case. We were nervous the whole time, not confident that my single filter jury-rig was reliable. There weren’t any detectable RPM variations the whole time.
So, here we are in Barra trying to figure this thing out. I have some clear hose and fittings on the way so that I should be able to see any obvious air bubbles or leaks in the system. There have been many suggestions for potential causes including a bad fitting, a bad piece of hose, tiny holes in the tube that draws fuel out of the tank, etc. In the meantime, I replaced the Racor filter housing that I suspected of leaking with the other one, that seems not to leak. We actually ran the boat for a couple of hours on this setup without any problems. While I will systematically address all the possible sources of leaks, I remain very suspicious of the filter and assembly. I have spare parts on order to completely rebuild that.
So that’s our puzzler for this week.