Bad Day on the way to Banderas Bay

Posted by John

A funny thing happened on the way to Banderas Bay. We ran out of fuel. Or maybe we didn’t. Maybe we just had clogged fuel filters. For whatever reason, which we still don’t know absolutely for sure, the engine just up and quit sometime around 3 AM on Tuesday. We were left going nowhere, in the ocean, at night, with zero wind. It was nine hours before we got on our way again.

Both Puerto Vallarta and La Cruz are on Banderas Bay. It’s a popular starting point for the Puddle Jump and others going to French Polynesia. There is a large cruising community there, a sailing school and many boating related businesses. We had ordered some flags made by a woman in La Cruz and needed to pick them up, and we wanted to meet up with some of the Puddle Jump boats. Some of Robyn’s friends from the Baja Ha-Ha were there. We emailed the La Cruz marina asking for a reservation, but hadn’t heard back before we left Mazatlan.

Back when we were in Puerto Escondido, Robyn swapped a paperback that she was done with (some teen fantasy) for a paperback copy of The Martian that had been on the bookshelf in the laundry room. I’m not convinced it was a fair trade, but the deal was done. We already had the DVD, now we had the book, too, so I read it. In The Martian, an astronaut is mistakenly left behind on Mars and has to figure out how to survive until the next Mars mission can rescue him. It’s one thing going wrong after another. Just when he solves one problem, another one pops up. I’ve felt a lot like that for this entire trip.

We did not get fuel before leaving Mazatlan at 7:30 AM. We had last filled our main tank in Puerto Escondido before crossing the Sea of Cortez. With zero wind on that trip across to Mazatlan, we ran the engine continuously for 53 hours. We have three fuel tanks. The center tank holds somewhere around 100 gallons. The port and starboard tanks hold about 50 gallons each. At an average burn rate of about one gallon per hour (we think), we planned on switching to the starboard tank around 3 AM. There should be another 20 to 25 gallons left in the center tank as a cushion. We know from experience that the center tank gauge will read empty, but there will still be 25 to 30 gallons in the tank. We’d hoped to sail at least part of the way to Banderas Bay, but again, there were very light winds and some fog so we were burning through our fuel.

The fuel gauge for each tank is mounted directly on the tank. The center tank is the easiest to read. It only requires pulling back a small rug on the floor, opening a hatch in the floor, and reading the gauge. The starboard tank is a little harder to read. It requires opening a storage cabinet, pulling out much of what is stored in the bottom of that cabinet, opening a panel in the floor inside the cabinet, getting a flashlight and maybe a magnifier, and reading the gauge. Due to the location of the port tank gauge, we very rarely even try to read it. Let’s just say that a small mirror is most helpful, and so is the ability to bend your body in ways it normally does not. To make it more difficult, the plastic face of the gauge is clouded so only the relative position of the needle is discernible through the haze.

Pre-purchase survey photo from 2005 of Starboard tank gauge inside a then mostly empty storage cabinet.

Slight digression: I’ve been experimenting with radio email for almost a year. I’ve been trying to use Winlink, which is run by ham radio operators. I had a frustrating time finding stations to actually answer and accept my email, and finally learned that I might’ve been locked out of the system for six months. That might explain the disconnections I was getting on the rare occasions I could connect. We decided to stop messing around and opened an account with the commercial SailMail system. For several reasons, we held off trying to transmit in the marina and waited until we were on our way to La Cruz. It was early Tuesday morning, right about the time I should’ve been checking the fuel gauge on the center tank, when I sent the first email. My email connected and sent on the very first try and I was excited about that. That email produced the previous post about email at sea. Except for not knowing how to handle an “o” with two dots above it and messing up the word, it worked! But I didn’t know that at the time. It was my turn to sleep, but instead I was awake and thinking about the email—not the fuel tank—when the engine quit.

I immediately checked the center fuel tank gauge. It read empty. I went into the engine room and switched the fuel to the starboard tank. The engine and engine room were both very hot. We tried to start the engine again, but it wouldn’t start. If the tank truly was empty then there was probably air in the fuel lines. It would need to be bled out, but I’d never tried that on this engine before. On the other hand, we hadn’t run the engine long enough —according to recent experience—to go through all 100 gallons in the center tank. A blockage, such as a completely clogged filter seemed like it would also be a good reason, as would a fuel pump that wasn’t pumping anymore.

Fuel manifold for selecting supply and return tanks and transferring fuel. Sight glass on lower right just above yellow handle allows for visual confirmation of flow when transferring fuel.

The fuel line between tank and engine is quite long. Each tank has a supply line for drawing fuel out and a return line for running excess fuel back in. Diesel engines are fed more fuel than they can use and the excess fuel is returned back to the tank. Our manifold (which we custom built) can select one of the three tanks as the supply, and one of the three for return. We also put in a fuel transfer pump so that fuel can be pumped from one tank to another, either directly or maybe through external polishing filters, and we can either take a sample of fuel, or put in an additive, simply by opening and closing valves. We can even pump fuel out into a jerry jug if another boat really needs it.

Once out of the manifold, the fuel on its way to the engine first passes through a centrifugal water separator, then one of two Racor filters (only one is used at a time, if it should need changing, the fuel flow can be switched to the second filter and the first one changed while under way). Then it passes through a small electric fuel pump that was installed by a previous owner for the purpose of helping to bleed air from the lines, then on to the engine lift pump which is mounted on the engine and driven by a cam. The lift pump pulls fuel all the way from the tank, through the manifold and filters, and pushes it through the final fuel filter, mounted on the engine, and on to the injection pump which feeds the injectors. That is just about all I know about diesel engines, this one in particular, except that there is a lot that can go wrong in the fuel’s path from the tank to the cylinders, including no fuel in the tank to start with. One of those things was most likely preventing the engine from starting, unless it was something entirely different and far more serious and beyond my simple understanding.

Sitting in the engine room, completely covered in sweat (did I say how hot it was in there?) and looking at the manifold, I noticed that no fuel was visible in the sight glass. I had put a sight glass in the manifold for the purpose of being able to see if any fuel was being pumped by the transfer pump (Sight glass purchased from MSC Industrial Supply). I ran the pump and it easily brought up fuel from the starboard tank into the sight glass. The engine still did not start (I had to try). Then I switched back to the center tank and ran the transfer pump again. I was hoping to get nothing but air, proving that the tank was, indeed empty, but the sight glass filled right up with fuel. That would mean that the center tank was NOT empty, wouldn’t it? I switched back to the starboard tank anyway.

Next up was the centrifugal water separator. Water separated from fuel collects in the bottom of a canister, which then can be drained through a valve. When was the last time I had done that? I found a container and drained whatever was filling that canister. It was very dirty and smelled like diesel, but I couldn’t tell how much might be water. I began to wonder what happens if that whole thing fills up with water. I don’t know. I looked at the Racor filters. The bowl on one of them looked amber, the one actually in use looked black. I switched the valve handle to the amber-colored filter.

Then I realized something. When designing the fuel manifold, I worried about the transfer pump sucking fuel back out of the filters and engine. It probably wouldn’t if there was no air anywhere in there, but to be safe I had installed a valve to close off the supply to the engine so the transfer pump could only pull fuel from the tank. I had forgotten to close that valve when I’d just run the transfer pump, so maybe, maybe, the fuel I saw in the sight glass had not come from the center tank, but from the filters and engine itself. Then I thought, if that supply valve had been left open, couldn’t air get into the system through the fuel return and I could, indeed, suck all that fuel back out? It was too much to think about at this time, but it meant to me that the possibility of the center tank being empty was again the most likely problem. After all, the gauge needle was below empty. I just didn’t want to believe that I had let it run dry.

We have three old Perkins 4-236 manuals onboard. All three of them describe pretty much the same fuel line air bleed procedure, but with slight differences in terminology. There are three ports that you open to let the air out as you push fuel through the lines, closing the ports each in turn as fuel squirts from them. I tried this much of the procedure several times using the electric bleed pump installed inline after the primary filters, and also verified that the lift pump worked with the manual lever, at least. Each time I thought I had made some progress I tried to start the engine. The battery was getting weak. It still didn’t start, but there was still more to the procedure that I hadn’t gotten to yet. I was beginning to wonder why we had never practiced bleeding the fuel lines before we left home.

What I had trouble with was the rest of procedure. After those three ports were closed again, it said to loosen the pipe between the secondary filter and the high pressure pump and manually pump some more until fuel came out of that loosened connection. I could not get any fuel to come out. I even removed the connection nut entirely, and never saw any fuel. So, thinking that perhaps the secondary fuel filter might be clogged (it didn’t make a lot of sense, but I was running out of ideas), I changed it out with our one and only spare. This filter is not meant to be changed often. It is not entirely easy just to get to it. I had to remove the fuel return pipe from the injectors just to have room to get a wrench onto the bolt to get the filter off. After changing that filter, I also changed out the black Racor filter (as good a time as any). By this time it was well into Tuesday, we had sails up, and we were slowly making progress toward Banderas Bay, albeit at less than two knots with some 60 miles left to go. I really wanted to go to bed and “call the guy” in the morning, but there was no guy to call. Joe, the engine mechanic on Slainte, was hundreds of miles away. I wished we were still buddy boating so, if nothing else, he could give me some guidance. Perhaps if we could eventually sail close enough to La Cruz to make contact with the marina, they could send someone out to tow us in and then we could find a mechanic to take a look. That sounded like a good plan, but when I went through the mental image of the whole process, and what it would cost, it just didn’t sound all that good anymore. We already had a full schedule in La Cruz if we were going to leave with other Puddle Jump boats.

I went through the bleed procedure again, and again got nothing out of the feed pipe connection on the high pressure pump. It made no sense. How could I bleed fuel out of ports that were before and after this connection, but get nothing here? I got tough with it, yanked on the pipe connection, and it popped off—and there was the fuel—a lot of it. It had just been a tight connection. Who knows when it may actually have been last taken off.

My next problem was putting that connection back together again. I could not get the screw threads to line up. I was now angry that the procedure had made me take it off when clearly it hadn’t been loosened in years, and there obviously was no air in there blocking things. I bent it this way and that way, in a very limited space, and finally, with even more sweat and bashed knuckles, got it lined up and the threads started. By this time everything—tools, the engine, my fingers, even my face (from wiping sweat) was very slippery from being covered in diesel. I know people who like to work on cars and engines. I think that’s great. It just isn’t my thing.

The last step of the procedure none of the three manuals had any illustrations for. They simply said to loosen the connections on the two high pressure pipes at the atomizer ends (spelled atomiser since it’s British), which I assume were the fuel injectors, and turn over the engine until fuel runs out. But there were four injectors, what are the two pipes? Then, after reading it several times, I finally noticed that it was worded “…two of the pipes…” It did not say THE two pipes. It didn’t actually say how many there could be. So I loosened two of the pipes and cranked the engine, but the battery didn’t have the strength left to turn it over.

When we put in the golf cart batteries for the house bank last year, we kept the two batteries we had been using previously. They are the same type as the starting battery we still use. We had recycled the weakest one and kept the other one as a backup starter battery. I swapped it with the now dead battery. Checking everything one last time, I cranked the engine, and I kept it cranking. At last, not only did a fair quantity of fuel start flowing from the loosened injector pipes, but the engine almost started. I retightened the pipes, cranked one last time, and, after nine hours of trying, the engine finally fired up.

One more problem had been solved in our continuing journey so far, and as a bonus, I finally got some practice bleeding the fuels lines.

I can almost guarantee we will never run a diesel tank dry again. But I will also be very curious as to how many gallons it takes to refill that center tank. There should’ve been as many as 30 gallons left in there.

We actually made it to Banderas Bay and La Cruz before the requested start date of our reservation, which we learned had been accepted to begin on Thursday. We spent all day Wednesday anchored near the marina, wondering why our nearly drained backup starter battery was not automatically recharging. Once safely in a marina slip on Thursday, we did some troubleshooting and discovered we’d blown a 50 Amp fuse (really) in the charging circuit. We’re just not sure yet why that happened.

If you’ve read this far, then you deserve to know that I purposely did not make this a “long story short.” I just wanted to give a sense of some of what takes place onboard. Maybe somebody reading this who is thinking about cruising will actually try bleeding their fuel lines before they are forced to in the middle of the night. We were lucky that the engine didn’t quit at a more critical time.

This post also kind of balances out the idea that we spent the last month living in a tropical beach resort, sipping Mai Tai’s by the pool and playing shuffle board.

Mysticeti (left of center) in La Cruz marina. Also a good view of the wind generator up on the mast.
Finally installing the wind generator in Mazatlan.
Yes Dear, I did sneak one in when you weren’t looking, but where’s the Mai Tai?