Symptoms and Initial Misdiagnosis
The engine was reported to be unusually ‘shaky’ by the boat’s current and previous owners. Because of this, the independent mobile mechanic who worked on the engine (and related systems on the boat) for about 6 years, attempted a series of remedies.
The first ‘fix’ was suggested to the mechanic by the (previous) owner. The idea was to replace the engine mounts with stiffer ones (from a Yanmar 3GM) that would dampen the vibrations. The stiffer mounts prevented the engine from moving around, but made the whole boat shake ferociously.
The mechanic came up with the idea for the second ‘fix’ some years later. He had discovered that the distance between the transmission and the ‘cutlass bearing’ (propeller shaft support) was longer than the maximum recommended by Yanmar. Additionally, the boat had a ‘dripless shaft seal‘ instead of a conventional (old-fashioned) stuffing box, so there was no additional support damping between the transmission and bearing. The mechanic replaced the ‘dripless shaft seal’ with a stuffing box, and also had the propeller shaft re-balanced (as it had been discovered to be slightly out of spec). After this repair, the engine was significantly smoother, but still quite rough.
Measuring RPM & Fuel Injector 'Open Times'
I (as the applications engineer at GTC) had known about this long struggle, and always wondered what the real problem was, so when we released the GTC605 Fuel Injection Analyzer, I went to check on the 2GM, and see what I could find. The first thing I noticed (when measuring with the GTC605) was that the engine was idling about 25% higher than specified in the repair manual. The (40-year-old) analog engine tachometer on the boat turned out to be out of calibration, and showing a deceptive, lower reading.
After measuring the RPM, I used the Fuel Injection Analyzer to measure how long each injector was opening for, as well as looking at the valve waveforms (also known as ‘patterns’). The valve open times turned out to be very different, with one being extremely inconsistent. The valve with inconsistent open times also had a very strange waveform, which indicated that it might be ‘sticking’ open or bouncing. A valve that sticks open or bounces can cause a poor pattern, which reduces the cylinder’s power output, and increases the amount of particulate (smoke) emitted from the engine.
Removing and Rebuilding the Fuel Injectors
After I found that at least one of the fuel injection valves looking suspicious, I decided to have it rebuilt by a fuel injection shop, who could also do a ‘pop test’ on it, to verify that the diagnosis was correct. I could have bought a set of replacement valves, or parts to rebuild the assembly myself, but I decided that it would be best to have a specialist double-check my work, especially because this was my first time using the GTC605 in the field.
Getting the injection valves off the engine was a lot tougher than I’d expected. I had to use a screwdriver for leverage, and a mallet for force to get it out. This would have been much easier if I’d brought a slide hammer, but I didn’t have one handy, because I’d been told that the injectors had been out recently (which seems unlikely).
After getting the parts off the engine, I took them to Fred Holmes Fuel Injection, a local shop that specializes in rebuilding fuel system parts. The friendly technician there pop-tested the valves for me, which showed that neither was working very well, and one of the two was getting stuck open. They rebuilt the valves for me, which cost almost as much as a new set of valves, but was done much sooner than I would have been able to get new parts.
Reinstalling and Remeasuring the Injectors
Putting the engine back together turned out to be much easier than taking it apart (I think that nobody had taken the injectors out since the engine was new). It took a while to bleed the (high-pressure and low-pressure) fuel pumps, but the engine started up easily after the fuel was flowing. The engine was noticeably quieter and smoother than I’d seen before, so we adjourned for the day, and opened the celebratory BEvERages.
A few weeks later, I decided to check the injectors again, (just out of curiosity,) and see how they were doing. The waveforms for both injectors were much cleaner, and I was able to reduce the engine’s idle speed, (it’s just a set-screw on the 2GM,) but not quite get it down to the manufacturer’s specification without having it ‘die’ on me. After adjusting the idle, I used the Fuel Injection Analyzer’s ‘Chart’ view to look at the open times of each injector, and learned that one of them was being opened for much longer than the other. The only reason that these fresh injectors could be opening for different amounts of time was that the high-pressure fuel injection pump was delivering more fuel to one than the other.
Rebuilding the Fuel Injection Pump
Taking out the fuel injection pump turned out to be easier than getting the valves out, though it was a bit tricky because the ‘engine kill’ lever had to be in a very precise location. After getting it out, I took another trip to the fuel injection shop, because a new pump cost about five times as much as a rebuild, and there was a long lead time on the pumps (which are only stocked by Yanmar in their North American headquarters).
The shop rebuilt the pump within a week or two, and removed the superficial rust on it. They returned the parts they’d replaced, which exhibited obvious wear, and must have been original. I took a look at them under a microscope, (a magnifying glass would have been enough, but we’ve got fancy tools at GTC, so I might as well used them,) and they had deep scratches and worn seats. I have no idea how many hours the engine has on it, because it only had an hour-meter installed a couple years ago, but I think it must have 3000-4000 hours on it, so the pump was well past its recommended service interval.
Problems Solved
The engine came back together even easier than it’d come apart, but by the time I was done, I was sick and tired of getting oily fuel all over my hands. I must admit that I tried to take a few short-cuts with bleeding the engine this time, but they didn’t work out, and I ended up having to take my time and do it all properly. It was worth it though!
The engine started up easily, and ran smoother (both in neutral and under load) than it’d ever done before. I don’t think it was any quieter than it had been after the injection valves were rebuilt, but it sounded beautiful to me, (I describe it as the sound of a cow snoring). I adjusted the intake & exhaust valve gaps before calling this project done, but they were all within tolerance, so my dream of finding more (easy) room for improvement was unfulfilled.
That's All!
Thanks for reading about my first adventure with the GTC605 Fuel Injection Analyzer. I appreciate any feedback that you might have, on this post’s subject, style, or anything really, so please send me an e-mail!
If you work on either common-rail or non-common-rail fuel injection systems, particularly Diesels, I hope you’ll check out the tool some time, and consider getting one. Diesel engine diagnostics are tough, the parts are expensive, and we hope our tool can help with your work.