About four weeks ago, our generator suddenly went from normal output to having far less power. Instead of the 10+ kW it normally produced, it could only generate 7 kW. We worked through the Northern Lights diagnostic procedure and concluded we had a valve or valve seat problem on #3 cylinder exhaust. The steps we took and the conclusions we reached are documented in the video below, Diagnosing Low Power Output on a 12kW Generator.
We concluded that the engine problem was very likely a failed exhaust valve seat on #3 cylinder. We were fairly confident we understood the problem, but you never know with certainty until the engine is apart. We’d been holding off taking the generator apart since it’s still our backup power source even though it’s suffering from internal mechanical problems.
Last Friday night the new cylinder head and other seals and gaskets needed for this job arrived here in Stornoway. Within minutes of the ferry arriving, a Stornoway Port Authority worker had transported the fairly heavy parts all the way to our boat, since they knew we were anxious to get this generator back online. Our thanks to everyone who works at the port. They have all been very helpful to us and we really appreciate it.
Saturday morning, the old head came off. The #3 exhaust valve, at the far left in the photo above, was visibly much higher in the cylinder head than the rest of the valves (click image for a larger view). For context, the clearance in the valve train, also known as valve lash, is adjusted in thousands of an inch. But this problem is visible from a meter away without accurate measuring tools.
In the picture above we’ve turned the old cylinder head over to have a more detailed look at the problem. The combustion chamber looks normal, where it’s all black with light carbon deposits. Here we’ve scrapped the carbon off and rust is visible beneath, which suggests there may have been water intrusion sometime in the distant past.
We have carefully checked the exhaust against the manufacturer’s specifications and this installation meets their specifications. We’ve also checked for excessive water flow back into the water lift muffler on engine shut-down. If excessive water is in the muffler, it can splash back up into the engine in rough water. That doesn’t appear to be a problem either. Nonetheless, it looks like there has been some water in this engine sometime in the past. We need to dig deeper into this issue.
We don’t have the right tools to remove the valves, but the valve springs are fairly light on this small engine so we were able to remove them and have a more detailed look at the valves.
In the picture above, we’ve removed the exhaust valves from cylinders #2 and #3. The #3 exhaust valve seat has recessed a long way into the cylinder head. Sufficiently so that this cylinder head no longer is serviceable, since these heads don’t use removable valve seats. Both of the removed valves have very poor sealing surfaces. A normal valve would have a polished section where the valve contacts the seat. These valves have been sealing poorly and leaking.
We also checked the intake valves and saw the same thing. They also have been leaking. All the valves and seats are in poor condition. Looking at the removed valve heads in the pictures above, we see that there is a section in each that is heavily corroded (click image for a larger view). But most of the valve is not corroded at all. Valves rotate in the cylinder head in normal operation, so seeing valves with corrosion on just one side suggests they were in sea water, but probably only once. And for damage to occur, they would have had to had sat like that for days, if not weeks.
As the valve lash reduces, it has little impact on engine operation, until it starts to hold the valve open. At that point, an entire cylinder stops producing power. That’s why the output suddenly dropped from 10+ kW to 7 kW. And it’s also why the generator can sustain 7 kW even though an entire cylinder is not operating. Unsurprisingly, it’s running at roughly two-thirds of its normal output, since it has two out of three cylinders operating.
During our North Atlantic crossing back in 2017, we saw some very rough water (see video North Atlantic Gales). Our hypothesis is some seawater splashed into the engine from the muffler during those storms. It may even be the case extreme heel angle or large wave pressures forced water in from outside through the exhaust system and all the way into the engine. And since we landed at a marina with good power in Kinsale, Ireland, the generator didn’t run for weeks after the trip.
We suspect some sea water was in this engine for a few weeks after the North Atlantic crossing. This caused only minor corrosion in the valve and seat area, but is probably why the engine needs valves and seats at 6,700 hours rather than 15,000 to 20,000 hours.
The lesson we take away from this inspection is, whenever the boat sees extreme weather or is knocked down, we will run all engines and the furnace to ensure all are clear and no moisture is anywhere in the system. It’s a good lesson and one we’ll follow going forward.
I really enjoy watching your detailed and well put together videos on the boat systems!
I have a friend that worked as a marine engine mechanic and Volvo Penta Dealer for 40 years here in Sweden. On his own boat he constructed a Y-valve for the seaside (Saltwater) side of the exhaust.
This valve redirects the waterflow from before the heat exchanger to flow just overboard. This is used maybe 10-30sec before turning the engine of. (Running remaining water out of the exhaust to keep it dry.
The thing he saw from all these years of engine work is damage to valve seats on these engines after many hours of use. Reason: Because for the last second when the engine is turned off it backstrokes just a little and pull just a little of that saltwater back to the exhaust valve. Then it sits and corrode just a tiny bit every time the engine rests.
It’s a super interesting theory. An engine will often bump against compression on shutdown and turn a bit the wrong way. It seems unlikely that it would turn enough to pull water up but it’s possible. I’ve checked every aspect of the exhaust design and there should be no way to get water in there in other than extreme conditions and these don’t really happen all that often. Your friends theory is worth considering. In support of his thinking, when tearing the engine down for the cylinder head change, I did notice that the exhaust elbow was wet and it troubled me a bit.
I don’t see a way to act on this theory since our gen often runs when we are not on the boat or during the night. I’m not around to go down and shut off the water supply each time it’s to be shut down. I do find the theory super interesting and fairly credible. Thanks for sending my way.
This backstroke as I call it, will probably only just make a small mist of moisture back from the elbow.
The thing that was getting my attention was that your cylinder #1 is the most worn and the one closest to the exhaust elbow, so that could maybe support this theory.
I agree this valve arrangement would be difficult to fit on a genset and it had to be automated in some way to work.
Another interesting thing is the similarity between this Lugger and a Volvo Penta D1-20. I guess only the parts around the actual engine is different between the brands ?
Yes, I think it’s entirely possible that mist and condensation around the exhaust elbow could make there way back up into the engine on shut down. It won’t have to be very wet cause problems and for engines that are off and on frequently as ours is, even more of a problem. If I could come up with a way of eliminating that I would do it.
Your right that the Volvo D1-20 is exactly same base engine. Perkins also offers this engine as part of their product line.
If a worn valve seat, wouldn’t you see poor performance immediately and not have gone 39 hours?
What happens in this situation is as the valve seat wears away the valve rises in the head and the valve lash goes to zero and then the valve starts to get held open. As long as there is still valve lash and the valve isn’t leaking badly, the engine will produce full power. As the valve lash closes up, the engine continues to have full power. But once the valve lash goes to zero and the valve is held partially open, that cylinder power will quickly drop to zero. If the valves are adjusted again, the bad cylinder will again return to near full output as long as the valve continues to seal. That’s why a valve adjustment would return the engine to full power even though the valves seats were eroding badly.
Taking off that valve cover: is it stuck down with some goop or just a simple gasketing?
The head gasket between the cylinder engine block and the cylinder head is installed dry but usually both sides stick so taking it off will require some force to tear the head gasket once the bolts are all free. Between the rocker carrier and the head there is an O-Ring without any required sealant and it won’t stick at all. The valve cover to rocker carrier interface is an O-ring as well. Again, no sealant is required.
Well, that just goes to show you- It’s always something.
Bummer that this was sort of self-inflicted, if you’d known you could have run the gen. Were you seeing un-burned fuel on the water when running? I wonder if a a flap on the discharge would have made a difference?
Hey Steve. On this one I don’t fully agree that the problem was “self inflicted” but I do agree that if we were more conscious of what might have been happening we could run the engines subsequent to those few really rough water events and would have either found the engine hydro-locked (easy to deal with if you deal with it very quickly) or running it would boil off any issues and leave the engine undamaged. So, by that measure its’s fair to call it self inflicted. We didn’t take a precaution that could have prevented the issue. But, the reason I’m not 100% on board with “self inflicted” is ocean going boats should be able to take weather without equipment damage and, specific to exhaust system design, the design should keep water out of the engine.
As an example, we experienced some rough weather when the boat was less than a year old and the wing took water up through the muffler system. I cleaned it out immediately, change the oil several times and run it under high load. I’m 100% confident the engine life will be in no way impacted. But, it really shouldn’t have happened. Yet, in looking at the exhaust design there appear to be no mistakes. It appears the PAE design meets the Northern Lights spec. It’s compliant with the manufacturers specifications. But, on the other hand, the engine did get flooded and many others had the same issue and had to have the engine rebuilt. We put a Centek check valve on the Wing and we have seen worse weather since and the problem never repeated. The specs should be adjusted to prevent this issue in my opinion and I’m arguing a check valve should be a standard component Given the protection is cost effective, I would love to see it adopted as the standard for ocean going boats.
Our generator has an AirSep that is up in the Salon settee seat back so our thinking is that it’s already protected as well as the wing with a check valve. It might be the case that the generator did have water forced past the AirSep or it might be the case that the AirSep has prevented excess water intrusion and it wsa just residual water in the muffler that has been splashed into the engine by rough seas. I have no way of knowing if the water was forced past the AirSep or if it was just splashed water that came up from the muffler on a 40+ degree heel from a wave action. If it’s the former, a check valve would help. If the latter, even a check valve protected engine would still be at risk. My thinking is that it’s probably residual water from the muffler being sloshed up into the engine on a heavy wave hit.
We also see that the exhaust valve is only corroded on one side which argues that this might have been a single event. Thinking this through, my general advice is check valves should be part of any wet exhaust design and, even with that precaution, water can still splash up from the muffler and, because of that, engines should be run after extreme conditions to ensure that salt water doesn’t sit in place. If those recommendations where both in place back in 2010 (or even today), then I agree this event was self inflicted :-).
I’m glad you got the head replaced with so little effort. It always seems to be my luck if I go any deeper than the valve covers everything snowballs.
I hope you figure out what’s causing the water/moisture intrusion.
One thing I noticed is if you can find the correct eye-bolt or shackle you now have a fine sentinel for your anchor.
:-). Because the valve heads have only rust on one side and they are components that, in normal operation, rotate in the head I suspect the damage was done in one incident where the engine sat for a while with water in it after the incident. I can’t say for sure that’s what happened but that is where the data point me and we have had a couple of horrendous weather days so I can think of when it might have happened. I suspect if we had run the engines subsequent to those events, we probably would have either found the engine hydro-locked (easy to deal with if you deal with it very quickly) or running it would boil off any issues and leave the engine undamaged.
Was your 69.1 degree roll at Fraser Island a possible timing for water ingress?
Hi James, Nice story and wonderful that you found the cause. I also have a NL (6KW) and will next to the water intake also close the ball valve in the outlet as well as the ball valve in the water outlet during navigation. Good luck with the oil seal change. Regards, Rob