diesel engines - a cautionary tale

[pappaecho]

I replaced a 28 year old Perkins 4108 last year with a Beta 35.
Two weeks ago coming out of Yarmouth harbour we had white smoke out of the exhaust. the engine was shut down and we sailed home. On inspection the sump has about 3 litres of sea water in it. Beta suspected that a faulty "antisiphon valve " was to blame, Not so!
It transpires that as the engine is below water level, if the sea cock is left on, water passes through the raw water pump and fills up the exhaust manifold, which then leaks in the cylinder head and via an open valve into the sump.(It was open for a couple of days )We always left the seacock open on the 4108, because some years ago we had to exit the berth very quickly when a boat caught fire, and on that occasion we over heated the engine because in haste the sea cock was left closed.The Perkins did not have this problem
So the moral of the story is to leave the seacock closed at all times when not in use.The Beta manual makes no mention of this possibility.
A second sorry tale as a result of the first. The engine stopped mid passage even though the emulsion was immediately removed from the sump and the engine flushed with 2 lots of flushing oil before a third fill was made.The fault was a seized injector pump caused by corrosion due to the seawater. The pump was changed and the oil was good and up to the dipstick mark.
The engineer cranked and cranked the engine to rid of air in the system, and after 3 long sessions of cranking it finally started..........but the oil pressure light came on. On stopping the engine it was found that the sump was full of water again!
Not the same reason. The excessive cranking pumps water through the engine and into the exhaust manifold, but as the engine is not running, no water is expelled via the exhaust, so the exhaust manifold fills up with water and drains into the sump.
So the moral of this story is to ensure that if you are cranking the engine for more than a few seconds, turn off the sea water cock, so no water is pumped through the engine into the exhaust. This must apply it seems to all engines whether above or below the waterline.
Again this problem is not documented in either Beta or for that matter the Perkins manual.
As the Beta is indirect, I shall in future start the engine before opening the seacock, and close the seacock before turning off the engine

 

[CaptainBob]

Won't you damage the impeller if you start with the inlet seacock closed?

 

[jerryat]

[ QUOTE ]
Won't you damage the impeller if you start with the inlet seacock closed?

[/ QUOTE ]

No, you won't cause damage if it's for short periods (like starting or trying to start the engine) because the speed it's truning is very slow relative to when the engine is running so it doesn't build up much heat.

It's a well known principle to shut the seacock before multiple starting tries. Just open the seacock smartly whe it starts. Better a new impeller than a flooded engine as noted above!!

 

[cliffordpope]

Standard Yanmar starting and shutting down advice.

 

[VicS]

[ QUOTE ]
Beta suspected that a faulty "antisiphon valve " was to blame, Not so!
It transpires that as the engine is below water level, if the sea cock is left on, water passes through the raw water pump and fills up the exhaust manifold, which then leaks in the cylinder head and via an open valve into the sump

[/ QUOTE ]

Surely in any installation in which the exhaust injection point is not well above the water level an antisiphon valve MUST be fitted. It will then prevent this sort of thing happening

It must be situated in a loop of pipe between the engine itself and the exhaust injection point and be 40cm above the waterline.

I don't understand how the situation you describe can arise if there is a correctly installed and operating antisiphon valve. In fact it wont. That's the whole point of the device.

IF you do not have an antisiphon valve fitted FIT ONE without further delay 16", or thereabouts, above the water line (12" at least).

If you do have one check that it is free of salt deposits etc regularly
[ QUOTE ]
The engineer cranked and cranked the engine to rid of air in the system, and after 3 long sessions of cranking

[/ QUOTE ] Any "Engineer" who cranks an engine for 3 long sessions with the seacock open is no engineer at all. He needs a f*****g good kick somewhere where it will be very painful for a long time

 

[Bilgediver]

I am afraid this is indicative of a poor installation and nothing to do with the quality of a Beta engine.

All impeller pumps leak to a degree so it is up to us or the installer to ensure he puts a barrier between the sea and the exhaust mixing elbow. This requires an anti syphon valve or loop fitted high enough that water cannot pass it to the mixing elbow. I prefer loops with an outlet to a skin teltale as there are so many tales of anti syphon valves sticking.

There is always the chance of flooding an engine with sea water due to excessive cranking as the water pump is doing its job but there is no exhaust flow to clear the water.

I have seen similar problems on installations of Yanmars etc when mounted below the waterline in deep hulls.


The correct anti syphon loop is essential and closing the sea cock when cranking for long periods likewise.

 

[VicS]

[ QUOTE ]
I prefer loops with an outlet to a skin teltale as there are so many tales of anti syphon valves sticking

[/ QUOTE ] To which end Vetus, for example, supply two versions of their antisiphon device, One with a valve, the other without a valve but with provision to discharge to a skin fitting.

The latter type I believe has a "restrictor" in the end of the discharge tube so that it discharges a stream like an outboard engine "pee hole"

 

[earlybird]

I have a Beta, not that the make matters, installed well below the waterline, but of course with an anti-syphon valve fitted, and also a water-lock, absolutely no problem. This is typical of many, many thousands of installations. There should be no need whatsoever to worry about the seacock being open so long as the a/s valve is doing its job properly. Mine is of the continuous bleed type, so there's little to go wrong.
Filling the engine with seawater by prolonged cranking is a rather different issue. A supposed "engineer" doing this is either very forgetful or incompetent.
Add; agree with lw395, I should have also said that water through the exhaust valve, whilst the engine is being cranked, is likely to cause severe damage before it gets into the sump.

 

[lw395]

This does not quite add up.
Although I agree that an anti syphon loop is required, I would be concerned that something else is afoot.
Water leaking into the exhaust while the engine is not running doesn't have a path to the sump until the exhaust has filled up. There should be no need to start the engine with the seacock closed as the impeller pump will take a long time to fill the exhaust to this level. Only exception is when it is slow to start for some reason, eg if you are bleeding the injection.
If there is water in the cylinder, turning the engine will result in bent conrods or similar. Unless the rings are shot, leakage from the cylinder to sump will be quite slow.
This isn't a raw water cooled motor is it?
Is there any other possible route from seawater to sump?
Best of luck with it.

 

[rob2]

when installing a "new" Yanmar on my mate's boat, we put a diverter valve between heat exchanger and injection elbow so we can send the water to a point downstream of the swan neck. All starts and stops are done with the flow diverted so as to prevent filling the exhaust and to allow the last few seconds of running to blow it clear. Simple, eh?

 

[lw395]

Simplest solution to exhaust filling up is to have a silencer at the bottom of the run from the mixing elbow, with enough capacity to take a minutes worth of cooling water. If the engine won't start in a minute, you definitely have a problem. If the starting and charging systems are as they should be it's a non-problem anyway.
Having non standard valves etc can catch people out if you lend someone the boat. I realise you have to tailor the system to the boat though.
We've drifted a bit from PapaEcho's problem, which I hope is swiftly fixed.

 

[starfire]

I prefer the latter, you can tell at a glance you have cooling water & if you are washing the deck & the sea is cold, you somewhere to get warm seawater !

 

[boatmike]

We've drifted a bit from PapaEcho's problem, which I hope is swiftly fixed.
__________________________________________________________________

No you haven't. This is a question of really poor installation without reference to normally accepted principles. Both you and others are right. I would only say its not a question of "minutes" it's static volume. There should be sufficient capacity in the exhaust wet box silencer to take all the volume of the water in the standing pipe from the wet box to the skin outlet in one direction and back to the manifold in the other. If not the entire volume, sufficient for the water in the exhaust system in total ABOVE the manifold to drain down to a level BELOW the manifold. On the inlet side the water delivery pipe should also be mounted with a swan neck that raises well above the waterline and there should be an anti-syphon valve in addition to this which is also well above the waterline. I agree with the post that prefers constant delivery ASVs to those with valves but the flow on these should be checked regularly too as they can become blocked. If this engine were installed that way the problems described would not be possible.

 

[sailorman]

Sorry to hear of your probs i have the nanni 37.5 hp
the inlet seawater supply loops as high as the cockpit sole will allow thia is about 200 m/m above the engine rocker cover. there is a syphon break @ the highest point of thia loop that "piddles" into the cockpit ( i can then see whats happening ) approx 200 m/m above the cockpit sole.
i always leave the inlet seacock "On " when aboard.
I did the engine installation 6 yrs ago (& clocked up some 650 hrs to date) my self after removing a perky 4108.
your probs are due to installation + an engineer who doesnt know his job

 

[petersto]

A very sad and I expect expensive tale. I have often worried about this happening on my boat. My problem seems to be grasping exactly where the ASV should be fitted between what and what exactly. My boat is a Centaur with an MDIIC.

 

[PuffTheMagicDragon]

[ QUOTE ]
[ QUOTE ]
I prefer loops with an outlet to a skin teltale as there are so many tales of anti syphon valves sticking

[/ QUOTE ] To which end Vetus, for example, supply two versions of their antisiphon device, One with a valve, the other without a valve but with provision to discharge to a skin fitting.

The latter type I believe has a "restrictor" in the end of the discharge tube so that it discharges a stream like an outboard engine "pee hole"

[/ QUOTE ]

That is what I installed when I changed my engine, the main reason being that there is no valve to get clogged.

I am not aware of any 'restrictor', however; there is just a little nylon fitting to which the bleed tube is fitted. I preferred to fit this 'pee hole' in the cockpit, rather than over the side because I can tell at a glance that the thing is working without having to look over the transom; especially useful at night. One other advantage when I still used to fish was that I could rinse the fish muck off of my hands in the tell-tale stream.

 

[pappaecho]

To clarify certain points in this post. This engine was professionally installed. The existing Perkins anti siphon loop was working and has been tested again and antisiphon is NOT the problem. The Perkins water trap/silencer, Swan neck and exhaust system are original and worked perfectly for 28 years, and the Perkins was a larger engine with more water throughput than the Beta

If you leave the sea cock on, and remove the hose to the exhaust elbow, and let it drain into a bucket, it fills up at a rate of 1.1 ltres per day ( 24hrs). This is nothing to do with siphoning, it is due to pressure of water through the raw water pump. Clearly the Jabsco on the Perkins did not leak, this one does.
As the raw water pump is filled with water even when the seacock is off I do not expect any problems with damage to the impellor for the short period of time before the water is allowed to flow.
The engine was fitted by a Beta agent it was not self installed.

I am thinking on the lines as follows to provide a belt and braces solution to the problem.

I can source a 3/4 inch 12 volt solenoid operated stop valve ( which draws 27 watts) If it is inserted close to the exhaust elbow, and wired into the ignition it will prevent the first problem of an open seacock allowing water via the raw water pump and heat exchanger to flood the exhaust system.

However it will not solve the second problem of prolonged cranking with the seacock open, again filling the exhaust.
It occurs to me that if a slave relay can be connected into the alternator charging circuit, then the engine has to be running before the relay will open, and so both the " drain down " and filling up due to excessive cranking will be eliminated. However as the device uses 27 watts, I assume I will have to fit a relay or possible a Mosfet?
Any ideas?

 

[moody1]

Friends - they had their bilges flood one time when boat was left and drains blocked. The water level reached sump level on the engine. They pumped out, cleaned the boat, dipped and checked the oil in sump. All appeared fine.
Started engine used it for a weekend trip. For return from final stop, when pressing start button all they heard was click of solenoid. Thinking battery may be too low, they borrowed another to see if they could jump it. Click.
They sailed the boat home without engine, even to berthing in home marina.
They tried everything to get the engine to turn over, even a socket and bar on the crankshaft pulley. Engine was completely locked solid.
When the engineer changed the engine, they asked him to check the old for reason. It was found that SW had entered the cylinders and sump, it had basically hydro-locked it and then corrosion / use had literally welded the piston rings to liners. The engineer couldn't get them to move even when engine was part stripped.

As to solutions to the original post - they have been covered by others, I just thought I'd pass on our friends experience with water in sump.

 

[lw395]

Your solution adds complication and failure points. The extra valve can either leak due to corrosion, scale, contamination or fail to open due to electrical or mechanical failure. Personally I would be much happier with the anti syphon loop and syphon breaker properly installed, plus adequate exhaust volume to accomodate a sensible amount of cranking. Note that if your exhaust volume is low the engine is also (more)vulnerable waves up the stern 'pooping' the exhaust, with consequences of filled cylinders etc (happened to a friend's boat). Sorry to be negative, but a solution with minimum moving parts is often better imho.
BTW, your comment about water pressure vs syphoning is kind of the point. The water level outside is above the mixing elbow. So you add the loop to raise the highest point above waterline. This becomes a syphon in the classic sense (a hose going from high source of water to lower outlet via a higher point), so you add a syphon breaker to let air into top of loop when water is not being pumped. The term 'anti syphon loop' is a bit misused, unless it means a loop with an air inlet.
Hope its all fixed soon,
Cheers,

 

[VicS]

[ QUOTE ]
If you leave the sea cock on, and remove the hose to the exhaust elbow, and let it drain into a bucket, it fills up at a rate of 1.1 ltres per day ( 24hrs). This is nothing to do with siphoning, it is due to pressure of water through the raw water pump. Clearly the Jabsco on the Perkins did not leak, this one does.


[/ QUOTE ] Yes but if the plumbing anywhere between inlet and exhaust injection is taken to a point well above water level and an air bleed (antisiphon valve, ) is fitted there you will not get your flow rate of 1.1liters per day because although you will have created a potential siphon it will be broken by air being admitted at the highest point.

Simple basic physics!

If your engine was merely installed with the exhaust injection point below water level without any form of siphon break as you are saying the so called professional who did it was a complete and utter incompetent tosser . You should get you hands on him and present him will bills for all the damage that has occurred as a result.

There is no need to go to the extent of solenoid valves etc. TOTALLY UNNECESSARY. Rubbish.

Take the hose between the engine and the exhaust injection point up 16" above the water level and fit an antisiphon valve at that point. If for some reason that hose cannot be so diverted then the hose between pump discharge and engine will do, even the hose between inlet and pump!

You imply that the original Perkins antisiphon valve is still fitted and works but where the hell in the system is it fitted. If it allowed your engine to fill with water it is not frigging well working!