Marine Diesel Thermostats: NOT WHAT THEY SEEM

[Andrew Troup]

Executive Summary:

1) Thermostats on marine diesels generally operate differently from automotive thermostats

2) Saltwater-cooled diesels are particularly different, and troubleshooters need to take note

3) Most small marine diesels will overheat if run without a thermostat

4) Overheating can be caused by thermostat assembly even when opening temperature OK


More to follow !

[Andrew Troup]

Thermostats on many (typically small) marine diesels don't seem to be well understood or documented, either in books or on the internet.

Searching through posts on this topic reveals only once or twice on this forum where this has been mentioned in the past, and as far as I can see the forum has seen no detailed explanation of a) how they differ
b) how to tell what sort you have, and
c) why the simple automotive sort would not be suitable

Raw-water cooled motors generally have a thermostat arrangement which closes off flow through the (cold) engine, diverting coolant to bypass the engine altogether and flow straight to the exhaust when the thermostat is closed. This is quite different from usual automotive practice, in which the flow is diverted back the pump to recycle through the engine, and it is the radiator which is closed off. This makes them counterintuitive to troubleshoot.

The most thorough coverage of the topic I can find in the literature is in the excellent book "Diesels Afloat" starting on page 79.

This can be viewed on the "Scribd" website.

However even in this book, things get confusing once they get to the topic of fresh-water "indirect" cooled engines.

The text on this type of cooling describes one sort of thermostat control, similar to an automotive circuit, (and to, say, the Yanmar 2GM20F) but the adjacent diagrams show the other sort of thermostat control, similar to a raw-water engine described above, where the thermostat diverter circulates coolant through the heat exchanger when the engine is cold, rather than through the engine as described in the text.

Then they compound the confusion by seeming to describe the thermostat circuit IN THE DIAGRAM, incorrectly, as being "identical to the way in which water is directed through the radiator on a car". This latter problem may be a typographical accident, but the fact that it was not picked up shows how confused even the experts can be.

[Andrew Troup]

Cars have different cooling systems from marine diesels in several respects. Firstly: they do not have a total loss, saltwater circuit, incorporating a positive displacement pump, and secondly, they do not have a wet (salt water cooled) exhaust system.

Car systems are particularly different from saltwater cooled engines, where the total-loss system runs through the engine, with no possibility of recirculation.

I imagine it is because of the latter that small marine diesels evolved a different type of thermostat from automotive installations: even many of those which are fresh-water cooled share design features with their salt-water cooled kin, and in the latter case, there needs to be a constant bypass flow to cool the exhaust system, even when the engine has not reached operating temperature.
And this constant flow is also essential to prevent the impeller pump from running into a brick wall: after all, it's a "positive displacement" pump.

This constant flow is achieved on small raw-cooled marine diesels by closing off all flow through the engine block* at the outlet (except for a small bleed flow through a hole in the thermostat valve disc) and sending the entire flow from the impeller pump via the bypass hose to the exhaust.

What makes it confusing is that the diverter valve is situated at the point where the flows re-converge, not at the divergence point.

It's like a cop controlling the proportion of traffic using a detour, by blocking traffic exiting from one and then the other of the two alternate routes.

* Small, heavily built diesel engines can get away with this, whereas in larger automotive engines it is necessary to circulate coolant through the block even when the entire engine is not up to temperature, to relieve local hot spots.

In automotive engines (and in most closed circuit, fresh water cooled marine engines), the bypass flow when the thermostat is closed travels in the reverse direction, back to the circulation pump, to do another trip through the (warming) engine.

[Andrew Troup]

Here's a cross-section through a Bukh thermostat from an old marine diesel book (now out of print)



I've photoshopped it to show the fully closed and fully open conditions, and labelled the ports.




This is similar to the arrangement on direct (raw salt water) cooled Yanmar GM engines, which I will show in later photos.

Notes to diagram:
In the case of "raw" (salt) water cooling: the bypass water from the right is cold salt water from the saltwater (impeller) pump.

In the case of closed circuit, fresh water cooling: the bypass water from the right is recirculating fresh water, from the freshwater (centrifugal) pump, drawn from the heat exchanger.

In both cases: The water flowing to the left is either that same bypass water, or (if the engine is nearly up to temperature) a mixture of that bypass cold water and warm water from the engine block. When the engine is at or above operating temperature, the water from the right is blocked off, and all the water flowing to the left is hot water from the block.

In the raw water case, the water flowing to the left goes to the exhaust injection elbow.
In the fresh water case, the water flowing to the left goes to the heat exchanger.

Referring now to the diagram showing a section through the thermostat housing of an indirect (fresh water closed circuit) cooled Yanmar GM engine:



At first glance this looks like a simple automotive "Open or Shut" valve. However if you look carefully at the upper extremity of the "Thermostat valve" the left hand diagram shows it shutting off the port to the heat exchanger when the engine is cold.

So this is also a diverter valve, like the saltwater item, but the port which opens is served by a different part of the valve from the port which closes. And rather than opening the bypass route when the engine warms up, the indirect cooling thermostat closes it.

As with any fresh water cooled engine, the salt water is pumped by an impeller pump through the core of the heat exchanger and from there goes to the exhaust injection elbow to be discharged overboard.

Unlike in a motor vehicle, this "diverter valve" characteristic means you CANNOT leave the thermostat out of a diesel with such a thermostat, and expect it to run cooler. It will almost certainly overheat.

This is not actually recommended for any marine diesels I've checked (or, for that matter, even for modern automobile engines, which in some cases will overheat or heat very unevenly)

[Andrew Troup]

Thermostats also acts as a thermal regulator, by changing the amount of flow proportionately as the temperature rises or falls; even in the simpler automotive circuit, it's not a simple on-off.



This is the thermostat from a 1GM10 direct-cooled Yanmar engine (after soaking in vinegar, scraping, brushing vigorously and blasting with an espresso steam wand). It is similar to the Bukh shown in the first diagrams in the previous post. In this picture it is fully open, at about 115 deg F

The small bleed hole through the shutoff plate is provided so that there is always a small flow through the valve, so that it is exposed to a more representative coolant temperature than it would otherwise be in the cul-de-sac at end of cooling passages it's situated in. It is important to run the correct sized drill (one which is just small enough not to take a metal shaving) through this, as they block easily.

The Bukh thermostat shown in a previous post has an airfilled bellows. Most modern thermostats use a wax capsule, which is much more reliable.



Another possible cause for overheating to check, while the thermostat is out of the engine, is to see if the cooling passage inside the block, leading to the thermostat, is partially blocked.

For cleaning the thermostat and housing: use white vinegar, and a toothbrush. The steam wand from an espresso machine is great both for cleaning duties and to vary the temperature in the tin mug for checking the opening temperature.

IMPORTANT:
The bypass control can degrade over time due to wasting away of the female port machined into the housing of the thermostat, and/or a buildup of calcareous and chloride deposit. This failure to seal may lead to overheating as too much coolant is allowed to bypass the block.
It may be possible to machine this smooth, taking a similar amount off the mating face where it bolts to the block. At a pinch, a decent machine shop will be able to make a serviceable housing from scratch, with some guidance as to what is important.



This is the thermostat housing from a 1GM10 direct-cooled Yanmar engine. I have not machined it yet: it just needs a lick off the inner face which lies parallel with the mounting surface. However when first removed from the engine it had a heavy buildup of scale and salt mixed with corrosion product.

- - - -

NOTE: Some recent automotive engines have more sophisticated thermostats which operate on the three-way valve, or "diverter" principle, like the one shown for the Yanmar indirect cooling circuit.

[Andrew Troup]

My underlying reason for my posting this info is rather perplexing, and it is essentially this:

Thermostats are rather reliable.

For this reason, on the rare and widely spaced occasions they have played up for me in a lifetime of sailing and messing around on boats, I have always muddled through, realising each time that something about the circuit was not as I expected, but not really putting the whole puzzle together.

And I rather suspect that's how it is for many of us: things don't quite add up, but we manage to get the system sorted and forget about our puzzlement.

And that's how it was and remained for me, until the last time I struck problems, and was lucky enough to be in the presence of a true master of troubleshooting (with, I might add, scant experience in this particular field, but a phenomenal ability to work things out from first principles).

In a couple of sentences he crystallised my thinking, and I realised the situation of a saltwater-cooled engine was unique and different.

So I did some research, found out that even some of the recognised sources are silent or confusing on this topic, and thought I would set out my best understanding of what is going on.

I would welcome correction or clarification on any points, or further enlightenment. Particularly it would be good to know which engines have what sort of thermostat circuit.

TYPE A: Three-way valve, similar to 1GM10/Bukh (direct cooling)

TYPE B: Three-way valve, like 2GM20F*(indirect cooling)

TYPE C: Three-way valve, similar to 2GM20F but circuit like 1GM10**(indirect cooling)

TYPE D: Two-way valve, similar to traditional automotive (please specify type of cooling)

TYPE E: Two-way valve, opposite to traditional automotive (please specify type of cooling)

TYPE F etc: Other (please describe in detail)

(Note that by "Three-way valve", I don't just mean "three ports", as many automotive thermostat housings have three ports - but that there are three ports whose mutual interconnection is affected when the thermostat changes state, as say in a Y-valve or divertor valve)

*2GM20F: refer diagram in earlier post showing "indirect (fresh water closed circuit) cooled Yanmar GM engine"

** In Other Words: when the engine is cold, there is no circulation through the block. Instead, fresh water circulates through the heat exchanger

[Blue Stocking]

Thanks Andrew, interesting reading, over my 2nd cup of tea this morning.

[Wotname]

OK old thread I know but pertinent for me today and perhaps my post will lure Andrew T back to posting again - I do note he hasn't been on line for some 6 or 7 months now.

Anyway others may also be able to help

My 2GM20 thermostat appears to be stuck in an half way open position but I wondering if anyone can confirm this and / or advise if it is in anyway user serviceable?

In essence, it does not move from when heated - I have tried heating it from 0 though 100 degrees C (32 to 212 F).

It has been cleaned with vinegar and water and can be opened further with finger pressure compressing the larger spring. Next step is to remove the cir-clips and disassemble to see what is stopping it closing but my guess is that a new one is required

No engine trouble has been experienced, I have come across the problem with carrying out a major inspection on the engine (12 years old, 6 years in service, 280 hours).

[charliehows]

my 2 bobs worth - 0 -100C with no movement = toast. Very much doubt if they are serviceable, i generally consider them a consumable. Always worth a try though, you've got nothing to lose pulling it apart. Trouble is those sort of simple heat activated units usually rely on varied expansion rate bi-metal levers of some sort which i doubt can be replaced/obtained. never know though.

[leftbrainstuff]

Originally Posted by Wotname

OK old thread I know but pertinent for me today and perhaps my post will lure Andrew T back to posting again - I do note he hasn't been on line for some 6 or 7 months now.

Anyway others may also be able to help

My 2GM20 thermostat appears to be stuck in an half way open position but I wondering if anyone can confirm this and / or advise if it is in anyway user serviceable?

In essence, it does not move from when heated - I have tried heating it from 0 though 100 degrees C (32 to 212 F).

It has been cleaned with vinegar and water and can be opened further with finger pressure compressing the larger spring. Next step is to remove the cir-clips and disassemble to see what is stopping it closing but my guess is that a new one is required

No engine trouble has been experienced, I have come across the problem with carrying out a major inspection on the engine (12 years old, 6 years in service, 280 hours).

Bin it. They're not serviceable.

Sent from my SM-N900T using Cruisers Sailing Forum mobile app

[Wotname]

Originally Posted by charliehows

my 2 bobs worth - 0 -100C with no movement = toast. Very much doubt if they are serviceable, i generally consider them a consumable. Always worth a try though, you've got nothing to lose pulling it apart. Trouble is those sort of simple heat activated units usually rely on varied expansion rate bi-metal levers of some sort which i doubt can be replaced/obtained. never know though.

Originally Posted by leftbrainstuff

Bin it. They're not serviceable.

Sent from my SM-N900T using Cruisers Sailing Forum mobile app

Yep, it was crook, new one on order.
Still I found it an interesting exercise to pull it apart, especially as it failed half open rather than fully open or closed.

[El Pinguino]

Originally Posted by Wotname

Yep, it was crook, new one on order.
Still I found it an interesting exercise to pull it apart, especially as it failed half open rather than fully open or closed.

At least it failed half open and not half closed

[nuku34]

A very late addition to this thread: As a Bukh DV20 owner, I’ve had experience with the overheat problem. I’ve never removed the thermostat to “fix” this problem because the thermostat is there for a reason (otherwise the engine designers would have left it out).
The thermostat is there so that a cold engine will come up to proper operating temperature soon after it starts, and then to regulate the engine temperature according to the load on the engine and temperature of the raw water coming into the engine.
The proper sequence is that the thermostat is “closed” when the engine is cold so that only a small amount of water from the “Y” is allowed into the block and the exhaust manifold via the “steam pipe“ which inserts into the block. That allows the water in the block to heat up quickly (engines are efficient when they run at designed temperature, not cold). Most of the pumped raw water is diverted via the bypass hose into the exhaust elbow to cool the exhaust and protect the rubber exhaust hose. As the water in the block heats up, it also heats the water in the exhaust manifold and that heated water causes the thermostat to begin to open allowing water from the block to flow out to the exhaust elbow. As the engine runs, the thermostat controls how much cold water is allowed into the block.

Some Bukh DV20’s only have a single connection on the water temperature sender which is screwed into the thermostat housing. This turns on an alarm if the water temperature goes above 96˚C. A much better system is a sender with 2 connections, the second one is for a water temperature gauge. With the gauge, one can actually see the thermostat in action. When the engine first starts, the gauge will show the temp of the water sitting in the engine (should be near the ambient temp in the engine space). As the engine warms up, the temp will rise to about 96˚C, at which point the thermostat begins to open (if the thermostat is working properly). The temp on the gauge will then fall until it reaches the proper operating temperature of the engine which is approximately 70˚C.

If you have a gauge, and its not showing this sequence, then you have a problem.
If the temp stays low and never rises, then the thermostat stuck open and/or the bypass hose is blocked; either way too much water is flowing through the block and the engine is running cold (not good for the engine in the long run).

If the temp rises and stays high, either the thermostat is stuck closed, not allowing water in the block to flow out, and/or the steam pipe is blocked which has the same effect, and or a bad impeller in the water pump (or the whole engine is full of salt deposits which usually means that the engine hasn’t been flushed with acid at regular intervals). If water can’t circulate through the block and out, the engine overheats.

Overheating can be caused by situations other than cooling water system problems: injector problems, fuel pump timing problems, etc, but mostly its cooling water system problems.

[Bill Seal]

Then you have the wierd (or not) problem I had on my fresh water cooled Volvo MD11C. It seems some mental midget designed the heat exchanger with the raw water going in the top and out the bottom. If, for any reason, air gets into the raw water side, be it a worn pump or heavy chop or whatever, it causes a bubble (Three Mile Island ) which can't get out, so obviously the tubes have no raw water circulating to cool them, and the engine intermittently overheated. Luckily, Volvo used standard rigid copper tubing (Home Depot stuff) so it was simple to make up new plumbing with sweat soldered elbows, and run the water the right way.