M25XP Overheating at wits end

[Kevin Henderson]

I have been thoroughly enjoying a warm summer and warm fall here in SOCAL.  A couple weeks ago, while coming back from a wonderful sail and approaching sunset, I had a sudden overheat issue.  I managed to limp back into port and observed the following symptoms during the overheat sail:
1.  Coolant steaming (later found that the thermostat butterfly valve had let some coolant out.
2.  I had plenty of raw water flow so no obstruction to the impeller.

After arriving back into the slip I have been steadily troubleshooting various causes for the overheating and I think I've eliminated the usual suspects.  Thus far I have verified and/or changed the following:
1.  Raw water strainer clear and no obstruction
2.  Removed and opened up HX.  Verified that all lines are clear.  (just cleaned it out in the spring)
3.  Changed out my oil after the overheat
4.  Flushed and changed and burped the coolant.
5.  changed hoses to the HX (Raw water and coolant)
6. Burped the coolant line to the hot water heater (per the tech wiki instructions
7.  Tested and verified that the thermostat is working properly. 

During the overheat the onset was sudden.  Although there was loss of coolant it was due to the butterfly valve on top of the thermostat housing and has since not leaked.  I have opened up the top of the coolant tank and watched for flow from the coolant pump after the thermostat is supposed to open up but no observable flow. 
I'm kind of at my wits end here.  The coolant pump has shown no signs of failure or weeping or making any noises but at this point it appears to be the last possible point of failure. 
As stated earlier, the onset of the failure was sudden.  No prior indications or noises.  It seems odd that the coolant pump would fail since I only have 1100 hours on the engine. 
Any ideas?  and...Although I have become very proficient in performing minor repairs and preventative maintenance, a water pump is making me pretty nervous. 
I would like to hear thoughts about possible causes of the overheating and any information about changing out the coolant pump (procedure tips and models number)

Anyway... Help!   I'm about to call in the engine guy and admit defeat.   

[KWKloeber]

Kevin,

You said you had cleaned the Hx.  Was it pressure tested (for a leak between the sea water and coolant sides)?

On the 1st overheat, were you low on coolant?  On subsequent overheats? 

Once you refill/burp the system, do you steadily run lower and lower on coolant?  If so that points to a hole in the Hx (or another coolant leak.)

If there's a pinhole leak in the Hx, coolant will be pushed to the sea water side and out the exhaust -- you may or may not see steam and have the sweet smell of engine coolant.  You can test the Hx by pinching off and disconnecting the coolant hoses, putting (low) water or air pressure to the seawater inlet and plugging off the the sea water outlet.  i water exits the coolant connections yo have a leak.

If you loosen the pressure cap, that will help to lessen the coolant loss - no pressure builds up in the engine, so coolant doesn't get pushed thru the pinhole to the seawater side of the Hx. 

You can also get a test pump that fits in place of the pressure cap to pressure test the coolant system.

With the year and in salt, I would suspect the Hx has a pinhole, regardless of the hours.  it wouldn't be unusual -- though there could be other causes.  My Hx developed a leak (100% fresh water C30) at year ~20. OEM was probably a standard and not a Copper-Nickle Hx.

Do you have a coolant overflow/reservoir tank or does the overflow go to the bilge?  Does that tank empty as the engine cools down (coolant gets sucked back past the pressure cap)?  If not, then it points to the vacuum being relieved at another place (Hx or other leak.)

Do you have a water heater?  A leak in the coolant coil could also loose coolant to the potable water tank, but it's rare because the potable water side is typically at a higher pressure than the coolant side (~13-15 psi.)

The coolant pump impeller could go bad - loosen the belt and check for play or noise (bad bearing) turning the pulley/shaft.  I can get you the Kubota pump p/n number if that's the problem.  More likely it's an obstruction in the Hx (impeller blades), an air lock, etc -- some of which you have already checked.

A bad exhaust manifold flange gasket can also be the problem (allows coolant to leak from the exhaust manifold into the exhaust riser/muffler, and out the exhaust.)

Let me know what you find out about the Hx, or a pressure test of the system, which is the easiest next step.

Cheers,
Ken K

[Stu Jackson]

Kevin, after doing all that work and checkouts, have you taken the boat back out and run at cruising speed?  How have you assured yourself that what you've done isn't working?

[Kevin Henderson]

I have run the engine up to 2500 RPM and it will quickly reach +200F before shutting down.
I have run the engine to get it hot enough for the thermostat to open up and allow the coolant pump to circulate coolant.  However, I never see what appears to be a flow of coolant in the manifold when the pump should be circulating coolant. 

[KWKloeber]

I have run the engine up to 2500 RPM and it will quickly reach +200F before shutting down.
I have run the engine to get it hot enough for the thermostat to open up and allow the coolant pump to circulate coolant.  However, I never see what appears to be a flow of coolant in the manifold when the pump should be circulating coolant. 

And you said that you pulled and verified that the Tstat is opening at ~160F?

in my experience, when I rev the engine, the full coolant will draw down in the exhaust manifold, and well up (and may overflow) when the rpm drops.  I don't know that I have actually seen a 'flow" per se.

To verify that the coolant pump is ok, you could check for flow in the:

Water heater lines (should be flow with Tstat closed or open.)
Return from the Hx to the coolant pump (Tstat open only.)
Hose from the Tstat housing to the exhaust manifold inlet elbow (T stat open only.)

I attached the flow schematic, if that helps.

[attachimg=#]

Ken

[2ndwish]

Kevin- Great advice from Ken. If there is any uncertainty about the thermostat- remove it from the system. That will guarantee flow through the housing if the pump is working (you can replace it later with a new unit if that was the problem). This may or may not help, but simplifying the system when diagnosing a problem can be useful for ruling out components. You can do the same with the water heater- simply replace the connections at the engine with a short section of hose, if the problem goes away, the trouble is in that loop. How much coolant did you lose? Where did it end up (the bilge I presume)?   

[Ken Juul]

Could also be a indicator problem.  Have you verified that the engine is really getting that hot? 

[KWKloeber]

Kevin- Great advice from Ken. If there is any uncertainty about the thermostat- remove it from the system. That will guarantee flow through the housing if the pump is working (you can replace it later with a new unit if that was the problem). This may or may not help, but simplifying the system when diagnosing a problem can be useful for ruling out components. You can do the same with the water heater- simply replace the connections at the engine with a short section of hose, if the problem goes away, the trouble is in that loop. How much coolant did you lose? Where did it end up (the bilge I presume)?   

Great idea to eliminate the Tshat completely!

Let me clarify about the WH though.  That loop doesn't affect the engine cooling much, especially if the water is already hot.   
That loop allows coolant to circulate thru the block (to prevent hot spots,) as well as the holes in the Tstat flange (that lets coolant circulate thru the exhaust manifold) while the Tstat is closed.  The WH hoses take the place of the water-pump to thermostat-bypass hose (indicated on the schematic) that are present on the Kubota tractors and the M25s.  That hose was eliminated with the C34/XPs.  It's just an easy way to check for coolant flow produced bu the pump.  After the Tstat opens, the WH loop could be pinched off and it won't affect engine temp.  Before the Tstat opens, of course, pinching it will could create disastrous hot spots in the engine block.

I presume she's truly overheating due to the 'steam' and coolant loss.  But JIC, I uploaded the troubleshooting steps for the sender/gauge to the TechWiki on this page:
http://www.c34.org/wiki/index.php?title=Tachometer_Malfunction_Q%26A

Ken K




Ken K

[Ron Hill]

Kevin : As Ken Juul said - confirm that the engine is really getting that hot!!  Take a temperature gun and get a reading!

Also if the engine is only "overheating" while the steps are in you may have the metallic insolation (inside the steps lid) is touching the temp wire connection!  If so, you can easily fix that with a wire boot.

A thought

[karista]

You have received a lot of good advice. I had a similar occurrence, that when I turned the engine on it quickly overheated to 200 plus. After doing a lot of checking as you did, it turned out to be a slipping belt. I would check your belt tension and also your fresh water pump, mine failed after only 600 hours. It has a pressed on metal impeller that was loose. So check your belt tension and then also remove your fresh water pump and make sure the impeller is ok.

[KWKloeber]

Kevin,

If you do need a new coolant pump, the Kubota p/n is 15534-73030 about 90 boat bucks, vs the 400-plus-boat-buck Westerbeke pump.  The gasket is p/n 15676-73430.

Ken K

[Kevin Henderson]

As always, I can count on getting a lot of great advice in this forum. 
I was thinking of calling in a mechanic, but seeing some of the responses here, I feel like I still have a lot options that I can do myself to isolate and find the true fault.  My work list for this weekend is growing and some of the things I'm planning to do is:
- verify the engine is actually getting as hot as the sender is indicating
- remove the thermostat and let the pump flow coolant regardless of temperature
- Verify belt tension  (I have a new belt at the ready and just got the "third hand" tensioner to make the job a bit easier)
- brace myself for learning how difficult (or easy) it is to swap out the fresh water coolant pump.    

[KWKloeber]

As always, I can count on getting a lot of great advice in this forum. 
I was thinking of calling in a mechanic, but seeing some of the responses here, I feel like I still have a lot options that I can do myself to isolate and find the true fault.  My work list for this weekend is growing and some of the things I'm planning to do is:
- verify the engine is actually getting as hot as the sender is indicating
- remove the thermostat and let the pump flow coolant regardless of temperature
- Verify belt tension  (I have a new belt at the ready and just got the "third hand" tensioner to make the job a bit easier)
- brace myself for learning how difficult (or easy) it is to swap out the fresh water coolant pump.    

Kevin,

The pump is actually relatively easy to replace.  The studs (4) are captive and 4 nuts up front.
You may have a harder time getting the Tsat cap off due to bolt corrosion.  While you have it off, consider drilling and tapping the stbd side to accept the upgrade high-temperature alarm switch (1/8" NPT thread.)

RE: Slippage, what belt are you replacing it with?  I recommend the Dayco Top Cog Gold Heavy Duty --- it's the best beot for heat dissipation and no slippage.

Ken K

[KWKloeber]

Oh also,  make sure that when you replace the Tstat cap that you maintain a good ground path --

if not the gauge will read "off."  With a poor ground path the gauge will read low. 

I have thought about ways to keep the ground intact -- using copper-based never seize on the bolt threads, coupled with a serrated locking washer under the bolt head to get a good "bite" on the Tstat cap, and making sure the bolt head/cap is painted over to prevent corrosion from getting a foothold. 

Alternately, using a stout ring terminal under the head and grounding the cap to a good engine ground point.  I am also trying to come up with some kind of clamp that will fit the temp sender (and new temp switch) and ground them directly.

Ken

[KWKloeber]

Kevin,

A follow up on some diagnosis before pulling parts.

I don't recall if you had indicated whether:

(1) low coolant (ie coolant loss) caused the overheating, OR
(2) overheating caused coolant loss (ie coolant being pushed put the overflow.)

(1) indicates a different set of problems than (2).

If the system is full, burped, etc and if

(2) happens, you SHOULD SEE coolant coming out the overflow hose as the engine overheat. Skip to (2) below.

(1) happens, you SHOULD'NT SEE a lot of coolant in the overflow reservoir (or in the bilge if you don't have a tank) as the engine overheats.

(1) points to a leak in the closed coolant system -- so pressure test it to confirm.
Autozone may loan out a radiator pressure tester.  (The website lists the pressure cap adapters, but not the tester -- it doesn't make sense to loan one and not the other.) Harbor Freight also has a tester kit ($95). 
But you can test it with a bicycle pump or 12v compressor -- make up your own fitting with pipe -- a 1/4" pipe tee, 1/4" x 1/8" reducing bushing, 1/8" nipple, and get a (1/8") Schrader valve (ie, tire valve from autozone), and a 1/4" pressure gauge (low pressure, like 0-30 or 0-50 psi.) 
Install the assembly in place of the bleed valve and pressurize the system to 5-10 psi -- no more, or it will just blow by the pressure cap.  (But that's also a way to test the cap to make sure it does release pressure at 12-15 psi.)

If the pressure DOESN'T hold, you have a leak -- in the Hx, the water heater coil, the exhaust flange gasket, or some other obscure place.  Pressure test the Hx (can do on the engine), test the WH coll (again can do in-place) and if both pass, the exhaust gasket is likely leaking.

If pressure DOES HOLD, it rules out a leak in the (cold) system.  However if you are pretty sure that you are loosing coolant before she overheats, there could be something obscure -- like a crack that opens or a gasket leaking only when the block temp rises.  So, run the engine until the temp comes up and you should not see the pressure rise (or it will rise and drop off, but nothing goes out past the overflow) because the pressure (and coolant) is leaking somewhere.  This is pretty obscure, so let's get into that later if that's the result!!

If (2) happens,

It points to engine coolant not circulating, or it not being cooled by the heat exchanger -- with the cause being:

(a) sea water not circulating (in the order of likelihood - blocked thru hull, bad impeller, blocked pump outlet, blocked Hx tubes on the sea water side, blockage in vented loop or hose to muffler.) OR
(b) coolant not circulating (in this order of likelihood - blocked Hx tubes on the coolant side, bad Tstat, blocked hose somewhere, bad coolant pump/impeller.)

Some quick thoughts, may add more later.

Ken K