Logo

Mercruiser 470 engine consumes coolant

NormW

New member
My Mercruiser 470 engine consumes a large amount of coolant. The last time that I ran the boat in the water it ran fine. It started quickly and the engine ran smoothly with normal power. The oil pressure and engine temperature were normal. There was no sign of engine overheating and the engine did not make any unusual sounds. No coolant was dripping into the bilge.


I discovered the problem when I drained the coolant since it was time to replace it. Prior to draining it, I noted that the coolant level in the recovery tank was at the “full” line so I expected to collect 10.5 quarts of coolant since the cooling system has a capacity of 10.5 quarts. But I collected only 7 quarts of coolant which surprised me since if coolant was being lost within the engine I expected that replacement coolant would be drawn from the recovery tank. This indicated to me that there was a significant leak in the coolant system both due to the loss of 3.5 quarts of coolant and the lack of suction to draw replacement coolant from the coolant recovery tank.

After draining the coolant, I filled the coolant system with water in order to run the engine and flush out the cooling system. It took 10.5 quarts of water to fill the system which is the coolant system capacity. I ran the engine for about 15 minutes at 2,000 rpm. The engine ran smoothly.

After the engine cooled I drained the cooling system and collected only 7 quarts of water. So I lost 3.5 quarts of water in just 15 minutes of engine running. There is a big leak somewhere internally. Nothing is dripping into the bilge.. I drained the oil to see if the oil looked “milky” but I could not tell. The oil was black as it always is when I drain it at the end of the boating season. I got approximately 5 quarts of oil when I drained it plus there was still some oil in the oil filter. The capacity is 5.5 quarts of oil so none of the missing coolant wound up in the oil.


I refilled the cooling system with water and did a pressure test. I pressurized the system to 10 psi. and the system held at that pressure. No sign of any leakage.


The only time that the coolant or water is lost is when the engine is running and hot.


Based on the tests described above, my non-expert opinion is that the head gasket is OK since:


1 There is no coolant in the engine oil.
2. The engine runs smoothly when passing large amounts of coolant. The engine consumed 3 quarts of water in just 15 minutes. It would probably not run at all with that amount of water being ingested. Also, since no coolant is winding up in the bilge, the loss of coolant must be through the exhaust.


I suspected that a leak in the heat exchanger was the cause of the large coolant loss since coolant only flows through it when the engine is running and hot. When the engine is cold, like when I did my pressure test, the thermostat bypasses the heat exchanger. I removed the heat exchanger from the engine and did a pressure test on the coolant leg using hot water at 75 psi. It held that pressure. No leaks.


What else could be causing my coolant loss problem?
 
What year is the engine/boat?

I would suggest going to a discount auto parts store and purchasing a radiator pressure checking kit. It comes with a hand pump, gauge and assorted fittings.

With the system drained, attach pump and adapter to radiator cap location on tank and pump to 17 ish PSI. Listen for leaks.
You can do it with antifreeze in system just make sure level is 2-3 inches below full, Same PSI 17 ish


470 coolant flow.jpg

If the antifreeze is coming out into exhaust it could be the exhaust manifold.

the infamous 470 has a very long history of the exhaust manifold leaking. Only a pressure test can reveal.
 
I bought the boat new in December 1979 so it is 42 years old.

As for pressure testing, I got a loaner tester from AutoZone. I filled the cooling system about half full with water and pressurized it to 10 psi. After 1/2 hour it was still at 10 psi. An hour after that it was down to 9 psi. I did not hear any hissing nor did I see any dripping in the bilge. I probably should have gone higher in pressure.

I thought that the manifold could be the cause of my problem but there would have to be a crack in the metal between the hot exhaust side and the coolant side. It would have to close tight enough when cold to hold the pressure in my above mentioned test. And it would have to open wide enough to pass 3 quarts of coolant when hot in my 15 minute engine run test.
 
I bought the boat new in December 1979 so it is 42 years old.

As for pressure testing, I got a loaner tester from AutoZone. I filled the cooling system about half full with water and pressurized it to 10 psi. After 1/2 hour it was still at 10 psi. An hour after that it was down to 9 psi. I did not hear any hissing nor did I see any dripping in the bilge. I probably should have gone higher in pressure.

I thought that the manifold could be the cause of my problem but there would have to be a crack in the metal between the hot exhaust side and the coolant side. It would have to close tight enough when cold to hold the pressure in my above mentioned test. And it would have to open wide enough to pass 3 quarts of coolant when hot in my 15 minute engine run test.

I believe the cap pressure is 13-15 psi. To test you must go higher. And having engine at running temp as suggested is a good idea as well.

""With the system drained, attach pump and adapter to radiator cap location on tank and pump to 17 ish PSI. Listen for leaks.
You can do it with antifreeze in system just make sure level is 2-3 inches below full, Same PSI 17 ish""
 
Last edited:
I attempted to pressurize the coolant system. Unfortunately the boat is in an unheated garage and the engine has been drained of all coolant, water, and oil. I borrowed a radiator pressure testing kit from AutoZone. It comes with 18 different radiator caps, none of which fit the cap on the 470 engine. The kit also has three rubber conical shaped caps one of which seemed to fit OK. I painted the area about the cap with soap water to see if any bubbles appeared, but none did so that cap must be sealing OK. Despite vigorous pumping the pressure gauge came slightly off zero and then went back to zero quickly when I stopped pumping. Obviously there must be a big hole somewhere in the cooling system. I suspect that the hole is in the exhaust manifold. I plan to remove the exhaust manifold and test it separately. I welcome any other ideas.
 
FTI

Radiator cap and reservoir tank fitting would be the same as any 1960's-1980's automotive standard type radiator and cap!

Attached here are two cooling system schematics. They are basically the same with the exception of slightly different parts. BUT the water flow both RAW and ANTIFREEZE paths are still the same with the exception of the exhaust ELBOW and Reservoir tank. Older model has separate reservoir tank where as the newer model, the reservoir tank and exhaust elbow are one assembly.

The DARK arrows are ANTIFREEZE
The WHITE arrows are RAW/SEA water.

Old system is on right side of picture
Newer system is on the left side of picture (reservoir tank and exhaust elbow in one assembly)


Old system with separate reservoir tank,
The exhaust ELBOW gasket could be a point of leakage as it is a BLOCK OFF gasket and if rotted or pin hole in it antifreeze from exhaust manifold could leak through (slow leak)

Newer system with all in one ELBOW (Reservoir tank and exhaust elbow) This part could be the issue as the casting may have corroded/warn away and is now leaking between antifreeze zone and raw water zone. Other possible issue is the same as old system and that would be heat exchanger.
The exhaust ELBOW gasket could also be a point of leakage as it is a BLOCK OFF gasket and if rotted or pin hole in it antifreeze from exhaust manifold could leak through (slow leak)

On both systems, the exhaust manifold has only antifreeze running in and out of it so if the casting has cracked/worn away the antifreeze could be entering the HOT exhaust from cylinders and being burnt off.

On both, the heat exchanger can also be the source of the leak as both raw water and antifreeze go through the assembly. This is rare but this can also be pressure checked.

It is best to remove ELBOW and Reservoir from exhaust manifold old system and pressure check all three if you can figure out how. (A lot to explain here)
Do the same for newer system but its only the Reservoir/Elbow and exhaust manifold.

When doing a pressure test you pressurize the water/antifreeze side only. so pressure in at one point, all other water/antifreeze holes must be sealed off real good.
If there is a leak it will be heard in exhaust gas flow areas (exhaust ports on exhaust manifold and exhaust gas exit to elbow) (for elbow, the exhaust gas flow area)

Make a fitting to use a compressed air connection (quick disconnect with ball valve in line to shut off air supply to test port).

If you are unable to make the fixtures to do testing, check with a radiator shop to see if they have or are able to do so. Or a marine business if you have one local

470.jpg
 
Last edited:
Per your suggestion, I began to remove engine components. I have the older model 470 engine with a separate Exhaust Elbow and Reservoir Tank. I started with the Exhaust Elbow. It did not come off easy but I finally got it off. The gaskets were damaged during removal so I could not tell if they were damaged while in place. The gasket material is very hard and almost impossible to scrape off. I will have to use a chemical to soften the gaskets.


The elbow itself looked to be in great shape despite its age. But I became perplexed as to how it works. The discharge sea water exits the heat exchanger and enters the exhaust elbow through a fitting on the side. I ran water into this fitting and it exited not into the exhaust gas stream but exited from the three slotted openings in the flange where the elbow attaches to the exhaust manifold. But at this attachment point are the block off gaskets and plate which appear to prevent sea water from entering the exhaust manifold and antifreeze from entering the exhaust elbow. I do not know the path that the sea water takes to exit from the elbow. Please enlighten me.


As for the reservoir tank, the four bolts that hold it on are stuck solid. I was fearful of breaking them off if I applied too much force so I decided to keep the reservoir attached to the manifold for now.


I plan to remove the manifold next and test it with the reservoir attached. First I need to block the opening where the exhaust elbow attached to the manifold Then I plan to run hot water (130F) into the thermostat housing which should exit from the fitting on the manifold end cap. Then after the manifold is up to temperature, I will cap off the end cap. My water pressure is 75 psi so that should be a good leak test of the manifold. The reservoir cap should vent at 14 psi but I can block the vent and let the pressure go up to 75 psi. Since I have already removed the heat exchanger, I made a bypass using hoses and PVC pipe to simulate the heat exchanger flow path.


I have already tested the heat exchanger using a similar method. I ran hot water through the coolant side to bring it up to 130F and then capped off the discharge end. The were no leaks into the sea water side with 130F water at 75 psi on the antifreeze side..
 
""The elbow itself looked to be in great shape despite its age. But I became perplexed as to how it works. The discharge sea water exits the heat exchanger and enters the exhaust elbow through a fitting on the side. I ran water into this fitting and it exited not into the exhaust gas stream but exited from the three slotted openings in the flange where the elbow attaches to the exhaust manifold. But at this attachment point are the block off gaskets and plate which appear to prevent sea water from entering the exhaust manifold and antifreeze from entering the exhaust elbow. I do not know the path that the sea water takes to exit from the elbow. Please enlighten me.""


Elbows are the same design basically for closed cooling or raw water cooling. The difference as you point out is the gasket used at exhaust manifold.
Closed cooling uses a BLOCK OFF gasket system so the antifreeze cannot exit the elbow and stays in exhaust manifold and circulates in engine.

The raw water connection as you describe puts raw water into elbow and with the block off plate/gasket all raw water exits out the other end of elbow which connects to large rubber reinforced exhaust hose.

In a RAW water cooled system, the raw water travels thru the exhaust manifold, through the gasket at elbow (gasket has open holes/passages) and up through the elbow and out the exhaust side. There may be a cold raw water connection as yours has also to make sure cold raw water flows out exhaust to cool hoses at all times when engine is running.

If you have determined the Heat exchanger is good, the elbow is good (rare the elbow has any issues causing your problem) there are only FOUR possible causes left.

1. Pin hole leak in head gasket.............You would find antifreeze in oil or one cylinder would be super clean and spark plug would be also super clean as the antifreeze would remove all carbon build up.

2. Exhaust manifold has a pin hole leak between antifreeze chamber and exhaust chamber and antifreeze is being exhausted out the exhaust. (you may notice excess steam coming out of exhaust even though engine is at proper temp).

3. Radiator cap has a slow leak allowing antifreeze to leak when pressure increases. You would see antifreeze trails running down sides of reservoir.

4. Block off gasket at elbow had a pin hole leak. I believe the set up is, gasket/stainless steel gasket/gasket sandwiched. So the stainless steel gasket would be the one having the pin hole. If you cannot see it, replacing it is your only choice.
 
I have done a lot of work but did not make much progress in determining if my exhaust manifold leaks.


First I removed the exhaust elbow from the manifold and then I removed the manifold from the engine keeping the End Cap, Thermostat Housing, and Reservoir attached to the manifold.


I made up a connection between my laundry tub faucet and the thermostat housing using hoses and PVC pipe and fittings. I then made using more hoses and PVC pipe and fittings a replacement for the heat exchanger.


And I removed the drain plug on the manifold and replaced it with a pressure gauge.


My big problem was sealing off the flange where the exhaust elbow attaches to the manifold. I made a gasket from 1/16 inch thick rubber and made an aluminum cover plate to cover the opening.


I ran hot water at 130F through the system until it was hot to the touch. The outlet on the end cap was open and the pressure gauge did not show any pressure as was expected. No leaks were apparent. Then I closed off some of the opening on the discharge end and the gasket leaked both externally and internally. The pressure gauge remained at zero which surprised me since the leaks were not drips but forceful streams of hot water.


I then replaced the aluminum cover plate with the exhaust elbow and got the same result.


I then bought some gasket material from NAPA and before I installed it, I coated the flange on the manifold with Permatex. I covered the gasket with the blocking plate and installed the exhaust elbow.


It too leaked both externally and internally but I believe that the pressure was higher than previously before the leak started. Still the pressure gauge read zero which surprised me.


I tested the pressure gauge by connecting it directly to the faucet and opened the faucet wide and it read 75 psi so it works OK. In my test I expected the water to fill up the reservoir right to the top, and if the pressure exceeded 14 psi, which is the cap rating, I expected to see a discharge down the hose. There was no discharge and the reservoir remained dry.




As a result of all this testing, I still cannot conclude if the manifold is sound or has internal leaks.


I thought of another way to test the manifold. I have an air compressor and could apply compressed air to the drain plug opening. I would position the manifold with the exhaust ports facing up and fill the manifold with water right up to the top of the manifold ports and look for bubbles.


Another thing that I noticed is that the flange on the manifold where the elbow attaches is not totally smooth even after extensive cleaning. There are a few bumps on the thin strip between the one coolant channel and the central exhaust duct. Perhaps this is the source of the coolant leak. If I could remove the 4 studs, I could get that flange machined down to a real flat surface.
 
1. Why did you use water to test with? Was not compressed air specifically mentioned?

2. Did you measure the water pressure from your supply first to understand just how much pressure it is making?

3. It is obvious you are not seeing how all the parts work so you are not understanding how to properly test.

take a step back, understand the water flow and the exhaust gas flow. Understand how block off water flow in order to allow pressure to be applied.

As mentioned, it may be necessary to make sure the parts are warmed up to expose a possible crack which my not be leaking when cold.

You said you used a gasket that was 1/16" thick...............if you can make a gasket make it at least 1/8 inch thick....................a 1/16" gasket may not be very effective in this scenario.

Also pictures of your test set up would be helpful so we can determine if what you are doing is correct.


I will also say, you did not do a proper pressure check before you took it all apart. It was suggested to use a radiator test kit to check and you simply said you could not make it work...........well It works and I have done many. There is also a check for the cap itself, Did you check that?

One other question, when running did you see any excessive steam coming from exhaust?
 
You've spent more in time and energy to diagnose a problem than you would have spent by buying a new manifold and riser. However, my gut feeling is that you're beating a dead horse, or rather in this case, a dead Mercruiser 470.
In all likelihood, you have a bad head. Just a guess tho. Are any of the spark plugs "steam cleaned"?
 
Last edited:
Just a quick update on my testing, this time using compressed air. I closed off the coolant input port as well as the output port using rubber hoses and PVC pipes and end caps. I also installed a bypass between the two connections to the heat exchanger. The interior of the exhaust manifold was totally dry. I turned on the air compressor and observed major leaks at the joint between the manifold and manifold elbow. The leak appeared to be mostly internally but there was one point externally that was leaking. I am using a NAPA 3075 gasket material that is intended for use in sealing water and coolant. It was the closest material that I could find to an OEM gasket. I also used some Permatex to enhance the seal.


I plan to order a pair of OEM gaskets and a backing plate and attach the manifold elbow to the manifold and see how it goes. I expect to still have leaks. The flange on the manifold has no rust but it does have some small bumps which must be the cause of the leaks. I will need to remove the 4 studs and find a machine shop that will reface the flange surface. The flange on the exhaust elbow looks OK.
IMG_1881.JPGIMG_1882.JPG
 
OK

You can remove the elbow completely and just cap off the manifold there. The Elbow does NOT get any antifreeze as there is two gaskets and a BLOCK OFF plate to keep the antifreeze in the manifold. So that is NOT a source for a possible leak.

This will allow you to make a plate/cap and get that leak stopped without using oem gaskets. Make a single thick gasket with 4 bolt holes and a thick flat plate to cap it off.

What pressure are you using? I am not sure I would go over 60 psi. I am thinking 30 psi may be enough. Maybe someone else can chime in and suggest what may tbe the right pressure to use.

You also may want to if you can, put the manifold and reservoir in the oven at 2-300 degrees for an hour or so before testing so if there is a crack it will expand like it would when engine is running.

Remember exhaust gas temps when exiting the head are around 1200 degrees so the inside of the manifold gets very hot!
 
It has been a while since I last posted my problem. As I mentioned previously, I have a coolant leak at the joint where the Exhaust Elbow joins the Manifold. It appears that the flanges became eroded and no longer made a seal. The leak became severe to where I lost a quart of coolant every five minutes of engine operation. The leak was entirely internal. There was no sign of an external leak at this joint. I took the Exhaust Manifold Assembly parts to a machine shop and had the flanges refaced. They came back very shiny, almost mirror like, and were very flat. I bought a complete set of gaskets for the entire exhaust manifold assembly plus a blocking plate that goes between the gaskets where the Exhaust Elbow attaches to the Manifold. I installed new studs since the original ones were slightly damaged when I removed them. Today I ran a pressure test by injecting compressed air into the coolant drain hole in the manifold. The joint still leaked. but the leak was external not internal as best I could tell. A jet of air came our of the end corner. I could not find any spec on the torque for the four nuts so I tightened them as fully as I could. I looked at leaking corner for any sign of a crack but there was none. I have no idea as to what I should do to correct this leak problem.
 
Back
Top