Need advice on diagnosis - water into oil - Concluded

Hi all,
To say I’m pissed of is an understatement. I have recently installed my rebuilt 3.8 into my XK, bolted all the ancillaries into place and filled with oil; no dramas.

Started filling the radiator, and once full I saw the level in the neck slowly drop down…
Ok, there’s probably airpockets etc and it’s just finding its level I thought.
However over time I put approx 5 more litres of water in.
No leaks under the car, so I checked the dipstick; a bit high and clean oil. Another 1/2 bucket and more water vanished and the oil level rose…:disappointed_relieved::disappointed_relieved:
So I’m hoping for a poorly sealing head gasket rather than a crack in the oil/water separation walls in the block.

My next step is to take the head off and see what I see…

I plan to drain oil, and then add water into one of the water passages in the top of the block and see if it holds a level…

Can anyone suggest other tests or observations to watch for?

There is no real way water from the head gasket can get into the oil.
Except in the front where the chains are, or through the cylinders.
Remove the camshaft covers, check if you see some leaking going on there.
Remove all spark plugs, to see if there is water in the cylinders.
Good luck.

1 Like

Broken or cracked timing cover where the water pump is, allowing water to drain into the block.?
You can check this without taking off the head.


Thanks guys both of those suggestions I’ll try this weekend . I’ll start with radiator out and water pump off so I can see if the timing cover is the culprit… that’s a lot less hassle than head off.

Agree with those suggestions: for the water to get in the oil that quickly suggests a leak in the water pump passage or a cracked front timing cover.

1 Like

Friday’s a public holiday here (Go Pies!!!) so it gave me opportunity to work this out.
Radiator Out, water pump off and I think I have found it…

This is the intermediate timing chain sprocket shaft. Could this be the path for the water?

Thanks to Peter, Paul and Wardell (sorry, I don’t know your first name) in your suggestion as to where to start…

maybe apply low psi compressed air to the suspect area, and listen for hissing ?


What does the back mating surface of the water pump look like in that area?

Also, is that a shadow from your screwdriver, or a discolored area on the timing chain cover?

Hi Jerry,
Yeah, that’s the one area that my theory falls flat… The circular area was covered by the water pump gasket, and appeared to be undamaged.
I just took a wire brush to the water pump backing plate, and although slightly pitted in a couple of areas it seemed intact That was a screwdriver shadow.

When I took the pump off, water came out with it, but not a heap (I had drained and removed the radiator, but not the block) The water in the block is at level in the right hand port of the cover, and holding level. This makes me think the crossover to the oil is at least higher than the bottom of the R.H. port.

I put a boroscope down the spark plug holes, and all cylinders are clean and no sign of water…

Jon, looks like limited sealing contact width around areas of that round hole/opening.

Have you checked the mating surfaces with a square edge straight edge to check for high and low areas?

Both surfaces may look good, but if not flat and in-plane, a gasket won’t seal properly.

Hi Jerry,

Good observations. the shaft is recessed approx 2mm or so. This means the gasket was only grabbing around the edge. I dont know if this is to allow for float of the shaft?

The gasket at this point was free of sealant and was ‘dished’ like a welsh plug as it took up space in the recess.

i could fab up a disk to fill this gap and use sealant to hold it in place. I had started to pull the cover off (harmonic balance was off, and I had just ground down a couple of nuts to use to extract the studs that hold the head to the cover.

Instead I will fab that disk up, button it all back together and give it another drink of water…
Cheers, Jon.

(oh, and that grey muck around the end of the shaft is aluminium powder from my wire brushing, mixed with water… A good clean is needed before I go any further.

There is no need for that: what is needed is a thin smear of RTV on both sides of the gasket. Do that, then install the water pump with the bolts just finger tight. Let it sit overnight, then bring them up to torque value. That’s more than adequate to stop any seepage that might have occurred along that area.


Just seems to me a small leak in that area won’t cause it to leak that quick, the sprocket shaft is a pretty snug fit in that hole. Refitting the water pump should confirm. Can’t think of any other source, a pressure test should show it up, good luck.

The rather quick leaking suggests a more direkt passage than that gasket. Have you double-checked all the hose connections?

An extra 5 L would cause the oil level to rise dramatically. Are you sure there’s water in the oil? Cold oil can take a while to fully drain down to the sump. Crack the sump drain plug, water should come out it it’s in the bottom there.


Hi All,
Sorry for the slow reply; I shot down to the store to get some RTV.
Randall - Definitely water in the oil. It didnt emulsify and I was able to actually drain off the water first, and then the oil separately.
When we say quickly, I left it for 8-10 hours whilst I was at work, and when I got home the water had drained and the oil level had risen.

I agree that the socket is snug, but it is also hollow, so the water would simply pass through the centre of it.

To be honest I cannot think of another area, unless it is a damaged passage inside the block. I have never run this engine; I got the bottom end fully rebuilt by a reputable shop and I installed the chains and head.

I certainly hope for your sake that the source of the leak is there. The only other place is beween the front cover & the block at the water channel.

Thanks Randall; I hope so too :crossed_fingers:

Water pump in place and finger-tight. Will tighten properly in about 1 hour as per instructions on tube

Did you put a straight edge across the faces of the cover and the pump? I just went out and had a look at my block and there would have to be a real mess in your block for corrosion to make a hole from the water passages to the sump. Let me know how you go.
If you have an air compressor put an air gun in the radiator fill and just lightly seal it with you hand and raise the pressure in the radiator and listen, if it is leaking a fair bit under gravity then you may hear it.

If it keeps happening you could pop off the sump and look for where it is coming in as well.

1 Like