coolant in oil, no oil in coolant

[Dash01]

'90 535ia M30 engine gets coolant in oil, but no oil in coolant.

Head and block have no no apparent crack, headgasket looks OK, with no obvious streaking or other signs of leakage or seepage between oil and coolant holes. Cylinders and plugs are dry. Freeze plugs all OK, as is plastic coolant reservoir and plastic radiator tanks. Car runs fine and does not overheat, with temp gauge pegged at 12 o'clock position.

Put in new oil and filter, and within a few miles the oil starts to get frothy with coolant mixing in.

Head was rebuilt and resurfaced only recently, and placing a mechanic's straightedge along the head and also along the block indicates no warping of either.

Where else could coolant find its way into the oil?

[Ross]

Headgasket or a crack, that's it.

[whiskychaser]

I dont know the M30 but on the M50 you can have a perfect seal between the head and block and still get coolant in your oil. The leak can come from either a bad seal between the head and the timing chain cover or the timing chain cover (behind the water pump) and the block. No prizes for guessing how I now :(

[Barney Paull-Edwards]

Water in oil, no oil in water? Water at 2psi,oil at 40 psi,gotta be combustion pressure forcing oil into water but no overheating and normal on gauge? If it was a crack surely there would be pressurization of one side or another?

[Dash01]

How about various screw plugs, etc.? I did not mess with any of these, but rather am trying to understand how the coolant could get into the oil, but no oil in the coolant. Head, block and headgasket appear to be OK.

Anybody know more about the potential for a passageway or seepage around the water pump or thermostat housing into the oil? I've looked down the front chain cover, but cannot see any likely signs of seepage. Whiskeychaser, can you tell us more about how that happens on your engine, or your engine and the M30 535i engine compare in this respect?

[Dash01]

Water in oil, no oil in water? Water at 2psi,oil at 40 psi,gotta be combustion pressure forcing oil into water but no overheating and normal on gauge? If it was a crack surely there would be pressurization of one side or another?

No overheating, has normal coolant temperature, and no oil in coolant. Coolant in oil, instead. And, when I unscrew the cap on the coolant reservoir (on the firewall), it hisses briefly, telling me that it is still airtight there. I am losing coolant at a slow pace, but there is no sign of water in the cylinders or on the plugs or white clouds from the tailpipe, so I think the coolant is escaping and running down into the oil pan, where it is mixed and recirculated with the oil. Low oil pressure light flickers on shutdown once the oil becomes milky or froth appears on the dipstick or oil filler cap. (I assume that is because the oil/coolant mix has lower viscosity than just oil, and/or the frothy mix causes cavitation registered at the oil pressure switch.)

No sign of cracked head, block, or telltale seepage marks on headgasket, which suggests that is not the source of the trouble.

[ironie]

a leakdown probably would have provided your answer, at least in regards to which cylinder is forceing water into the oil. My bet would be the head gasket either way. or you have a true fluke and it is a casting defect. The leakdown might have pointed it out one way or another. I can't picture a crack between water and oil jackets without some freezing. Even a plain old compresion test probably would have been useful.

[Barney Paull-Edwards]

How about looking at only oil/water interface I can think of,tranny cooler in radiator or engine oilcooler? Some cars have cooler/heater in circuit to heat oil in cold climates,took mine out ages ago for the reason of potential contamination.

[Bin_jammin]

How about looking at only oil/water interface I can think of,tranny cooler in radiator or engine oilcooler? Some cars have cooler/heater in circuit to heat oil in cold climates,took mine out ages ago for the reason of potential contamination.


Mind if I throw my hat in the ring? I've seen this type of thing happen more often than you might think, and first of all let me just say DO NOT DRIVE THE CAR. DO NOT EVEN RUN THE CAR, until you figure it out. If engines were made to use coolant as a lube they would, you're going to do massive damage to your bearings running with coolant in your oil

Now, so far as the no oil in your coolant, that's an easier one, oil doesn't milk coolant up like coolant milks oil up. The oil may be sludged in a high point in your engine, but it is possible that you don't have any oil in your cooling system, sometimes you can have a leak that works one way but not the other.

Like a head bolt. I've seen them leak coolant plenty of times. Pull your valve cover and look for wetness around your head bolts. Did you re-use the old head bolts when you pulled the head? And please, for god's sake, tell me you put a new gasket on....

[Dash01]

Mind if I throw my hat in the ring? I've seen this type of thing happen more often than you might think, and first of all let me just say DO NOT DRIVE THE CAR. DO NOT EVEN RUN THE CAR, until you figure it out. If engines were made to use coolant as a lube they would, you're going to do massive damage to your bearings running with coolant in your oil

Now, so far as the no oil in your coolant, that's an easier one, oil doesn't milk coolant up like coolant milks oil up. The oil may be sludged in a high point in your engine, but it is possible that you don't have any oil in your cooling system, sometimes you can have a leak that works one way but not the other.

Like a head bolt. I've seen them leak coolant plenty of times. Pull your valve cover and look for wetness around your head bolts. Did you re-use the old head bolts when you pulled the head? And please, for god's sake, tell me you put a new gasket on....

Well, since oil floats on water and since coolant is mostly water, if there were oil in the coolant it would be at least somewhat discolored and there would be oil floating on top of it in the translucent bucket into which I drained the coolant. None. Looks normal, smells normal, indicating no oil in the coolant. Before removing the head, a sniffer test revealed little or no indication of exhaust gas in the coolant, which tells us the cylinders are not compromised. (Upon removal of the head, cylinders and plugs were completely dry. A mechanic's straitedge laid on the head and on the block at various angles showed no daylight, so presumably the head and block are not warped. If they were warped, telltake leakage streaks or discoloration would presumably show on the removed headgasket, but it looks OK.)

The oil drained from the pan, however, started with water coming out, being heavier than oil and therefore at the lowest point. Then came milky oil. Some frothy latte' stuff was on the bottom of the oil filler cap and on the dipstick, on the oil filter, etc..

Again, the headgasket shows no signs of leakage or seepage between oil and coolant holes, toward the drain galleries, etc.. It is a new Reinz headgasket installed with new bolts, torqued to spec. If it were the source of the compromise, there presumably would have been coolant in the cylinders or white smoke from the tailpipe, but there was not. All freeze plugs are intact, as are the plastic radiator and coolant reservoir parts, which hold pressure, with a brief hiss when the reservoir cap is removed. No external signs of leakage or seepage, either.

Somehow, coolant is getting into the oil, but apparently not via past the head gasket.

How does a head bolt leak coolant in an M30 engine?

Can anybody direct me to a good schematic that shows the passages for both oil and coolant in the M30 engine?

[The Bigfella]

Looking for daylight under a mechanic's straightedge is not the answer. you need to do this with feeler guages. By mechanic's straightedge, I hope you mean a proper 1/4" (or more) thick straightedge.

[Bin_jammin]

Well, since oil floats on water and since coolant is mostly water, if there were oil in the coolant it would be at least somewhat discolored and there would be oil floating on top of it in the translucent bucket into which I drained the coolant. None. Looks normal, smells normal, indicating no oil in the coolant. Before removing the head, a sniffer test revealed little or no indication of exhaust gas in the coolant, which tells us the cylinders are not compromised. (Upon removal of the head, cylinders and plugs were completely dry. A mechanic's straitedge laid on the head and on the block at various angles showed no daylight, so presumably the head and block are not warped. If they were warped, telltake leakage streaks or discoloration would presumably show on the removed headgasket, but it looks OK.)

The oil drained from the pan, however, started with water coming out, being heavier than oil and therefore at the lowest point. Then came milky oil. Some frothy latte' stuff was on the bottom of the oil filler cap and on the dipstick, on the oil filter, etc..

Again, the headgasket shows no signs of leakage or seepage between oil and coolant holes, toward the drain galleries, etc.. It is a new Reinz headgasket installed with new bolts, torqued to spec. If it were the source of the compromise, there presumably would have been coolant in the cylinders or white smoke from the tailpipe, but there was not. All freeze plugs are intact, as are the plastic radiator and coolant reservoir parts, which hold pressure, with a brief hiss when the reservoir cap is removed. No external signs of leakage or seepage, either.

Somehow, coolant is getting into the oil, but apparently not via past the head gasket.

How does a head bolt leak coolant in an M30 engine?

Can anybody direct me to a good schematic that shows the passages for both oil and coolant in the M30 engine?


First off, the water wouldn't be significantly discolored by oil in it, because while the oil passages are at times higher in pressure than the water passages, the VOLUME of oil that it takes to discolor water is much greater than the volume of water it takes to discolor oil. This may seem like an easy question, but you DID change the oil after changing the head gasket right?

Your old head gasket would be the best place to look for passages between the head and block. You're in luck, I happen to have my spare m30 with the head off down in the shop, so I just took a peek. The head bolts go into their own holes.

How it normally would work with a head bolt leaking oil is that the head bolt would thread into a coolant passage in the block, then when the block heats the coolant, the coolant expands and climbs the threads of the head bolt. I still wouldn't rule this out, you could have a crack in your block around the base of a head bolt thread. This would leak coolant in and no oil out.

One way to check for your leak (at least in the cam valley) is to use compressed air to pressurize your oil passages, then check for leaks. I'd recommend making a sealing cap to fit your oil filter canister. Also, do the same thing, pressurizing the coolant passages with a cap tester. If you have the tools, you may be able to find it using a vacuum setup.

One other thing, using a straightedge with daylight is a big no-no. You need a calibrated straightedge with feeler guages. You could be out of spec by .001 or less, and that would be enough to block light but still be warped. Maximum warpage allowed is .004. That's pretty damn thin, like less than a sheet of paper. What pattern did you follow when checking for warpage? Did you get the head pressure checked? Did you check the block for warpage?


I'm sorry to say it bud, but I think the only way you're going to figure this one out is to start tearing the motor apart. Even if you can figure it out without doing a tear-down, you're not likely to be able to fix it without one.

[Bin_jammin]

Looking for daylight under a mechanic's straightedge is not the answer. you need to do this with feeler guages. By mechanic's straightedge, I hope you mean a proper 1/4" (or more) thick straightedge.


Hah, I was downstairs checking my engine for my previous post in this thread when you type the same thing I was in the middle of typing. I'm glad to have seconded the emotion.

*** EDIT ***

Also, what kind of mechanic let you use his straightedge without telling you to use the feeler gauges? I hope like hell that's not how he's doing it, and if he is, have him talk to me, please. I'll explain it over the phone if I have to.

[Dash01]

Thanks, Bin, for the info and checking on your M30. While you have the head off, can you pls. tell me which of the 14 head bolt holes may be a pathway for coolant, or any other likely suspect sources?

I sure wish I could find a good schematic drawing that shows the internal pathways of oil and coolant in the block and in the head.

I will probably take the head to a proper machinist and have it pressure tested.

Hopefully, the source of the trouble is something simple and cheaply corrected, such as an installation mistake by me.

[Bin_jammin]

Thanks, Bin, for the info and checking on your M30. While you have the head off, can you pls. tell me which of the 14 head bolt holes may be a pathway for coolant, or any other likely suspect sources?

I sure wish I could find a good schematic drawing that shows the internal pathways of oil and coolant in the block and in the head.

I will probably take the head to a proper machinist and have it pressure tested.

Hopefully, the source of the trouble is something simple and cheaply corrected, such as an installation mistake by me.


No sweat, I had the motor on a shipping block and the head off already, no harm in checking it out :)

Basically both all the bolts and none of the bolt holes are suspect. What I mean is that you'd have to have a crack in the block in order to have coolant traveling through your head bolt passages. I'd go with the machine shop checking the head before anything else, make sure the head's not warped before anything else.

[whiskychaser]

Whiskeychaser, can you tell us more about how that happens on your engine, or your engine and the M30 535i engine compare in this respect?
This is what an M50 timing cover looks like:
http://www.realoem.com/bmw/showparts.do?model=HD68&mospid=47405&btnr=11_0157&hg=11&fg=10
On the M50 the head overhangs the block at the front and the timing chain cover sits under that. Coolant enters the cover through the rectangular hole in the top. The pump goes through the hole in the front and sits on studs (no. 2 in diagram). The back end of the timing cover where it mates with the block must be watertight, as does the join with the head. I got latte in my sump because I missed a tiny bit of old gasket. The head to block seal was fine but I redid it anyway

Just had a look at a 535i on real.oem. The above is not relevant

[Dash01]

As far as testing the M30 head (while off the car), it it safe to assume that ALL coolant should travel into and out of the head via the water pump and thermostat ports and the coolant holes on the lower side of the head?

If so, then I could do a vacuum test by sealing those holes with duct tape and applying vacuum, using WD40 and watching to see if and where it is sucked in.

Does this seem advisable?

[Bin_jammin]

As far as testing the M30 head (while off the car), it it safe to assume that ALL coolant should travel into and out of the head via the water pump and thermostat ports and the coolant holes on the lower side of the head?

If so, then I could do a vacuum test by sealing those holes with duct tape and applying vacuum, using WD40 and watching to see if and where it is sucked in.

Does this seem advisable?


If you've already got the head off the car, spend the $50 and have it checked by a pro. They do it for a living and will give you much better results than duct tape and wd40. I'd also inspect the block after a good cleaning looking for cracks in the deck.

[Dash01]

If you've already got the head off the car, spend the $50 and have it checked by a pro. They do it for a living and will give you much better results than duct tape and wd40. I'd also inspect the block after a good cleaning looking for cracks in the deck.

Well, Bin, I took your advice and had the best cylinder head guy in the county pressure test the head: Perfect, no leaks, and properly surfaced to within 1/2 of 0.001". He looked at the slightly used Reinz head gasket and found that to be just fine, too, with no signs or streaks indicating any compromise of the HG seal.

He says if there is coolant in the oil, but no oil in the coolant, the source has to be coolant gravity-draining down into the oil pan, because as if there were a leak somewhere allowing oil and coolant to come into direct contact, the higher oil pressure would send oil into the coolant and it would be obvious in the coolant reservoir, radiator, etc.. I drained all coolant into a translucent plastic bucket, and there is not even a sheen on the surface, no discoloration, and no hydrocarbons per sniffer test: There is no oil in the coolant.

So, the expert says to check the timing chain cover, which I'd removed to change the timing chain during the HG job a few months ago.

That's puzzling, since although the water pump is next to the timing chain cover, it seems to me that any water leaking from the water pump would just flow down the outside of the engine front and fall on the ground. Further, I used the narrow paper seals provided for the timing chain cover, and included Permatex gooey sealant on them.

Has any M30 engine had coolant get past the timing chain cover and into the pan?

If not this, then what sort of cracked block or whatever else could lead to coolant in the oil, given the circumstances previously noted?

[attack eagle]

Does the m30 use a water/oil cooler element? or anything of that nature?

BTW guys water pressure is higher than 2 psi... it's usually 12 psi or whatever the pressure cap is good for... And I have seen water in oil without any oil in water as well. Here is a thought, you have had the head checked now... but have you had the Block checked???

I have seen leaks in Multilayer headgaskets in between layers where they are not visible.
After cehcking the block, I'd probably by a new OEM HG, and new OEM bolts, and cross my fingers one more time that it worked out this time, before ripping the block and head out to be disassembled, tanked, and magafluxed to search for cracks. :(

[Dash01]

Does the m30 use a water/oil cooler element? or anything of that nature?

BTW guys water pressure is higher than 2 psi... it's usually 12 psi or whatever the pressure cap is good for... And I have seen water in oil without any oil in water as well. Here is a thought, you have had the head checked now... but have you had the Block checked???

I have seen leaks in Multilayer headgaskets in between layers where they are not visible.
After cehcking the block, I'd probably by a new OEM HG, and new OEM bolts, and cross my fingers one more time that it worked out this time, before ripping the block and head out to be disassembled, tanked, and magafluxed to search for cracks. :(

What about adding dye to the coolant, to help spot any leak source? Anybody tried this?

BTW, two very seasoned car mechanics have checked the HG that I just installed a few miles back, and both found it to be fine. Whatever leak source there is, is not related to the head or head gasket. Since the coolant cap is rated to probably less than 15lbs and since the oil runs at much more than that, it seems safe to say there is no DIRECT connection between coolant and oil, but rather a gravity flow of coolant to oil.

[Bin_jammin]

What about adding dye to the coolant, to help spot any leak source? Anybody tried this?

BTW, two very seasoned car mechanics have checked the HG that I just installed a few miles back, and both found it to be fine. Whatever leak source there is, is not related to the head or head gasket. Since the coolant cap is rated to probably less than 15lbs and since the oil runs at much more than that, it seems safe to say there is no DIRECT connection between coolant and oil, but rather a gravity flow of coolant to oil.


Trust me, that's the wrong way to think about it. Cars can put out high oil pressure, sure, but most of the time they don't. At idle, some cars have as little as 5 psi. I'm serious, before you waste time and money putting it back together, send your head to the machine shop and have it checked. Think of it as the absolute best money you can invest in a situation like this. The time you'll have into it will be roughly equal to the time in both parts and labor you'll have into doing a head gasket again.

[Dash01]

Trust me, that's the wrong way to think about it. Cars can put out high oil pressure, sure, but most of the time they don't. At idle, some cars have as little as 5 psi. I'm serious, before you waste time and money putting it back together, send your head to the machine shop and have it checked. Think of it as the absolute best money you can invest in a situation like this. The time you'll have into it will be roughly equal to the time in both parts and labor you'll have into doing a head gasket again.

Per my prior post #19, I did that this morning, and the head is perfect with no leaks in pressure test, etc.. HG was OK, too, and no signs of leak on either head or HG.

So, now I'm down to some other source of leak, perhaps the block or maybe around the timing chain, which does not seem too likely to me, but was suggested by the mechanic.

[Bin_jammin]

I'll take a closer look at my spare motor again tomorrow when I get back to the shop... maybe I missed something.

[whiskychaser]

Per my prior post #19, I did that this morning, and the head is perfect with no leaks in pressure test, etc.. HG was OK, too, and no signs of leak on either head or HG.

So, now I'm down to some other source of leak, perhaps the block or maybe around the timing chain, which does not seem too likely to me, but was suggested by the mechanic.

As I posted previously, that happened to me on the M50 but I cant see coolant getting across two seperate gaskets on yours. Having said that, did you need to remove the water pump when you changed the timing chain cover?

[Dash01]

As I posted previously, that happened to me on the M50 but I cant see coolant getting across two seperate gaskets on yours. Having said that, did you need to remove the water pump when you changed the timing chain cover?

Thanks Whiskeychaser, those are my thoughts too. Seems to me, if the water pump seal were leaking, the coolant would just run down the front of the engine, but could not go laterally and somehow penetrate the adjoining timing chain gasket. I did not remove the water pump when changing the timing chain or removing its cover.

The ETK shows a couple of screw plugs in its illustrations of the engine block, along with a gasket ring, but is unclear exactly how one of these screw plugs w/gasket ring fits. (ETK is also wrong sometimes, as with its depiction of rear cover plates on the head.)

Anyway, if there are in fact screw plugs, and one of them has a gasket ring that is loose or missing, and if the screw holes penetrate the front of the engine block into the coolant cavity, it could be a source of a coolant leak down the timing chain race and into the oil pan. (See ETK engine block illustration, part No. 02 SCREW PLUG M14X1,5 ZNS 07119919014; part No. 03 SCREW PLUG AM14X1,5 7119919228; and part No. 04 GASKET RING 14X18-AL 07119963200)

In other words, coolant when pressurized by heat and gravity flow could push past a loose or un-gasketed screw plug and then run down the front of the block and into the oil pan. IF said screw plugs in fact exist, IF their holes penetrate the front of the block to the coolant cavity, and IF they are loose or not properly sealed.

I'll have to take the timing chain cover off to find out, and that is a gross PITA.

Or, I could have a cracked block. Oh, happy day!

[whiskychaser]

Just to really piss you off, can I point out there also appear to be 2 screw plugs in the head? I can only guess they are there from when the head or block was drilled out and doubt very much they are to blame.

[DaveVoorhis]

If there are screw plugs in the head, it sounds like one or both are leaking. Pressurised coolant can be forced out of the coolant passages past a leaky plug, where it then drains into the pan. However, the oil is not pressurised on top of the plug, so it doesn't go into the coolant. Pull the valve cover -- if you've got screw plugs, you'll see 'em.

EDIT: I just examined one of the broken M20 heads I've got knocking around the shed. It has a screw plug between the coolant passages above cylinder 3 and the top of the head. If the M30 has similar plugs, you've got a leaky one.

[Dash01]

If there are screw plugs in the head, it sounds like one or both are leaking. Pressurised coolant can be forced out of the coolant passages past a leaky plug, where it then drains into the pan. However, the oil is not pressurised on top of the plug, so it doesn't go into the coolant. Pull the valve cover -- if you've got screw plugs, you'll see 'em.

EDIT: I just examined one of the broken M20 heads I've got knocking around the shed. It has a screw plug between the coolant passages above cylinder 3 and the top of the head. If the M30 has similar plugs, you've got a leaky one.

This is an March 1990 vintage M30 535i engine and its head and headgasket are fine, per pressure test & inspection. The source of the leak is not the head or headgasket or top of the engine block, which are all OK.

The screw plugs in question are in the BLOCK, behind or near the timing chain cover. The ETK shows two screw plugs there, one with a ring gasket. But, ETK has errors, so I don't know if is correct about screw plugs.

What I'm trying to learn is IF there are screw plugs there, and IF they are loose or lack proper gasket(s), IF that may be the source of coolant running down into the oil pan.

At the moment I don't have the proper tools to remove the big nut on the front of the crankshaft, prerequisite to removing the timing chain cover for inspection of screw plugs, etc..

[Gene in NC]

'90 535ia M30 engine gets coolant in oil, but no oil in coolant.

Head and block have no no apparent crack, headgasket looks OK, with no obvious streaking or other signs of leakage or seepage between oil and coolant holes. Cylinders and plugs are dry. Freeze plugs all OK, as is plastic coolant reservoir and plastic radiator tanks. Car runs fine and does not overheat, with temp gauge pegged at 12 o'clock position.

Put in new oil and filter, and within a few miles the oil starts to get frothy with coolant mixing in.

Head was rebuilt and resurfaced only recently, and placing a mechanic's straightedge along the head and also along the block indicates no warping of either.

Where else could coolant find its way into the oil?
Dash01,
What was final answer on your water in oil thread. You left us hanging.