Engine mixing water and oil after replacing head gasket

[eltonjohan]

I'm depressed right now
I basically been living in the garage since three month pimpin' up my e9. Engine and transmission out, repainting the engine bay, totally renewing everything on front axle, steering etc (I will hopefully come back with some pictures of this later).
I have also been renewing some parts in the engine, like rocker shafts/arms, valve stems gaskets etc.
Today was the big start up day. The engine started at first attempt - happy days! But when removing the valve cover for final adjustment of the valves I noticed that it was full of glycol/water in the oil!! The glycol seem to come along the one or two rocker shafts. Ok, removing the cylinder head again (puh) and having a look at the head gasket. It looks like it has hardly been compressed (?). The red "lines" on the gasket where not flattened as they usually are after tightening the head bolts. According to the book it should only be tightened to 70 Nm which seems like a quite small torque. Is that really correct?

I hope some of you have some good ideas to solve this problem!
To sum it up:
-The engine wasn't having any problems similar to this before.
- I have not damaged the head or block during this period and they don't seem to be uneven.
- The oil is totally mixed with water/glycol after only apprx 2 minutes of running
- No oil in the glycol.

Oh, this turns out to be a very long post, sorry!

thanks
/Johan

 

[eltonjohan]

I'm depressed right now
I basically been living in the garage since three month pimpin' up my e9. Engine and transmission out, repainting the engine bay, totally renewing everything on front axle, steering etc (I will hopefully come back with some pictures of this later).
I have also been renewing some parts in the engine, like rocker shafts/arms, valve stems gaskets etc.
Today was the big start up day. The engine started at first attempt - happy days! But when removing the valve cover for final adjustment of the valves I noticed that it was full of glycol/water in the oil!! The glycol seem to come along the one or two rocker shafts. Ok, removing the cylinder head again (puh) and having a look at the head gasket. It looks like it has hardly been compressed (?). The red "lines" on the gasket where not flattened as they usually are after tightening the head bolts. According to the book it should only be tightened to 70 Nm which seems like a quite small torque. Is that really correct?

I hope some of you have some good ideas to solve this problem!
To sum it up:
-The engine wasn't having any problems similar to this before.
- I have not damaged the head or block during this period and they don't seem to be uneven.
- The oil is totally mixed with water/glycol after only apprx 2 minutes of running
- No oil in the glycol.

Oh, this turns out to be a very long post, sorry!

thanks
/Johan

 

[gazzol]

I'm not sure weather 70Nm is enough or not as I don't have a conversion table to hand and I'm only familiar with Imperial measurements. However assuming that the measurement is correct did you clean out the holes in the block where the bolts screw into? If you had oil or water in the bolt holes this would prevent the bolts squashing down the head gasket before the torque figure was reached also if the threads were to dry this would have the same effect because the friction between the threads adds to the effort required to tighten the bolt so the effect is that whist the torque figure is reached the ammount of clamping force applied is reduced. This is why modern practice is to tighten heads bolts not by a torque figure but by degrees of rotation. That way if one bolt has oil on the threads and the other has rusty threads (not that you would assemble it that way but you get the point) they would both exert the same clamping force because they have been turned through the same number of degrees.
The ideal situation when using a torque figure to tighten bolts is to have clean bolt holes and threads on the bolts with a very light smearing of oil on the bolt threads

 

[gazzol]

I'm not sure weather 70Nm is enough or not as I don't have a conversion table to hand and I'm only familiar with Imperial measurements. However assuming that the measurement is correct did you clean out the holes in the block where the bolts screw into? If you had oil or water in the bolt holes this would prevent the bolts squashing down the head gasket before the torque figure was reached also if the threads were to dry this would have the same effect because the friction between the threads adds to the effort required to tighten the bolt so the effect is that whist the torque figure is reached the ammount of clamping force applied is reduced. This is why modern practice is to tighten heads bolts not by a torque figure but by degrees of rotation. That way if one bolt has oil on the threads and the other has rusty threads (not that you would assemble it that way but you get the point) they would both exert the same clamping force because they have been turned through the same number of degrees.
The ideal situation when using a torque figure to tighten bolts is to have clean bolt holes and threads on the bolts with a very light smearing of oil on the bolt threads

 

[30csl]

Other than that - maybe she has had the head skimmed in the past and the gasket you took off was actually thicker than original? I doubt this would make enough difference but just a thought.

 

[30csl]

Other than that - maybe she has had the head skimmed in the past and the gasket you took off was actually thicker than original? I doubt this would make enough difference but just a thought.

 

[x_atlas0]

Did you do the rest of the tightening sequnce? It is a multi-part deal.

According to the E34 BMW service manual, (granted, this is for a B35, but it should be the same) you are supposed to tighten the bolts to 60 Nm in the proper sequence, then wait ~20 min, then redo the sequence up to 80 Nm, then run the car for ~25 min, then turn each bolt another 35 degrees.

 

[x_atlas0]

Did you do the rest of the tightening sequnce? It is a multi-part deal.

According to the E34 BMW service manual, (granted, this is for a B35, but it should be the same) you are supposed to tighten the bolts to 60 Nm in the proper sequence, then wait ~20 min, then redo the sequence up to 80 Nm, then run the car for ~25 min, then turn each bolt another 35 degrees.

 

[jhjacobs]

The torque is supposed to be 51ft-lb +/- 1.4 which is 69nm +/- so you are doing the right thing.

Gazzol's assessment seems pretty reasonable and likely. There is one more possibility that I can offer up. There are two head alignment sleeves that rest in the block and head; they are about 1 cm deep and they insert half way into the block and half way into the head. If one or both of these were not properly seated it could keep the head from fully compressing the gasket. Also, if the head was milled there is a chance that these sleeve could be a hair too long. You can grind them down a bit without any concern.

One last possibility is that the gasket was not torqued in the right order. If the bottom was torqued before the top or front before rear if could have cantilevered the head. Make sure that you first hand tighten from a center bolt working your way out clockwise or anticlockwise with ever increasing tension - perhaps 5nm, 20nm, 50nm, 60nm, and finally 70nm.

There is also the obvious, any residual gasket old material or debris could cause this but I suspect you did a thorough clean up before reassembling.

Good luck.

 

[jhjacobs]

The torque is supposed to be 51ft-lb +/- 1.4 which is 69nm +/- so you are doing the right thing.

Gazzol's assessment seems pretty reasonable and likely. There is one more possibility that I can offer up. There are two head alignment sleeves that rest in the block and head; they are about 1 cm deep and they insert half way into the block and half way into the head. If one or both of these were not properly seated it could keep the head from fully compressing the gasket. Also, if the head was milled there is a chance that these sleeve could be a hair too long. You can grind them down a bit without any concern.

One last possibility is that the gasket was not torqued in the right order. If the bottom was torqued before the top or front before rear if could have cantilevered the head. Make sure that you first hand tighten from a center bolt working your way out clockwise or anticlockwise with ever increasing tension - perhaps 5nm, 20nm, 50nm, 60nm, and finally 70nm.

There is also the obvious, any residual gasket old material or debris could cause this but I suspect you did a thorough clean up before reassembling.

Good luck.

 

[bengal taiga]

many things to consider

when removing the valve cover for final adjustment of the valves I noticed that it was full of glycol/water in the oil!! The glycol seem to come along the one or two rocker shafts. Ok, removing the cylinder head again (puh) and having a look at the head gasket. It looks like it has hardly been compressed (?). The red "lines" on the gasket where not flattened as they usually are after tightening the head bolts. According to the book it should only be tightened to 70 Nm which seems like a quite small torque. Is that really correct?

I hope some of you have some good ideas to solve this problem!
To sum it up:
-The engine wasn't having any problems similar to this before.
- I have not damaged the head or block during this period and they don't seem to be uneven.
- The oil is totally mixed with water/glycol after only apprx 2 minutes of running

I am sorry to read about your problems. I have certainly had my fair share, so I can relate. I am also sorry to say that as simple as replacing a head gasket is, it is just as simple to miss something obvious. A few things come to mind:

1. You really don't specify what engine you are working with and exactly what equipment you are using. If you are working with anything other than the original stock block and head, there is a good possibility that you have different head bolts too. If that is the case, you might be using one-time-only torque to yield bolts. They stretch and cannot be reused. If you are using the stock headbolts, they still may be stretched or worn so they are not reuseable. If you are reusing the stock bolts, as I have done, get yourself a known good bolt, perhaps from the dealer and measure it and compare that measurement with the used bolts.

2. Are you using a new head gasket? Unless you are using a solid copper gasket, I understand stock steel composite gasket shelf life is relatively short (not indefinite). So, unless you obtained a shrink wrapped gasket, you should consider getting a newer one.

3. It does not sound as though you are following the gasket manufacturer's torquing instructions. I suppose the old three step tightening method described in the manual will work, but BMW has revised its torquing procedure so that the final step involves turning each bolt a specific number of degrees. This is separate from using a torque wrench. Again, consult the instructions that came with the gasket. This raises another question, is your torque wrench accurate? Has it been calibrated? If it is the simple beam type wrench, it is probably accurate, but then how is your eyesight to read the bloody thing?

4. Are you sure you put the gasket on correctly? It is possible that you flipped it over and a couple of the galley passages did not match up. I can't specifically recall the M30 gasket, but I believe one side is marked "Top" for a reason.

5. You state: "I have not damaged the head or block during this period and they don't seem to be uneven." How do you know this? Unless you have measured the head and deck of the block with a machinist's straight edge and a feeler gauge, you really do not know whether you have any warpage. Most repair manuals explain the procedure of measuring for straightness, but when in doubt, I would take the head to a competent machinist and have him/her/it measure it. If your cylinder head is even slightly warped, you may never get a decent seal with only one head gasket.
6. How clean were the threads in the block? If they were not all cleaned out, as others have mentioned, you will not get an accurate torque reading from your wrench and might as well not use a calibrated torque wrench at all. Those threads must be clean and I would want to run a clean head bolt down each set of threads to the point where you encounter only minimal resistance. Remember to lightly coat each bolt with engine oil when reinstalling them.

7. the manual and probably the head gasket instructions call for a dry gasket (except for the dab of adhesive where the timing cover meets the head). Due to all of the unknowns respecting your prior installations, I might consider using one of several head gasket adhesives, such as high tack or the spray version of copper gasket adhesive.

8. Are you sure your cylinder head is not cracked somewhere in or near the water jackets? Was your head ever pressure tested? While you state you never experienced the antifreeze mixing with oil prior to your latest head work, you do not explain what work you did to the head. If, for example, you hot tanked the head or decarbonized it with a wire wheel, you may have exposed a preexisting weakness in the casting. Early cylinder heads were known for this, but I suppose no cylinder head is immune from such things.

9. It has been mentioned before, all mating surfaces must be flat and CLEAN and Debris-Free. The gasket is somewhat forgiving and will obvious make up for certain irregularities, but it will not compress well over miscellaneous parts. I seem to recall that there are two alignment dowels. I also seem to recall once having a problem where one of the dowels was too tall and made tightening the head a nightmare. I do not know if the dowel had been inadvertently switched with another or exactly what happened. I ultimately tossed both of them and used a couple of head bolts as a guide and experienced no known problems as a result.

10. The gasket. Was it in anyway bent, creased, or kinked?

Hoppa på att hjälper. Good luck.

 

[bengal taiga]

many things to consider

when removing the valve cover for final adjustment of the valves I noticed that it was full of glycol/water in the oil!! The glycol seem to come along the one or two rocker shafts. Ok, removing the cylinder head again (puh) and having a look at the head gasket. It looks like it has hardly been compressed (?). The red "lines" on the gasket where not flattened as they usually are after tightening the head bolts. According to the book it should only be tightened to 70 Nm which seems like a quite small torque. Is that really correct?

I hope some of you have some good ideas to solve this problem!
To sum it up:
-The engine wasn't having any problems similar to this before.
- I have not damaged the head or block during this period and they don't seem to be uneven.
- The oil is totally mixed with water/glycol after only apprx 2 minutes of running

I am sorry to read about your problems. I have certainly had my fair share, so I can relate. I am also sorry to say that as simple as replacing a head gasket is, it is just as simple to miss something obvious. A few things come to mind:

1. You really don't specify what engine you are working with and exactly what equipment you are using. If you are working with anything other than the original stock block and head, there is a good possibility that you have different head bolts too. If that is the case, you might be using one-time-only torque to yield bolts. They stretch and cannot be reused. If you are using the stock headbolts, they still may be stretched or worn so they are not reuseable. If you are reusing the stock bolts, as I have done, get yourself a known good bolt, perhaps from the dealer and measure it and compare that measurement with the used bolts.

2. Are you using a new head gasket? Unless you are using a solid copper gasket, I understand stock steel composite gasket shelf life is relatively short (not indefinite). So, unless you obtained a shrink wrapped gasket, you should consider getting a newer one.

3. It does not sound as though you are following the gasket manufacturer's torquing instructions. I suppose the old three step tightening method described in the manual will work, but BMW has revised its torquing procedure so that the final step involves turning each bolt a specific number of degrees. This is separate from using a torque wrench. Again, consult the instructions that came with the gasket. This raises another question, is your torque wrench accurate? Has it been calibrated? If it is the simple beam type wrench, it is probably accurate, but then how is your eyesight to read the bloody thing?

4. Are you sure you put the gasket on correctly? It is possible that you flipped it over and a couple of the galley passages did not match up. I can't specifically recall the M30 gasket, but I believe one side is marked "Top" for a reason.

5. You state: "I have not damaged the head or block during this period and they don't seem to be uneven." How do you know this? Unless you have measured the head and deck of the block with a machinist's straight edge and a feeler gauge, you really do not know whether you have any warpage. Most repair manuals explain the procedure of measuring for straightness, but when in doubt, I would take the head to a competent machinist and have him/her/it measure it. If your cylinder head is even slightly warped, you may never get a decent seal with only one head gasket.
6. How clean were the threads in the block? If they were not all cleaned out, as others have mentioned, you will not get an accurate torque reading from your wrench and might as well not use a calibrated torque wrench at all. Those threads must be clean and I would want to run a clean head bolt down each set of threads to the point where you encounter only minimal resistance. Remember to lightly coat each bolt with engine oil when reinstalling them.

7. the manual and probably the head gasket instructions call for a dry gasket (except for the dab of adhesive where the timing cover meets the head). Due to all of the unknowns respecting your prior installations, I might consider using one of several head gasket adhesives, such as high tack or the spray version of copper gasket adhesive.

8. Are you sure your cylinder head is not cracked somewhere in or near the water jackets? Was your head ever pressure tested? While you state you never experienced the antifreeze mixing with oil prior to your latest head work, you do not explain what work you did to the head. If, for example, you hot tanked the head or decarbonized it with a wire wheel, you may have exposed a preexisting weakness in the casting. Early cylinder heads were known for this, but I suppose no cylinder head is immune from such things.

9. It has been mentioned before, all mating surfaces must be flat and CLEAN and Debris-Free. The gasket is somewhat forgiving and will obvious make up for certain irregularities, but it will not compress well over miscellaneous parts. I seem to recall that there are two alignment dowels. I also seem to recall once having a problem where one of the dowels was too tall and made tightening the head a nightmare. I do not know if the dowel had been inadvertently switched with another or exactly what happened. I ultimately tossed both of them and used a couple of head bolts as a guide and experienced no known problems as a result.

10. The gasket. Was it in anyway bent, creased, or kinked?

Hoppa på att hjälper. Good luck.

 

[sfdon]

# 1 No torque angle on M12 bolts.

# 2 Bolts are washed and oiled, bores are clean

#3 Torque 3 times- 50 Nm
80 Nm
wait 15 minutes
100 NM

#4 Torque angle is for M10 bolts

 

[sfdon]

# 1 No torque angle on M12 bolts.

# 2 Bolts are washed and oiled, bores are clean

#3 Torque 3 times- 50 Nm
80 Nm
wait 15 minutes
100 NM

#4 Torque angle is for M10 bolts

 

[sfdon]

Posted the bmw spec page for you...........

http://bmwturbos.scottiesharpe.com/topic.asp?TOPIC_ID=5392

 

[sfdon]

Posted the bmw spec page for you...........

http://bmwturbos.scottiesharpe.com/topic.asp?TOPIC_ID=5392

 

[bengal taiga]

# 1 No torque angle on M12 bolts.
# 2 Bolts are washed and oiled, bores are clean
#3 Torque 3 times- 50 Nm
80 Nm
wait 15 minutes
100 NM
#4 Torque angle is for M10 bolts

Perhaps I misspoke regarding BMW's recommended torquing procedures which is why I also suggested reviewing the gasket maker's instructions. Nevertheless, I have been given to understand that BMW superseded its original headbolt tightening recommendations in 1986 with service bulletin 11 1985(1106).

For some reason, perhaps mistakenly, I annotated my original page to indicate that "M30 headbolts were torqued to 60NMs, wait 15-20 min, angle torque 33 degrees followed by a 25 min run in and an additional torque angle of 35 degrees - regardless of temp."

The information you posted does not bear a date, so it is not clear whether this information may have been modified or superseded. Nor does it specify exactly to which engine it applies. Take a look at p15 of this site: http://www.bmw-m.net/TechData/torq/bmw_torq.pdf It does not distinguish between bolt sizes nor does the service bulletin do so. This is probably because the M30 did not use M10 bolts. M12 (part no. 11120621144). I mention this since 02's use the same M12 bolts and I recently noticed the headbolt torquing instructions that accompanied the gasket set and I am pretty sure it mentioned a final angle torque measurement.

While this is hardly a definitive answer, posters on other sites recognize this torquing dichotomy. See: http://www.mye28.com/viewtopic.php?t=59706 and http://forums.bimmerforums.com/forum/showthread.php?t=983975

Since this information is garnered from the net, one couldn't say which recommendation is reliable and/or preferred. However, it stands to reason that service bulletins ordinarily contain improved methodology and that the latest angle torque method is probably more accurate than the factory's original 3 stage method.

For the original poster, you might wish to look at the following link as it describes some of the differences between early and later water jacket/head design. http://www.firstfives.org/faq/cylinder_head/cylinder_head_faq.htm

 

[bengal taiga]

# 1 No torque angle on M12 bolts.
# 2 Bolts are washed and oiled, bores are clean
#3 Torque 3 times- 50 Nm
80 Nm
wait 15 minutes
100 NM
#4 Torque angle is for M10 bolts

Perhaps I misspoke regarding BMW's recommended torquing procedures which is why I also suggested reviewing the gasket maker's instructions. Nevertheless, I have been given to understand that BMW superseded its original headbolt tightening recommendations in 1986 with service bulletin 11 1985(1106).

For some reason, perhaps mistakenly, I annotated my original page to indicate that "M30 headbolts were torqued to 60NMs, wait 15-20 min, angle torque 33 degrees followed by a 25 min run in and an additional torque angle of 35 degrees - regardless of temp."

The information you posted does not bear a date, so it is not clear whether this information may have been modified or superseded. Nor does it specify exactly to which engine it applies. Take a look at p15 of this site: http://www.bmw-m.net/TechData/torq/bmw_torq.pdf It does not distinguish between bolt sizes nor does the service bulletin do so. This is probably because the M30 did not use M10 bolts. M12 (part no. 11120621144). I mention this since 02's use the same M12 bolts and I recently noticed the headbolt torquing instructions that accompanied the gasket set and I am pretty sure it mentioned a final angle torque measurement.

While this is hardly a definitive answer, posters on other sites recognize this torquing dichotomy. See: http://www.mye28.com/viewtopic.php?t=59706 and http://forums.bimmerforums.com/forum/showthread.php?t=983975

Since this information is garnered from the net, one couldn't say which recommendation is reliable and/or preferred. However, it stands to reason that service bulletins ordinarily contain improved methodology and that the latest angle torque method is probably more accurate than the factory's original 3 stage method.

For the original poster, you might wish to look at the following link as it describes some of the differences between early and later water jacket/head design. http://www.firstfives.org/faq/cylinder_head/cylinder_head_faq.htm

 

[eltonjohan]

Thanks a lot for all pcs of advice!! Really appreciate it.
I'm still confused what has been going wrong though!

As bengal taiga's points:
1) I'm working on my stock 3-litre engine (CS) with the improved cylinder head from a -79. All bolts (m12) are bought new from W&N.

2) New head gasket from W&N http://www.wallothnesch.com/d/pe/e55.jpg

3) No instructions came with the gasket, I followed the instructions in the e9 repair manual (70 Nm), then loosened the bolts and re-tightened them again according to m30-instructions (66Nm + 33 deg). The torque wrench is new and a note came along were it said it was calibrated with some +/- figures. But of course, it wasn't the most expensive wrench so you never know.

4) Yes, rather sure - not really possible to place it wrong.

5) I will measure it today with a straigth edge, but it seems strange that it worked fine with the old gasket if the block or head is warped.

6) This might have been a problem. The holes were clean and dry (I also used a threaded tap (?) in the holes. Though - I did not oil the bolts, but I'm rather sure that they were a bit greasy from the start.

7) Dry gasket. Gasket adhesiv at the timing cover junction.

8) No, I'm not sure that the head is crack free. But since it was ok before and since I have been treating it careful I assumed that it should be ok.
I applied pressure with a compressor through the cooling system and the clycol came along the front intake rocker shaft (which is new).

9) Good tip. I will grind of the dowels to make sure that they are causing the problem (especially if the new gasket is thinner).

10) The gasket looked almost unused. You could only notice that it had been compressed at the ends...

I guess I have to order a new gasket and try again after grinding down the dowels and measuring the head/block with a straight edge.

everyone - thanks once again![/img]

 

[eltonjohan]

Thanks a lot for all pcs of advice!! Really appreciate it.
I'm still confused what has been going wrong though!

As bengal taiga's points:
1) I'm working on my stock 3-litre engine (CS) with the improved cylinder head from a -79. All bolts (m12) are bought new from W&N.

2) New head gasket from W&N http://www.wallothnesch.com/d/pe/e55.jpg

3) No instructions came with the gasket, I followed the instructions in the e9 repair manual (70 Nm), then loosened the bolts and re-tightened them again according to m30-instructions (66Nm + 33 deg). The torque wrench is new and a note came along were it said it was calibrated with some +/- figures. But of course, it wasn't the most expensive wrench so you never know.

4) Yes, rather sure - not really possible to place it wrong.

5) I will measure it today with a straigth edge, but it seems strange that it worked fine with the old gasket if the block or head is warped.

6) This might have been a problem. The holes were clean and dry (I also used a threaded tap (?) in the holes. Though - I did not oil the bolts, but I'm rather sure that they were a bit greasy from the start.

7) Dry gasket. Gasket adhesiv at the timing cover junction.

8) No, I'm not sure that the head is crack free. But since it was ok before and since I have been treating it careful I assumed that it should be ok.
I applied pressure with a compressor through the cooling system and the clycol came along the front intake rocker shaft (which is new).

9) Good tip. I will grind of the dowels to make sure that they are causing the problem (especially if the new gasket is thinner).

10) The gasket looked almost unused. You could only notice that it had been compressed at the ends...

I guess I have to order a new gasket and try again after grinding down the dowels and measuring the head/block with a straight edge.

everyone - thanks once again![/img]