Oil consumption after CCV replacement

[bmwdirtracer]

You're really going to have to get over your distrust, and find a good shop for certain things, like testing.

Your intake/ccv system is far too complex to address by just throwing parts: every time you throw one part, you stress three more. And the obvious parts may not be the leaking parts: what about the dipstick tube's o-ring, the valvecover gasket, the throttle body o-ring? A smoke test is necessary, as is the crankcase vacuum test.

The OP of this thread is an experienced pro: he's even made his own slack-tube manometer, to accomplish the needed tests, because he knew the tests will usually lead to the answer.

There's little gain in throwing parts at a fault before doing the research and testing for the common failures.Duplicating my smoke machine is expensive, duplicating my manometer is an endeavour, but can be done.

[mahmoudh]

You're really going to have to get over your distrust, and find a good shop for certain things, like testing.

Your intake/ccv system is far too complex to address by just throwing parts: every time you throw one part, you stress three more. And the obvious parts may not be the leaking parts: what about the dipstick tube's o-ring, the valvecover gasket, the throttle body o-ring? A smoke test is necessary, as is the crankcase vacuum test.

....

Thanks for the advice. I am slowly coming around to that realization but will not until after I fix what I can on my own. VCG was done last winter with Vanos seals. Dipstick tube and its o-ring as well as dipstick o-rings are new from a few weeks ago when I replaced the CCV again. Replaced vaccum tube and cap on intake as well. CCV Hoses are new as of last winter.

New oil filler cap has a more snug fit. Will wait to see if that changes anything and then do the throttle body and DISA o-rings. Intake boot was replaced less than 40K miles ago but I will check again for cracks etc.

[bmwdirtracer]

I appreciate your dedication, but throwing gaskets and parts at things that aren't leaking often disturbs other parts in the process, causing the replacements to be inconclusive, at best.

You could make your own manometer, the details are above, somewhere. Or they cost about $65, as I recall.

Don't bother trying to make a smoke machine: I tried twice before I gave up and bought one. One of two of my personal finest diag tools - the other one's my GT1. Both were expensive, and worth every cent.

[mahmoudh]

I appreciate your dedication, but throwing gaskets and parts at things that aren't leaking often disturbs other parts in the process, causing the replacements to be inconclusive, at best.

You could make your own manometer, the details are above, somewhere. Or they cost about $65, as I recall.

Don't bother trying to make a smoke machine: I tried twice before I gave up and bought one. One of two of my personal finest diag tools - the other one's my GT1. Both were expensive, and worth every cent.

You said it. Since the new oil filler cap, SES light has come and gone twice, on cold days - Codes the first time were P0171 and P0174. Did not get a chance to read 2nd set. Still sucking up oil at same rate.
Definitely looks like a vaccum leak. Both intake boots look good. Next on my list if the DISA o-ring, ICV grommet and TB gasket.

I could try to rig up a manometer from the instructions but the smoke machine seems more useful - have had a lot of vaccum leaks in the winter.

How expensive is the smoke machine and how easy or difficult to use?

[mahmoudh]

Car feeling more responsive and smoother after working on it Saturday to correct vaccum leak (SES light with P0171 and P0174 fault codes - Have a P0313 which I am ignoring since fuel level was low.

Changed TB gasket and DISA o-ring. Cleaned ICV while it was out and made sure intake boots were correctly seated and all clamps tight.

Replaced dipstick tube o-ring which had a kink in it from last time - this time I installed it onto dipstick tube then forced tube into crankcase.

All of this is on top of replacing Oil filler cap.

Puzzled to see no sign of oil inside drain hose from CCV to dipstick tube Y.
Been losing a quart every 600 miles or so before.

Will monitor and report back on oil consumption and check for oil in tube.

[ptarditi]

Any chance you put the hose from the valve cover to the ccv on backwards? I've seen this and it kinks the hose.

[mahmoudh]

Any chance you put the hose from the valve cover to the ccv on backwards? I've seen this and it kinks the hose.

Hmm, it looks difficult to put that hose on backwards based on the angle of the hose at the ends but I will double check that. Thanks

[mahmoudh]

Hmm, it looks difficult to put that hose on backwards based on the angle of the hose at the ends but I will double check that. Thanks

I checked the hose and there are no kinks nor is it installed backwards.
I removed the end of this hose at the Valve Cover and saw a good amount of oil vapors coming from the VC so CCV is getting blow by gases.
However, I still cannot find any sign of oil in the dipstick tube end of the drain hose coming from the CCV after driving 350 miles from last check.

Puzzled - What can be causing this?

I felt a vaccum at the end of the hose when I covered it with by finger - Was it supposed to be a pressure instead?

The engine idled rough when I removed my finger from the end of the hose so is the new CCV (a month old) completely dead and never worked hence no oil in the hose?

Can it still be a vaccum leak that is small enough to prevent CCV operation but not big enough to throw a CEL or code?

Is the CCV not getting enough vaccum to do the oil/gas separation from the blow by gases?

[02Pilot]

I still haven't had a chance to further diagnose my own situation, so I've been staying out of this, but I've noticed a similar condition to what you are describing re: the lack of oil in the drain tube. I have a strong suspicion that this is due to a) a very small oil separator (compare it with other units; it's small), and b) the fact the the cold-weather CCV reduced the temperature differential between the crankcase and the CCV, which has the unintended side effect of reducing the separation of oil from crankcase vapors. The net effect, I suspect, is that the cold-weather CCV is likely to increase oil consumption in anything other than (surprise, surprise) cold weather. A larger oil separator and/or a greater temperature differential (i.e., time for the crankcase vapors to cool and allow the oil to separate) would, I'm suspecting, likely help to limit oil usage.

On a related note, this is one of the more relevant technical documents I've found relating to the issue at hand: http://dspace.mit.edu/bitstream/hand...pdf?sequence=1

[02Pilot]

They haven't overlooked it; on the contrary, they've been involved since the beginning. The problem is that they need concrete evidence to work with, the kind that comes from diagnostic testing. I haven't done much yet (crankcase vacuum test notwithstanding). All my rambling speculation may be well and good for entertainment purposes, but it doesn't mean much from a "fixing what's broken" standpoint.

Giving this one a bump in hopes that the pros might have simply overlooked it.

[bmwdirtracer]

Actually, it's interesting to read your speculation about temps, and the "cold weather" set-up. See, I've never actually seen a ccv failure, though I've done the crankcase vacuum test many times, and I've found many trashed ccv hoses, with my smoke machine.

I know - every mainland tech does them all the time, but in Hawaii, ccv failures are nearly unheard of - even by my friend, who's the lead tech at our dealership. And of course, in Hawaii, it's always relatively "balmy" in temperature. I'd be curious whether BMW is delivering the cold-weather part pretty much universally now. Guess I'll call my buddy and ask...

Heather, this thread is, and has been since the very beginning, pretty much pros, talking to each other. 02 pilot even made his own Slack Tube Manometer, a diagnostic instrument most mechanics have never heard of, and wouldn't know how to use.There's no lack of experience here, don't worry.

[02Pilot]

It was that MIT paper I linked above that got me thinking. While it doesn't specifically address the effects of separator design, it does make clear that it (along with a lot of other factors) plays a role in oil consumption. BMW seems to prefer the cyclonic separator design, but the M54 separator is really small, which makes me wonder how good a job it can do even without the additional issue of high gas temps. Other designs, or at least larger volume cyclonic separators, seem better suited to slowing and cooling gas flow, allowing oil droplets to fall out of the gas stream and be returned to the oil pan.

[mahmoudh]

O2Pilot,
The MIT paper is very interesting - especially all the variables affecting blow by gases and oil consumption. Thanks for the link.
I will read this paper some more and try to come up with some numbers later.

Your points on the M54 oil separator size and thermal response due to insulated hoses are valid. The question now becomes how much should that impact oil separation efficiency and oil consumption?

Note that essentially this same separator is used in the M52 which have reported less issues with oil consumption.

Thanks for confirming you are also not seeing oil in drain tube - it seems that we are getting no oil separation.

[02Pilot]

IIRC the M52 separator is of similar dimensions, but has a more elongated cone shape, whereas the M54 piece is very compact in the vertical dimension. As a point of contrast, look at the S62 separator featured in this post (yes, they have the hoses installed upside down; the inlet is supposed to be below the outlet): http://vsetrack.com/track_reports/20..._separator.htm

[mahmoudh]

IIRC the M52 separator is of similar dimensions, but has a more elongated cone shape, whereas the M54 piece is very compact in the vertical dimension. As a point of contrast, look at the S62 separator featured in this post (yes, they have the hoses installed upside down; the inlet is supposed to be below the outlet): http://vsetrack.com/track_reports/20..._separator.htm

Sorry. You are correct on the earlier M52 separator being different but the M52 from 9/1998 onwards have the same CCV as the M54 - See attached pdf

The CCV P/N at the link you provided is used on the e39 M5. See M5 CCV in attached pdf.
That is an interesting Mod to consider for a daily driver application. Need to find out if it can be done without upsetting the delicate balance of M54 vaccum, pressure and flow and avoiding any CEL conditions.

[mahmoudh]

It was that MIT paper I linked above that got me thinking. While it doesn't specifically address the effects of separator design, it does make clear that it (along with a lot of other factors) plays a role in oil consumption. BMW seems to prefer the cyclonic separator design, but the M54 separator is really small, which makes me wonder how good a job it can do even without the additional issue of high gas temps. Other designs, or at least larger volume cyclonic separators, seem better suited to slowing and cooling gas flow, allowing oil droplets to fall out of the gas stream and be returned to the oil pan.

O2Pilot,

Using the MIT paper which you shared with us at the link below
http://dspace.mit.edu/bitstream/hand...pdf?sequence=1

AND

From Castrol's Specifications sheet showing that
Castrol 0W30 Synthethic oil has a density of 841.7 kg/m3 (0.842 g/cm3 or 0.842g/ml) at 15°C

http://www.castrol.com/liveassets/bp..._Sept_2011.pdf

I went thru the following analysis.

Although I agree with the concepts, variables, relationships and trends in the MIT paper, I have no idea how close the numbers presented correlate to real world applications.
For the sake of discussion and analysis, I will accept the numbers in the paper at face value.

The MIT paper on pg 88 measured total oil consumption (blow by and power) at 0.55 g/h on the 2.0 liter 4 cylinder engine.
Let us be generous and assume that the M54 consumes at least double that amount of oil because of 6 cylinders, loose piston rings, higher loads and temps etc.

So, assuming M54 engine uses twice oil consumption of 4 cyl=0.55x2 or 1.1g/hr
Using Castrol's 0W30 Synthetic oil density of 841kg/cubic meter or 0.842g/ml to convert weight to volume

Then M54 consumes (0.55x2)g x 0.842g/ml=0.9262ml of oil per hr.

For easier math adn discussion, let us round this number (0.962ml/hr) to 1ml per hour.
So we calculate the M54 consumes 1ml of oil per hour compared to the 0.46 ml per hour for the 4 cyl engine tested at MIT.

Since 1 qt(US Liq) = 0.946352L or 946 ml and we calculate the M54 burns 1ml per hour
Then to burn 1 qt or 946ml of synthetic oil, the M54 engine should run for at least 946 hours.

So using the calculation above, and lumping start, stop, idle and driving into an avg mph
1. Driving for 946 hours at 1 mile/hr average speed will cover 946 miles and burn 1 qt of oil
2. Driving for 946 hours at 10 miles/hr average speed will cover 9460 miles and burn 1 qt of oil

Are the assumptions and math above reasonable?

Real World Example:

In my current situation, I am seeing way much higher consumption than what is calculated above

When I first got the 2001, it burned/leaked 1 qt between oil change (5K) which is closer to #2 above.
So I consumed 1 qt out of 7 quarts total capacity = 14% total consumption
If this is the best I can get from my M54 with no leaks, then I should expect to consume 0.1 qt over 500 miles (1 qt/5000 miles)

The MIT paper concludes on pg 91 that blow by gases contributed a maximum of 7% of the total oil consumption for a 4 cylinder.
Let us be generous and assume on the M54 with 6 cylinders, higher loads and temps, blow by gases contribute double that amount (14%) to the total oil consumption.

How then do we account for the following?

When my VCG and oil filter housing gaskets were leaking at their worst, the 2001 was using 1 qt every 400-500 miles.
Since fixing VCG, oil filter housing gaskets and replacing CCV, consumption is now at 1 qt every 500-600 miles - small improvement but still unacceptable.
I see no oil in the hose going to the dipstick tube. This implies there is zero oil separated from the blowby gases.

The MIT paper discusses on pg 78 the relationship of pressure, vaccum and flow and requiring a pressure difference across CCV (Blowby Inlet and Drain tube)
If I am getting a vaccum at the bottom of the dipstick tube hose, then I do not meet the pressure difference requirements for flow.

I am starting to think my CCV is never getting enough vaccum at idle and much less at high rpms to ever close the path of oil vapors from CCV to intake.
Since the blow by gases just flow right back into the intake and burn, my consumption of 1 qt every 500-600 miles is 10 times more than what I should normally expect.

How do I test and confirm this?

[02Pilot]

As the ambient temperature has been dropping, I've been noticing that the frequency and intensity of oil smoke from the tailpipe has been dropping. This seems to support the idea that oil droplets are having trouble coalescing when the temperature of the CCV plumbing is closer to that of the crankcase; if true, this indicates that the cold weather CCV can increase oil consumption in warmer temperatures. I've found oil in the distribution piece, so I know it's getting past the CCV (the lack of oil in the drain supports this as well); I'm going to have to check the drain again and monitor consumption carefully now that it's cooled off around here.

I'm also thinking about a catch can design that can work with or without the CCV, and with a sealable oil return so I can accurately measure how much oil is being pulled through - I'll post more details on this as I work them out.

[thejlevie]

I believe the answer here is contained on page one of this thread. But it may not be obvious. A better understanding of what the CCV is and how it works may help.

The CCV is both an oil separator and a vacuum regulator. The valve in the CCV opens to the intake when the pressure in the crank case is higher than it should be. That results in crank case vapors (and oil) being pulled into the intake. Since there is always gas flow into the crank case (from ring blowby) there will always be some flow. Less at idle and more at higher engine speeds. The oil separator is supposed to limit the amount of oil that goes into the intake, draining coalesced oil back into the sump.

More than normal ring blowby or air leaks into the crank case will increase the flow through the CCV as it attempts to maintain the desired crank case pressure and result in higher oil usage. A smoke test (done properly) will find any air leaks. Leak down & compression tests will tell if the rings are at fault. If the results of those tests don't reveal a cause, the CCV isn't working correctly and needs to be replaced.

Another symptom of a bad CCV is a greater vacuum in the crank case than should be. On the first page you mentioned seeing 8" of water at high idle then dropping to 6". That should have been a red flag. I don't think the crankcase pressure should not be changing with engine speed like that and really should not be above the high end of the acceptable range (4-6" of water).

[02Pilot]

I appreciate your sticking with this, Jim, and I accept fully that I need to get those diagnostics done to have a fuller understanding of what's going on here. I do understand how the CCV functions and the relationship between inflow (blowby and/or vacuum leaks) and outflow through the valve.

The 8" of water was with the engine cold, and I believe I took that measurement before I modified the dipstick to replicate the new design. I should retest when the engine is cold.

But the question regarding ambient temperature remains; why would colder ambient temps affect the amount of oil being burned (or at least the visual manifestation of it)?

[thejlevie]

I don't know that I can explain why the oil consumption is worse in cold weather. But the valve in the CCV is operated by a rubber diaphragm which will be stiffer when colder and on cold starts engine clearances will be different. The oil will tend to pick up blowby condensation which may also be a factor.

[mahmoudh]

I believe the answer here is contained on page one of this thread. But it may not be obvious. A better understanding of what the CCV is and how it works may help.

The CCV is both an oil separator and a vacuum regulator. The valve in the CCV opens to the intake when the pressure in the crank case is higher than it should be. That results in crank case vapors (and oil) being pulled into the intake. Since there is always gas flow into the crank case (from ring blowby) there will always be some flow. Less at idle and more at higher engine speeds. The oil separator is supposed to limit the amount of oil that goes into the intake, draining coalesced oil back into the sump.

More than normal ring blowby or air leaks into the crank case will increase the flow through the CCV as it attempts to maintain the desired crank case pressure and result in higher oil usage. A smoke test (done properly) will find any air leaks. Leak down & compression tests will tell if the rings are at fault. If the results of those tests don't reveal a cause, the CCV isn't working correctly and needs to be replaced.

Another symptom of a bad CCV is a greater vacuum in the crank case than should be. On the first page you mentioned seeing 8" of water at high idle then dropping to 6". That should have been a red flag. I don't think the crankcase pressure should not be changing with engine speed like that and really should not be above the high end of the acceptable range (4-6" of water).

Jim, thanks for your detailed post.
I was initially thinking weak vaccum to the intake hose of my CCV is not activating its diaphragm to regulate blowby - either due to clogged air distribution piece or a small vaccum leak - but am thinking that should also result in weak pull of blowby gases and less oil consumption.
Also, if vaccum is weak, I should not feel the vaccum at the bottom of the hose going to the dispstick tube.

Another thought is that the CCV diaphram on my M54 is stuck open, not regulating any flow and all the blowby goes directly to the intake resulting in my high oil consumption.

As a comparison, if I remove the Hose at the M54 Valve Cover, I do not get any sounds from its CCV, whereas on my M52TU (which consumes no oil) I get a groaning sound from its CCV (99/528i uses same CCV)
I am going to change by CCV a third time but am going to test its diaphragm for vaccum operation before I do so.

If this does not work, I am going to do a compression and smoke test.

[Jason5driver]

Thanks to all for your posts which helped me tackle this project.

I just bought the new Guide Tube 1P/N 1437531258 from ECS Tuning and the bottom has concentric tubes below the Y. It also has the small vent hole below the Y as seen in pictures of the original tube design.
The tube is new with the correct part number.

Air blown from the CCV drain hose inlet does exit both at the inner and outer concentric tubes as well as small vent hole.

The tube that was in there looks and operates the same as the new one.

I did not have any blockages, mayo etc but am burning oil excessive even though I changed the CCV and hoses last winter.

All leaky gaskets were fixed and none in exhaust tailpipe

I did notice the old tube was loose and the O ring came out with it.
I replaced the O ring in the oil pan before inserting the tube and got a nice tight fit. Also replaced the two dipstick O rings

Not sure what to make of this - I am the 2nd owner from 72K in 2005 so this must be the original tube.
- Is BMW selling the original design with new P/N?
- Can worn O rings at tube and dipstick cause that much oil burn?

Can you buy new dipstick O-rings (the O-rings at the TOP of the dipstick) from the dealer?

Thanks!
Jason

[mahmoudh]

Can you buy new dipstick O-rings (the O-rings at the TOP of the dipstick) from the dealer?

Thanks!
Jason

Jason
I did replace the dipstick O-rings with new ones from VICTOR REINZ - P/N 11431717666. Are you suggesting BMW sourced are somehow different?
Now that you mention it, the dipstick does not feel as tight as on the other car. I will try swapping them as I will not be able to get to the CCV for a while due to cold weather and travel plans the next few weeks.

Jim, O2Pilot, Chris,
Your posts had me thinking everything over and helped advance my understanding of the variables that impact CCV operation under various scenarios. Here is what I understand. Let me know if I have any of this wrong.

I noted when I opened my old CCV that its intake vacuum port and oil separator output ports are both on the same side of the diaphragm - the vacuum port trying to pull the diaphragm close against the force of the spring and blowby gases from the separator output trying to push it open - so the position of the diaphragm (valve) at any particular instant depends on the net result of these two opposing forces applied at these 2 ports.

(See attached pdf for picture - Credit for this picture to Bluebee)


1. If CCV hose at VC is removed at idle, vacuum dominates, there is no blowby force from the separator port and the diaphragm should close against the force of the spring and block that port (hence the groaning sound from my M52?). No sound on my M54 leads me to believe CCV diaphragm is stuck open.
2. With all CCV hoses connected, at idle, normal vacuum and blowby forces compete against each other, the net force against the diaphragm fluctuating at the lower rpms. Smallest blowby flow and greatest separation efficiency should result in visible oil separation at drain hose going to dipstick tube. (I have no visible oil separation)
3. If vacuum force is weak at idle (blocked or leaky air piece or intake), blowby has less force from vacuum to overcome, diaphragm opens wider, increases blowby gas flow into the intake and result in higher oil consumption.
4. If rings are leaky at idle, higher crankcase pressure and blowby gas forces dominate, diaphragm opens wider, blowby flows with greater velocity into the intake, separation efficiency drops and oil consumption climbs.
5. If vacuum is weak and rings are leaky, there is a perfect storm for excessive oil consumption.
6. If CCV diaphragm is stuck open in addition to weak vacuum and leaky rings, this may be the worst case scenario.
7. As engine load increases, (increasing rpms), vacuum force on diaphragm drops and force from crankcase pressure dominates, keeping the diaphragm mostly open.


Is it reasonable to expect the spring tension on a new CVV to provide a greater force than that of a used CCV - which puts an older car with lower vacuum and higher blowby from worn rings at a disadvantage - increasing oil consumption as other have reported after changing the CCV?

M54B30 was designed with looser rings (results in higher blowby) and larger intake (increased vacuum) compared to my M52B28.

While the absolute pressure and vacuum values on the M54 may be higher, should that change the balance of the vacuum and blowby forces at the CCV?
From an oil separation perspective, I can see the M54B30 consuming more oil due as a result of its higher crankcase flow, thermals, etc as covered in the MIT paper. Just how much more should we expect and accept?

Note that my M52TU mates to a 4 speed transmission and runs at higher rpms for the same MPH as the 530i - should this increase M52TU blowby gas flow?

Does the new dipstick tube design change the vacuum-pressure balance?

I think I just proved to myself why I need to take the 530i to a shop for tests if the new (third) CCV does not decrease oil consumption. (BTW I may not get to this until the spring for a number of reasons)

Any suggestions of what range of numbers are normal for compression, crankcase pressure, vacuum etc for the M54B30?

Mahmoud Hussein

[02Pilot]

Be forewarned, this is going to be long.

When last I left this problem, I was waiting on time to tear in and do some testing. Well, the time didn't appear, so the car went to a local indy I've known forever to delve into things. We talked about it extensively beforehand and came to the general conclusion that if testing didn't turn up anything significant the valve seals were the most likely cause. Leakdown testing might not prove definitive; assuming the valves themselves were sealing tightly, it might be hard to determine if the seals were leaking or not.

Nothing showed up. Noting that he had an E46 with similar symptoms, and that several customers have this issue as well, we were both anxious to see if the valve seals really were the problem. Knowing the potential issues with removing the head on the M54, I was none too keen on doing this the old-fashioned way. He is always up for a challenge, and pretty good at fabricating tools, so he built a rig to compress the valve springs in the head so that the seals could be replaced without removing it. Still time-consuming, but better than the alternative IMHO. Obviously, this precluded measuring the valve guides for wear, but manually and visually checking for wear in them during the seal replacement did not indicate anything obvious. The seals themselves were rock-hard, which seemed to suggest that we were on the right track.

Car buttoned up and off I go. Initial indications were good: still a tiny bit of smoke in high manifold vacuum situations, but we both figured this was likely to be residual oil in the intake manifold. On the positive side, engine braking was notably stronger, suggesting that the cylinders were sealing better than before.

A few days later, I drove out to a buddy's place some 60 miles away on a very cold day (4degF in the morning, rising to the mid-teens). Drive out was fine, but on the drive back I saw smoke again on deceleration. Bugger. I surmised that the CCV diaphragm may have frozen and ruptured due to the cold. I changed out the CCV (I'm getting way too good at this job): no dice, still smoking. Talked to the shop again, they said drive it a little to remove any residual oil potentially introduced by the CCV failure (if it did in fact fail).

Still no good. Engine braking remains stronger than before the valve seals were replaced, but I'm still using oil at a pretty significant rate (maybe a bit less than before the seals), and it's always on deceleration (high manifold vacuum), never under load. I'm running out of ideas and getting well and truly frustrated when I stumbled across a mention on another forum, in a thread unrelated to my particular issue, of something called "ring flutter". I'd never come across this, and apparently it is not a well-understood phenomenon, but it seems to occur when there is a pressure differential between the forces acting on the top and middle rings, causing the top ring to lift off the bottom of its groove and disrupting its ability to seal. Hmmmmm.....

I figured I had nothing to lose at this point. If the problem was actually related to ring flutter, high vacuum in the cylinder and very low vacuum in the crankcase would be the most likely time for it to occur. If I could introduce more vacuum into the crankcase, especially when manifold vacuum was high, perhaps I could stop the rings from fluttering.

The obvious method would be a direct line from the valve cover vent to the distribution piece on the manifold, perhaps with a catch can inline. Rather than go nuclear right off the bat, I decided to see if I could do something a bit less radical. Knowing the CCV configuration all too well at this point, I knew there was an unused small-diameter port above the oil separator portion of the valve but before the diaphragm. I found there is also a similarly-sized unused port on the back of the intake manifold. I grabbed some appropriately-sized hose and connected the two. Vacuum gauge connected at the valve cover shows that vacuum builds slowly, topping out at about 5inHg at idle. Everything runs normally otherwise. Time for a test drive....

The car drives normally; you'd never know the crankcase was under considerably more vacuum than designed. There's a reason for this: with the throttle plate open at all, the vacuum drops down to almost nothing, i.e., back where it was designed to be. But close the throttle and it rises, eventually reaching about 10inHg on a long descent at about 3000rpm. And the oil smoke after that descent? GONE. Zero. Nothing.

Now, I haven't had this set up long enough to say that it's completely eliminated oil consumption; it may not. But it has certainly reduced it considerably. And I can't say with certainty that the problem is indeed ring flutter, but having eliminated the CCV and the valve seals from the equation, I can't think of any other explanation.

There are a couple of caveats. I would not perform this modification without making absolutely sure that the hose from the valve cover to the CCV is completely unobstructed, as a blockage here could easily cause oil to be drawn up the dipstick tube and into the intake, albeit through a tiny orifice. Similarly, I would not do this without replacing with the updated part or modifying the dipstick guide tube to eliminate the narrow sleeve through which the oil drains back into the pan on the old design.

I must also mention that another poster on these forums, Poolman, has done something similar, although he built a full catch can setup rather than modifying the existing system. He reports similar results, though without explanation as to why (at least as far as I can find). His research encouraged me to go in this direction (eventually).

I'll post back here when I have some better data on consumption, but finally I feel like I've made some progress on identifying and resolving this extremely frustrating issue.

[bmwdirtracer]

Wow. 02 Pilot, mahmoudh, Jim, I'm very impressed; I was going to cry "Uncle" with the MIT papers, but it was all just too interesting, because you guys really followed through with depth of knowledge, and I'm beginning to see light.

I'd like to repeat one thing here, though, for posterity. I've talked to three BMW masters in Hawaii about this: every single one reports the same thing as I've experienced: zero failures of the valve, and never increased oil consumption or smoke related to the ccv system or valve on a six cylinder engine. Every single failure has been a major intake/vacuum leak, manifesting as fuel/air issues, caused by a hot-oil rotted oil return hose to the dipstick tube, right at the valve nipple (I did another one of these, just today)/

I personally find this very weird; because I fully recognize the frequency of these issues on the mainland. Particularly with reference to the difference in consumption vs. temp referred to above, I wonder if there's some correllation beyond that we're even considering, caused by occasionally freezing temps. Certainly, I understand the freezing condensation failure mode: but perhaps we're dealing with a rubber durometer issue of the ccvalve's diaphragm, instead?

It just seems very odd, to me, that I can account for several hundred oilreturn hose failures, between myself and the abovementioned pros in Hawaii, but never a ccvalve failure, or even oil consumption as the complaint. I'm still waiting, and testing everything I can with my manometer...

Nonetheless, I really like the extra vacuum tube you did, 02. Are these ports open and capped, or do they need drilling?

Mainly, I'm just going to watch this very interesting thread, with a hope to finding out whether it'd pure dumb luck that I can't find a similar issue in this very mild climate.. After all, I may be turning wrenches on the mainland, in the not-too-distant future, because autocross isn't enough for me.....