Engine squeal and vibration - crankcase ventilation / air-oil separator?

[SpacemanSpiff]

Hello -

I have an odd situation I can't find solid info about online, and my Porsche-specialty mechanic is stumped. Can anyone here help?

 

Since December/January, I have had an increasingly prominent squeal/whine sound coming from the rear of my 997.2. Initially, it only occasionally occurred quietly when I actuated the clutch pedal. Now, it's louder, sometimes whines for short periods or cycles (wooo...wooo...wooo - here's a video) without clutch activity, and I can make it happen nearly on demand holding the clutch pedal about halfway through release.

 

Background:

• 2009 911 Targa 4, manual, 43k miles (19k miles until 2017 when I bought it, now a daily driver in western WA)
• No CEL illuminated.
• The clutch was replaced at the end of October due to pressure plate failure.
• 40k "major" service performed in mid-November.

 

Diagnosis info:

• My mechanic found that a crankcase vent line vibrates noticeably in sync with the sound. #2 on this parts diagram.
• That line apparently connects to the air-oil separator, and he suggested the AOS may be failing (squeal caused by the diaphragm inside leaking?).
• I found an article by Tony Callas noting that a failing AOS can honk and how to test it. My mechanic tested the crankcase vacuum and found it "within spec".
• We replaced the vent line in case it had a leak. He thinks the sound is a bit quieter, but I don't notice any change in volume, and it still occurs as often.

 

Neither of us is keen to spend time and money replacing parts just to try them. I can't help but think the clutch replacement or 40k service is related, since this started not long after.

 

Perhaps the AOS really is failing? The Callas article claims 996/997 AOS failure often doesn't show smoke (unclear whether he means 997.1 or 997.2). Does anyone have more solid info on diagnostics?

 

I was once able to get a tiny rpm increase (50?) visible on the tach while holding the clutch pedal halfway on release while sitting in my garage. How might clutch actuation cause an rpm increase (electronic smarts to prep the engine for drivetrain engagement?) and is that even relevant?

 

Your thoughtful replies are much appreciated! I'd love to not spend $900 on the AOS without having a better idea what's happening.

[dammad]

I believe I've read that if the engine is running and you remove the oil cap, then the engine should stumble because vacuum is lost. However, if the AOS is leaking, then the engine note won't change appreciably because of the much higher vacuums being generated. Also, if the AOS is leaking the cap should be hard to remove at idle. 

[JFP in PA]

9 hours ago, SpacemanSpiff said:

Hello -

I have an odd situation I can't find solid info about online, and my Porsche-specialty mechanic is stumped. Can anyone here help?

 

Since December/January, I have had an increasingly prominent squeal/whine sound coming from the rear of my 997.2. Initially, it only occasionally occurred quietly when I actuated the clutch pedal. Now, it's louder, sometimes whines for short periods or cycles (wooo...wooo...wooo - here's a video) without clutch activity, and I can make it happen nearly on demand holding the clutch pedal about halfway through release.

 

Background:

• 2009 911 Targa 4, manual, 43k miles (19k miles until 2017 when I bought it, now a daily driver in western WA)
• No CEL illuminated.
• The clutch was replaced at the end of October due to pressure plate failure.
• 40k "major" service performed in mid-November.

 

Diagnosis info:

• My mechanic found that a crankcase vent line vibrates noticeably in sync with the sound. #2 on this parts diagram.
• That line apparently connects to the air-oil separator, and he suggested the AOS may be failing (squeal caused by the diaphragm inside leaking?).
• I found an article by Tony Callas noting that a failing AOScan honk and how to test it. My mechanic tested the crankcase vacuum and found it "within spec".
• We replaced the vent line in case it had a leak. He thinks the sound is a bit quieter, but I don't notice any change in volume, and it still occurs as often.

 

Neither of us is keen to spend time and money replacing parts just to try them. I can't help but think the clutch replacement or 40k service is related, since this started not long after.

 

Perhaps the AOS really is failing? The Callas article claims 996/997 AOS failure often doesn't show smoke (unclear whether he means 997.1 or 997.2). Does anyone have more solid info on diagnostics?

 

I was once able to get a tiny rpm increase (50?) visible on the tach while holding the clutch pedal halfway on release while sitting in my garage. How might clutch actuation cause an rpm increase (electronic smarts to prep the engine for drivetrain engagement?) and is that even relevant?

 

Your thoughtful replies are much appreciated! I'd love to not spend $900 on the AOS without having a better idea what's happening.

 

Connect a digital manometer in place of the oil fill cap and start the car; fully warm, you should see no more than 5 inches of water vacuum signal.  Any higher than that, the AOS is toast.

[SpacemanSpiff]

It turns out the shop did the "is the oil filler cap tough to remove" test, not a manometer:
 

Quote

 

1) Is the oil cap a pain to remove?  Tested as fine / normal.
2) Once removed, does the engine stall?  No, just a minor variation in idle speed determined to be normal.
3) Is there a strong vacuum present at the oil fill tube?  No.

Based on these results the separator was determined to be in good working order without concern at this time.

I didn't have any luck Friday afternoon looking for an official [test] procedure but I will continue the search today.

 

 

dammad - Thanks, I'd seen that on forums, too, but nothing more official or explanative. Tried it the other night and the cap was about as easy to remove as with the engine off. No appreciable stumble, but rpms do rise when replacing the cap. I was also able to get the squeal when doing that. More video!

 

JFP in PA - Thanks for the recommendation. Can you point me to a spec on what I should expect for that engine (MA1.02 / 9A1) at x rpm?

[JFP in PA]

10 hours ago, SpacemanSpiff said:

It turns out the shop did the "is the oil filler cap tough to remove" test, not a manometer:
 

 

dammad - Thanks, I'd seen that on forums, too, but nothing more official or explanative. Tried it the other night and the cap was about as easy to remove as with the engine off. No appreciable stumble, but rpms do rise when replacing the cap. I was also able to get the squeal when doing that. More video!

 

JFP in PA - Thanks for the recommendation. Can you point me to a spec on what I should expect for that engine (MA1.02 / 9A1) at x rpm?

 

Test is run at idle, engine fully warmed up (15-20 min drive), should not see greater than 5 inches of water vacuum signal.

[EMC2]

On 2/19/2020 at 6:18 AM, JFP in PA said:

 

Test is run at idle, engine fully warmed up (15-20 min drive), should not see greater than 5 inches of water vacuum signal.

Just to clarify - I know the post is old but I just came across it;

 

5 inches of water column was for some of the earlier generations.

The 997.2 will have more in the range of 12 to 18inches.

This is based on measurements across several form member vehicles tested and also confirmed recently in one of Jake Raby's videos.

[JFP in PA]

10 hours ago, EMC2 said:

Just to clarify - I know the post is old but I just came across it;

 

5 inches of water column was for some of the earlier generations.

The 997.2 will have more in the range of 12 to 18inches.

This is based on measurements across several form member vehicles tested and also confirmed recently in one of Jake Raby's videos.

 

On the 9X6 and 9X7 Porsches, the engine crankcase vacuum reading is normally in the region of -4.0” to -6.0” H2O (Inches of water). When an oil separator fails, the crankcase vacuum can rise to -10.0” to -15.0” H2O, or even much worse (-40.0” H2O) depending on the extent of AOS failure.

 

On the 9X7.2 and 9X1 Porsches, the engine crankcase vacuum reading is normally in the region of -14.0” to -16.0” H2O (Inches of water). When an oil separator fails, the crankcase vacuum can rise to -20.0” to -25.0” H2O, or even much worse (-30 to -40.0” H2O) depending on the extent of AOS failure.

[Silver_TT]

Good info!  My two cents because I find this stuff interesting, may not be telling you anything you don't already know:

 

DFI cars (not just Porsche) generally run a much higher signal of crankcase vacuum.  As a matter of fact in several cases this was done after the fact.  For example, Audi, who was an early adopter of DFI, bumped up the vacuum signal almost 10 times higher (~5" H20 to ~40" H20) several years after the introduction of DFI in their engines.  This required a new AOS/PCV, ECU update, etc.  This was obviously done to reduce/eliminate oil consumption, help ring seal, lower fuel dilution, etc (these are all inter-related).  OEM's have done all sorts of things to address issues with DFI not limited to changes in oil control rings, vacuum levels, etc.  I personally find this fascinating.

 

Don't be surprised to hook your manometer up to a newer DFI vehicle and find it's running a much higher signal of vacuum than older generation cars.  As both posters show, it is critical to know the "healthy" range.

 

I have been measuring my crankcase vacuum but will replace my AOS/PCV preemptively every 40-50K miles just because the part is so cheap and a failing unit can cause so many issues.  When I just did mine a few months ago I found it was starting to fail but I wasn't getting a CEL.  Guess it's important to remember that every "failure" of a part doesn't always come with an error code.  For some parts IMHO it makes sense to get to the part before it gets to you.

 

DFI cars also control fuel trims much tighter and have evolved to have extremely precise timing and control with ultra fast injectors like the piezoelectric injectors.  In some sense these cars are totally different animals than the port injected predecessors (I'm not necessarily saying they are better).

Edited March 25, 2021 by Silver_TT