Parts Catalog Accessories Catalog Tech Info Tech Forums
 
  Search our site:    
 Cart  | Project List | Order Status | Help    
  We salute a legendary member of the community.
Thank you for selflessly sharing your wisdom with us...
Go Back   Pelican Parts Technical BBS > 1- Porsche Technical Forums > Porsche 911 Technical Forum

Reply
 
LinkBack Thread Tools Rate Thread
Author
Thread Post New Thread    Reply
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Leak Down Numbers

I decided to post a new thread with my leak down numbers rather than add them to my previous thread about Supertec.

While warming the motor before the test I used a "ticking rod" to isolate the knocking noise to the valve train, and the noise was concentrated on the righthand bank (cyls 4-5-6).

So here we go:

Cyl # %leakage
1----------11%
6----------22%
2----------10%
4----------6%
3----------10%
5----------21%

Obviously cyls 5-6 are a problem, and where most of the noise is coming from. So the way I see it I either have 2 broken intake valve springs (intake is where most of the leakage was coming from) or I have 2 bent valves.

My Question, what are the symptoms of a bent valve? AND if this is broken springs, can someone address the procedure for removing and replacing these?

Thanks!
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 08:25 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #1 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
bumpity..

Nobody can tell me how to replace a few valve springs?
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 11:23 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #2 (permalink)
Registered User
 
na2ub's Avatar
 
Join Date: Sep 2001
Location: Columbus, OH
Posts: 918
Way above my head. Did you try the engine rebuilding forum?
__________________
Doug
79 SC Targa, 22/29 torsion, ER revalved bilsteins, ER rubber bushings, carrera cooler
Old 03-05-2004, 11:27 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #3 (permalink)
Registered User
 
chrisp's Avatar
 
Join Date: Jul 2001
Location: Connecticut
Posts: 1,942
Get Wayne's rebuild book if you have access to one. Not to outdo our host but if you need one urgently barnes and Noble may have it. They stock them at the one near my house (and I don't live in a Porsche-centric community). Perhaps they do at a B&N near you.
__________________
'83 SC
Old 03-05-2004, 11:30 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #4 (permalink)
Registered User
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Doug,
One of the best features of the cylinder leak test is that you can listen to where the air leakage is. You can listen in the intake for intake valve leakage. You can listen in the exhaust for exhaust leakage. You can listen in the oil filler for leakage past the rings. You can listen under the engine for head gasket leaks. These are subjective “measurements” but will provide a lot of useful information.
Repeat your test and report back to Pelican.

Remove your valve covers. Look at the valve springs. This may take some complicated lights, mirrors, scopes, etc. While you have air to the cylinder, tap the valve adjusting screw with a soft hammer. If there is something on the valve seat, this may dislodge it and the leakage will come down.

(One of the procedures I have always done is to not remove the spark plug until I’m at TDC compression and ready to take the measurement. This prevents some carbon, etc. from falling on the exhaust seat. Also, always rotate the engine CW, never back-up.)

While you have the covers off, check the cam timing and valve clearance.

If you don’t find something wrong. Go drive it on the highway, put a load to it, and repeat your measurements.

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 03-05-2004, 11:51 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #5 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Just as a reminder, a persistant knocking sound is what prompted the leakdown procedure. I did take notes when doing the test as to where the majority of noise was coming from, and on cylinders 5-6 it was definately the intake valves. I will take your advice and tap on the adjuster nuts on those 2 valves and retest.

"One of the procedures I have always done is to not remove the spark plug until I’m at TDC compression and ready to take the measurement. This prevents some carbon, etc. from falling on the exhaust seat. Also, always rotate the engine CW, never back-up"

I'd read that in another post of yours I believe, and thats exactly what I did. And I replaced each plug as I went around the firing order, and I did the whole thing twice while engine was warm.

I'm hoping this is a couple broken valve springs, how can I confirm those are broken or not? Will they ahve a different sound when tapped with a hammer?

Thanks!
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 12:10 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #6 (permalink)
 
Now Available for Ordering:   101 Projects For Your BMW 3 Series 1982-2000  [more info]
Registered User
 
Join Date: Jul 2000
Location: So. Calif.
Posts: 16,853
I'm pretty sure one of the two valve springs should have enough tension to hold the valve shut for a static leakdown test. Where is the air leaking from?

On our engines, you're more likely to have bent valves along with broken rocker arms. Pull the valve covers and take a look.

Sherwood
Old 03-05-2004, 02:39 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #7 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
The air is "mostly" coming from the intake valves on cyls 5-6. I've got the covers off, and I can see the springs, but have no way of knowing if they're broke or not. The rockers are fine, which is what makes me think (hope) I don't have 2 bent intake valves. But I really have no idea what 2 bent valves would sound like, both in exhaust note and engine noise (knock?).

Thanks..
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 02:55 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #8 (permalink)
CEO, Pelican Parts Inc.
 
Wayne at Pelican Parts's Avatar
Check out this article (experted from the Engine Rebuild Book):

http://www.pelicanparts.com/techarticles/mult_Engine_Rebuild/mult_engine_rebuild-1.htm

In summary, the leakdown test is only one symptom - I can't recall what you said in the other thread, but I would at least do a compression check as well before concluding that there's a big problem.

That said, these leakdown numbers are not good. You, of course, want to double-check your test to make sure that the cam is not slightly opening any of the valves while you're performing the test.

Also, if you do a compression test, make sure that you turn the engine over with the throttle wide open...

-Wayne
__________________
Wayne R. Dempsey, CEO, Pelican Parts Inc., and Author of:
101 Projects for Your BMW 3-Series101 Projects for Your Porsche 911How to Rebuild & Modify Porsche 911 Engines101 Projects for Your Porsche Boxster & Cayman101 Projects for Your Porsche 996 / 997
Coming in 2014:
• 101 Projects for Your MINI Cooper
Old 03-05-2004, 03:09 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #9 (permalink)
Registered User
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Doug,
I know it’s difficult but try depressing the valves by hand at TDC. Perhaps you can tell if 5&6 are different from #4. That is easy to do when the engine is out of the car; I’ve never tried it in the car.

All the “big iron” (NASCAR, etc.) have valve spring seat pressure testers where they load the spring (in situ) and see how much force it takes to lift the valve off the seat. They have the same issues where they can’t easily see if a spring is broken and the piston will try and help close the valve. Perhaps Wayne can engineer such a tool for 911s - hint, hint.

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 03-05-2004, 03:17 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #10 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Ouch..

Grady,

That hurt the ole hand

I put each cyl 4-5-6 at their respective TDC's and the springs felt pretty much the same (could only compress them about 1/4").

One thing I had done just before this started was change the plugs. Is it possible I knocked some carbon down into the valve seat? That would explain the higher leak-down numbers, but not the knocking sound... I don't think.

Wayne,

I've read those articles of yours before I'll admit I don't have your Engine Rebuild book (yet...) but I have been using your 101 Projects book quite a bit.

Based on your article though my mostly 10% numbers don't sound too bad, just the 2 clinders at 21-22% seem problematic, correct?

Keep 'em coming guys!
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 03:59 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #11 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Also, at TDC the rockers seem to have a little side to side play on their shafts, is this normal? If I rattle them back and forth it seems I can almost make the noise I'm hearing when the car is running...
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-05-2004, 04:05 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #12 (permalink)
Registered User
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Doug,
If something fell into the combustion chamber, it would be the exhaust that would show leakage.

There is a spec on the side clearance of the rocker arms. I think that would have to be way out of spec to be an issue.

Have you checked the valve clearance and cam timing? Compare left to right. Remove the rockers and inspect (one at a time). While rockers are out, inspect the right cam lobes. Be careful to not get dirt in the engine, particularly rocker shafts.

It appears your big issue is there is something different from left to right banks. Are you certain that is true?

I have two (now antique) Sun Electric Corporation Model 228 Cylinder Leakage Testers. I compare them against each other regularly to insure my measurements. While they are old, I have had consistent readings over some 35 years. On perfect engines they measure 1.5-2.5%, with no cylinder out of that range and all leakage past the rings. Really good street engines are 2-4% and are typically all within a 1% range. Sorta normal street engines are 2-10% and very all over the range. Only now do I see leakage past the valves. These engines will usually settle down if run on the highway and a full power run up a hill; back to 2-5% and a 1% range. Engines that are in need have 4-16% and don’t respond well to driving. This means that some cylinders improve and others degrade in leakage, measurement is different each time. If one cylinder is dramatically different, that indicates a problem. Numbers above about 12% (where the others are 2-5%) I start looking for a burnt or bent valve, or broken rings. At regular 20%, there is no question, pull the heads.
Remember, these are the numbers from these old (but not too lame) Sun testers and may not be the same numbers as other cylinder leak testers. Don’t rely on a single measurement, repeat several (many) times. Go aggressively drive the 911 between measurements. Record everything. Statistics counts.

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 03-05-2004, 05:28 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #13 (permalink)
Registered User
 
Join Date: Jan 2002
Location: Long Beach CA
Posts: 29,965
I am far from an expert, but I wonder why broken valve springs are being considered as the problem. Under a leak down, even the slightest pressure of the spring holding the vavle against the seat, and helped by the pressure in the cylinder, should be sufficient to determine the source(s) of the leak. The danger of broken valve springs is that the valve may not follow the cam and close fast enough at higher RPMs. This ususally means the piston hits the valve as Grady said. And, it gets bent. While valve springs may be the cause of this senario, the problem now is the valve itself.

My point is that while the springs need to be in good shape, they by themselves don't cause a bad leakdown. I doubt that at this point replaceing valve springs will help this engine.
Old 03-05-2004, 06:41 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #14 (permalink)
Registered User
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Doug,
What is new today?

Wayne is right on about using an automotive stethoscope. He is correct that a cranking compression test may give useful additional information.

Some manufacturer makes a cranking compression gage where part threads into the spark plug hole and can be tightened and then the gage connected. This allows one person to get good readings by himself.

When doing a cylinder leak test, never try and hold the engine from rotating, just get it exactly at TDC compression. You can break something – like your hand. It can be powerful enough to unscrew the pulley bolt.

I wouldn’t act on any of this information until I had driven the car in between the tests. That is unless you feel the 5-6 “tick” portends some immediate failure.

Milt,
I agree. The leakage difference between the left and right banks may or may not be related to the 5-6 “tick.”

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 03-06-2004, 05:56 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #15 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
All good info here... I'm taking a break from it this morning and I'm driving my 356 to an RGruppe get-to-gether. When I get back I'll take the 911 out for a long drive and see what happens.

I borrowed the tester I'm using from my friend who owns a shop, its an ancient Snap-on unit, looks about 30yrs old and its mounted in a box. As I said before, the readings seemed pretty consistent as I went around the crank several times.

I'll post again tonite after I drive her a bit and retest.
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-06-2004, 07:45 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #16 (permalink)
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Well I took the car for a long drive yesterday and noted a couple things:

1) The knocking is still there
2) The knocking is mostly under acceleration.
3) The car still pulls hard to redline

After the drive I re-did the leak-down test and the numbers were only slightly lower, in other words the bad cyl were now 19-20% instead of 22%.

I drained the oil and took the lower valve covers off, nothing unusual in the strainer screen, with the valve train, or in the magnetic plugs themselves.

So, I'm at my witts end with this thing. I'm pretty much ready to give up and take it to a shop and have them start tearing it down. What a drag...
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-07-2004, 08:20 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #17 (permalink)
Registered User
 
Join Date: Jul 2000
Location: So. Calif.
Posts: 16,853
And the compression test readings are what?

And the ignition timing is set at what?

Did you try rotating the cylinder head studs with a wrench?

Sherwood
Old 03-07-2004, 10:53 AM
  Recommend this thread for the PelicanWiki    Reply With Quote #18 (permalink)
 
Now available:  101 Projects for Your Porsche Boxster!
Registered User
 
Spud's Avatar
 
Join Date: Feb 2003
Location: Livermore, CA
Posts: 530
Sherwood,

I do not have a compression tester handy, I'll pick one up and give it a go this week. The head studs were all tight. Timing and dwell are right on.

I'll post here again after I test compression.

Thanks,
__________________
Doug
1964 356C - "Olivia"
1968 912 - Slate Grey
Old 03-07-2004, 01:35 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #19 (permalink)
Registered User
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Doug,

In the interest of caution and preserving the engine and your pocketbook – STOP NOW. Knocking under load is a usual symptom of a rod bearing failure.

It can be other, like low octane fuel or something else mechanical but you have the symptoms.

No offence but when you pulled the sump plate, do you know what to look for tell-tail signs of bearing failure? Look in the little area between the case and the edge of the sump screen. You are looking for (and hopefully not finding) little copper flakes. These can range from almost dust size to 2 mm diameter but no more than 0.05 mm thick (or so.)

Can you find a shop that has the oil filter “can opener”? You can take the filter element out of the metal housing and carefully inspect. You won’t find any big pieces but small flakes will get through the sump screen.

I’ve got my fingers crossed for you.

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 03-07-2004, 02:07 PM
  Recommend this thread for the PelicanWiki    Reply With Quote #20 (permalink)
Old 03-07-2004, 02:07 PM
Reply

Thread Tools
Rate This Thread
Rate This Thread:

 


All times are GMT -8. The time now is 07:41 AM.


 
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2011 Pelican Parts - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page
 

DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.