Roadfly Home | Features | Car Review Videos | Car Reviews | Cars For Sale | Used Car Parts Classifieds | Forum | Car Review Archives | Forum Archives Index


Chet Juall
10-01-2001, 03:20 PM
After rebuilding the 2.7L engine from my 1976 911S, the engine
smokes from the exhaust at startup. I installed brand new Mahle pistons and cylinders. A leak down test shows 6% total leakage in all cylinders but most of this leakage is from the valves. There is little ring leakage. The engine has 40IDA carburetors. Can someone explain to me what is going on here ?



10-02-2001, 10:26 AM
It could be oil leaking past the valve seals into the combustion chambers while the car sits. Another possibility, if the smoke is black instead of blue, is a rich fuel mixture at startup. If I'm not mistaken, your car has CIS injection. Is the CO set properly and is the cold start circuit wirking as it should?

Chet Juall
10-03-2001, 08:32 AM
Actually, the car has 40IDA weber carburetors. The smoke
is bluish and smells of oil. I did install valve stem seals.
Why would the oil get past them ?



06-16-2011, 09:57 PM
After living with the car for a few years the smoke at start up went away. I think the cylinder leakage and smoke were unrelated. The 6% leakage was due to using a local machine shop who did not grind the valves well. 3% leakage would be appropriate for a new grind using Mahle piston/cylinders.
I have heard from several sources including a highly regarded Porsche mechanic in New Jersey that startup smoke is common.

It seems that startup smoke is common on some newer 911 models (and Formula One engines) also. The following link shows evidence of these facts.

Keep in mind 911's use a dry-sump oil system. These systems are commonly found on drag racing cars to gain a few horsepower by nearly eliminating the "windage" or drag caused by the crankshaft having to rotate through a viscous sump of oil.

Dry sump systems are also used on road racing cars and on cars such as 911's which are designed to have excellent cornering characteristics.
Dry sump systems help to improve handling characteristics because they allow the engine to be mounted more closely to the road. A wet sump system on the other hand requires a relatively deep oil pan which must hang well below the crankshaft center line and would scrape on the road if the engine were mounted too low and the vehicle was driven over a bump.
When it comes to mounting the engine as closely as possible to the road, the Porsche 911 design is very effective in this regard. The "oil pan" is basically just a circular plate which does not protrude downward beneath the engine at all. When the engine is close to the road, so is the transaxle and when the two are taken together, they add up to quite a bit of weight.
Now lets return to the issue of startup smoke. With a dry sump system there are more than one oil pumps in the crankcase. One pump is used to drive pressurized oil to the bearings and valve train and one or more other pumps are used to scavenge or remove oil from the crankcase and return it to a reservoir tank which in the 911 is mounted in the read fender well. When the engine is running, the scavenging pump(s) should keep most of the oil out of the crankcase and the reservoir tank should be nearly full. When the engine is turned off there is a tendency for the oil to seep back from the reservoir tank into the crankcase. If the car sits long enough, this seepage will continue until the level in the tank and crankcase are the same. In this case there will be a significant amount of oil in the crankcase and the level of this oil I think will be higher than the lowest part of the cylinders. Because the 911 engine is a horizontally-opposed boxer type engine, the cylinders sit unusually low, close to the crankshaft center line. For these reasons, the oil may seep past the rings to some extent when the car sits for a period of time.
On my car, it is my feeling that the smoke was due to oil getting past the piston rings and not the valve stem seals. This possibility is consistent with the fact that the smoke diminished as the engine became older (and the rings seated more fully).
In contrast, I would expect oil leaking past the valve stem seals and through the guides to increase in quantity as the engine aged.

If in fact the source of 911 start up smoke is oil passing the rings, another factor which could effect this behavior is the position and size of the piston ring end gaps.
I remember complying with Porsche's recommendation for the position of piston ring end gaps. The size of the end gaps however were set by Malhe and I assume were okay. (Unless Malhe sells the piston and cylinder sets which do not meet Porsche's specs to the aftermarket). If the ring end gaps were inadvertently positioned at the bottom of the cylinders and/or the end gaps were too large, oil would get past them more rapidly when the car is turned off and sitting.

A common belief for older 911's is that the oil pumps are indestructible and do not need to be replaced on rebuild. I'm wondering if even though these pumps will continue to make adequate pressure and continue to scavenge effectively, the internal clearances increase substantially over time. If they do, then on a car with an old oil pump, oil could seep past the pump more rapidly when the engine is at rest. The clearances between the gear-like members of these pumps as well as the clearances between each gear-like member and the pump housing could increase over time.

The issue of oil getting past the rings when the engine is turned off is not an issue on your common V8,V6, or typical four cylinder engine because the piston rings always remain above the oil sump in the oil pan. i.e. gravity keeps the oil away from the rings.

As ADias says on,

"This (smoke on startup) is common behavior for boxer engines for the last 50 years."

The engineer in me want to know exactly "Why"?

A camera with a light and fiber optic cable placed down the spark-plug holes would likely have answered the question, such a device however was beyond my budget.

Chet Juall

06-16-2011, 10:05 PM
Here is another quote from 911dev on Rennlist

"Mine seems to smoke when it has sat for a couple of days or so. I always allow for a cool down before turning off the engine and will allow for an extended idle -actually elevated a little as the manual reads. Anyway, it will still smoke as it's inherent in the boxer design. No biggie. Any oil burned at start up is so minimal it doesn't even change the level reading."

06-16-2011, 11:18 PM
More quotes from on the topic of startup smoke.

"smoke anyway, upon startup is a characteristic of the flat-6 engine -- at least the Porsche engines."

"mhm993 is correct. You'll notice your "smoke" is blue because it's the oil. Very common with boxer style engines. An occasional puff on start up every now and then is nothing to be concerned about."

"if my car sits for a few days i get a little smoke at startup . no big deal."

"My '03 only has 20k on it and it does it now and then - just for a few seconds at startup. Usually when it hasn't been driven in a while. It concerned me at first, but look on any Porsche discussion board out there, and you'll quickly see (as I did) that it is very common for the Porsche flat six to do this."

Roadfly Home | Car Reviews | Forum Archives Index