How much oil in the crankcase?

[rcaradimos]

Where would you say the oil level would be in the case and would it rise after sitting for a few weeks?

[cgarr]

Upper part of the oil return tubes.

[JFairman]

With the top valve cover off shine a flashlight down in there and you'll see where the oil has come up to after sitting overnight or longer.

Usually around half way up the lower valve cover when I've done that.

[J P Stein]

All of it.

[911 tweaks]

what yr car and/or engine...??

[rcaradimos]

3.2l

Think it could be this high?
I'm getting some oil in my exhaust manifold and wondering if it seeps in from the cylinders (stuck oil rings) or up the exhaust valve guides. (bad stem seals / worn guides)
Car starts with small puff now and then, recently it smokes once the exhaust gets hot 3 min. after start-up... then it burns off. No smoke when hot and very little in between shifts. My thoughts are bad stem seals on the exhaust causing the smoke. Pulled all the spark plugs and they are clean, thats why I suspect oil in the exhaust manifold.

[rcaradimos]

If oil levels are this high I would think oil would find its way in the exhaust manifold from a bad stem seal right?

[911 tweaks]

sure would...a/o the valve head has carbon built up on it & it too, possibly, is adding to the issue...I tend to lean to the latter part of my comment "IF" oil is in the h/e's...

[911 tweaks]

pull the h/e's off...check them out...clean out any & all oil that may be in them & inspect ALL of the exhaust valves for seating properly & side to side wiggle = possible worn out valve stem seals...

[Walt Fricke]

Porsche was of the opinion that startup puffs of smoke in the 959 were due to water condensing in the exhaust and then steaming off on startup. This from the 959 owner's manual.

I don't use exhaust seals on my track motors, and they don't smoke on cold start any more than my SC does. And the track car often sits, alas, for months at a time between startups.

In theory, the oil in the tank should flow out over time until the oil in the case rises to equal the tank's level. The only time I've had issues approaching this was when I had the front of the track car propped way up for weeks repairing things. The top oil level in the cylindrical 4 gallon front oil tank is ratherhigh to start with. Wasn't exactly hydrolock, but made a lot of smoke for a while. Luckily the pump is pretty high volume.

I don't recall seeing a nice cutaway of an oil tank to the same scale as the fine engine cutaway you posted. With one a guy could perhaps do some measuring and juxtapose the two to get some idea of the hydraulic head and so on.

[lucittm]

These engine are "dry" sump, no? I would estimate that the Porsche engineers designed the engine and lubrication system so that the oil level would never go above the oil pump regardless of the year or model.

Can anyone estimate how many quarts (liters) that would fit in the bottom of the engine case to the top of the oil pump? It might be a little different for the early engines that had the sump plate, but I would guess that all the later engines up to the 996 would be pretty close to the same.

Mark

[Walt Fricke]

I don't think Porsche designed the 911 to have any particular level of oil in the case sump. The dryer the sump, the less parasitic drag. The oil pumps have always had a substantial scavange capacity, and over time that has only increased. The scavange side of a GT3 pump looks to be almost as large as the whole of a 1966 pump.

So the design for running is for nearly all the oil to be in the tank, with some being cooled or piped around, and some lubricating and squirting out to be recurculated through the tank.

What I suspect was not as a result of direct design was the slow migration of the oil from the tank of a stopped engine into the sump. I believe most if this is a result of oil leaking out the ends of the pump, where shafts without seals provide a very small, but still available, opening. But I have never mocked up an oil pump with an oil tank to test this theory. There otherwise really isn't much of a way for oil to travel from tank to sump.

But travel it does, hence the mistake most of us have made back when of reading the dipstick with the engine cold, concluding we were a couple of quarts down, and ending up overfilling and making a bit of a mess.

Or the other mistake of not putting a large enough pan under the sump drain when changing oil on a cold engine.

[rcaradimos]

Changing oil hot, about 20 min. after shut down. I get about 3 quarts from the case and 5 from the oil tank (total actual measurement 8 quarts) I have not measured the two separately. I have never changed the oil cold so I have no idea what would come out of the case cold? 6,7 quarts?, raising the question of level in case on cold start and if oil is above the ex-guides (@ piston end ) seeping into exhaust. I have the front fender cooler... so I would imagine from T-stat closed (rear wheel well), the two oil lines from there to the cooler, including cooler capacity would be holding 3 quarts min. ???

[E Sully]

Bob,
Is your engine high mileage? The 3.2's have a tendancy for worn valve guides. Mine only had @ 65,000 miles and needede the guides replaced. The newer guides are better materials so will hold up better.

[304065]

Originally Posted by Walt Fricke

I don't think Porsche designed the 911 to have any particular level of oil in the case sump. The dryer the sump, the less parasitic drag. The oil pumps have always had a substantial scavange capacity, and over time that has only increased. The scavange side of a GT3 pump looks to be almost as large as the whole of a 1966 pump.

So the design for running is for nearly all the oil to be in the tank, with some being cooled or piped around, and some lubricating and squirting out to be recurculated through the tank.

What I suspect was not as a result of direct design was the slow migration of the oil from the tank of a stopped engine into the sump. I believe most if this is a result of oil leaking out the ends of the pump, where shafts without seals provide a very small, but still available, opening. But I have never mocked up an oil pump with an oil tank to test this theory. There otherwise really isn't much of a way for oil to travel from tank to sump.

But travel it does, hence the mistake most of us have made back when of reading the dipstick with the engine cold, concluding we were a couple of quarts down, and ending up overfilling and making a bit of a mess.

Or the other mistake of not putting a large enough pan under the sump drain when changing oil on a cold engine.

Well, wasn't the oil bypass mod in 1976 a conscious decision by Porsche to lower the oil level in the sump? Dumping the overpressure into the sump vs. recirculating it back into the pump inlet was intended to reduce oil foaming and windage, and also allow the pressure side to be made larger relative to the suction side.

If we visualize the oil flow through the "S" hose from the tank, it goes into the solid tube in the oil cooler and then into the pressure side of the pump. In the pressure side, the oil runs up against the gears in the pump. Remembering that the pump is a positive displacement type, there's no way for the oil to get around the gears, except through the clearances between the gear teeth and around the ends.

The same small clearances that make the pump a positive displacement type also limit the oil flow into the case with the engine off.

Now, there's probably a couple liters of oil up wandering around the heads and oil return tubes at any given time. When the engine sits some of that will leak down due to gravity, but I doubt it's able to fill the case up very far.

[J P Stein]

After sitting for a couple weeks or more, *all* of the oil...9-10 qts worth came out the sump on my old car. It has to be coming back through the scavenge side of the oil pump. The pressure side is fed after oil passes thru the thermostat much higher on the engine.

When restarting after its sleep, I crank the engine over for 30-60 sec to get oil back into the tank before attempting to fire it up.

[Walt Fricke]

I will have to beg to differ with our esteemed moderator on this.

I believe the tolerances around the oil pump gears, both between gears, between teeth and the housing, between the ends of the housings and the end faces of the gears, and between the gear shafts and their bearing holes through the housing ends and separators, are sufficient for oil to leak through. At what rate I don't know.

It can exit on the flywheel side through the shaft openings there, and from the other side after passing through the intermediate plate or whatnot on into the pickup tube.

After passing through the mesh and around the gears, I once thought perhaps it could clamber up into bearings, but it has to go up too high to do that.

It could leak past either of the oil pressure pistons. The vertical one would just leak it back into the inlet, so that wouldn't count if it could get up that high. But the safety relief just dumps right back into the sump.

But from my experiences with trying to stop oil from leaking elsewhere, I am convinced that it is slippery stuff, and zero is the only tolerance which will stop it. None of these parts need to be leakproof.

A guy with a pump apart could measure its tolerances, especially from gear to case at the teeth, easily with feeler gauges. Since most of us are baffled on how to tell if a used pump is good, maybe someone or some discussion has this information already? Then one might be able to calculate/approximate the flow rate of 50 weight oil at room temperature through openings of such a size.

Were I possessed of even more free time and curiosity than I already have (on showing someone my latest fitting with small lightening holes, my friends exclaimed: "Walt, you have to get a job!"), I could figure out a way to do an experiment. I have the oil pump from a 1968 or so engine. Not likely to be useful to much of anyone. I could make a jig - maybe from the mag case with pulled timecerts I keep for who knows what use - to hold the seals, and rig a gallon jug a foot or so above the pump, fill with oil, and watch what happens.

What I do know experientially is that the level in the oil tank drops after the engine is shut off. Where does that oil go? Since it enters the tank from the oil filter up top on the tank, it can hardly go back that way. That leaves only the S tube directly into the pump.

I've never measured this, but I don't believe that the dribble down of oil from the various surfaces inside the case and up in the heads can amount to more than a pint, if that. Not all that much oil is sequestered in the heads below the level of the return tubes.

Though I admit that when I need to get into the exhaust valves outside of an oil change, I jack the car way up on one side, and then on the other, to avoid the mess. So maybe a) the oil level is higher quicker than one might think in the sump, or b) oil drains from those areas slower than I am asserting it does.

Myself, I think what is important about valve guides is to replace worn ones (and worn valve stems, too) before more problems than just a few ephemeral puffs of smoke show up.

To get back on Bob's topic, I think he needs to measure oil consumption. If it is low, I think that is some indication that his guides may still be in OK shape. And if it is down in the 500 mile per quart range, maybe he needs some serious examination of his guides.

That's my case, and I'm sticking with it.

[304065]

Gear backlash is .03 to .08mm. End play is .035 to .1 new .2mm is the wear limit. Please mark this in evidence as "Short Wheelbase Guy's A."

The clearance on the oil pressure and overpressure pistons is probably (guess) 0.008mm. "Short Wheelbase Guy's B."

All that oil is coming through orifices that small in a timeframe of weeks? I guess so.

[Walt Fricke]

Thank you, counselor. Without objection they will be so entered.

Long ago I read about Ignatius Donnelley, a post-Civil war attorney and politician from Minnesota. A multi-talented fellow, but he set about "proving" to his entire satisfaction that Atlantis existed (maybe even where it was). I took that as a warning about too much reliance on legal reasoning in matters of science or engineering.

Still, when one eliminates possibilities, suspicion mounts against possibilities not yet disproven.

When you turn off your engine, there will be oil in the oil filter and the return line to the tank which feeds it, and it is basically open to atmospheric pressure (because the tank into which it will eventually drop is vented). If no external thermostat, that oil has a hydraulic head back into the scavange pump. But it, too, would have to exit past those gears. And I don't think there is all that much of it.

With an external thermostat things get a bit more complicated, with more oil/pipe friction, humps for oil from front coolers to have to get over, thermostat originally open, but closing and rerouting, etc. But I'm pretty sure that most of the oil in the external cooler circuit pretty much stays put after the pump stops. At least there is plenty to deal with when one needs to remove a front cooler or remove one of the long oil lines.

[304065]

Walt: I agree-- where physics is concerned it's a slightly higher standard of proof.