Porsche 944 Oiling System Explained

Due to much discussion regarding why the Porsche 944 engine can tend to trash the #2 rod bearing, I took the opportunity to look closely at an engine to see how the oiling system worked.  As you look at the oiling system layout, you can begin to understand how the #2 bearing might be at risk.   And to jump to the conclusion, #2 appears to be at risk for seeing an oil supply with a high percentage of oil with air bubbles in it.  Stay tuned and all will be revealed.

Note:  A picture is worth many words so I’m going to try and minimize words and maximize pictures.  Note that most of the pictures are of an engine turned upside down so don’t let that confuse you. The engine shown is from a 1984 car.

Most of us understand that the oil starts its journey in the sump.  Pooled in the bottom of the sump, it enters the oil pickup tube.

Oil pickup tube with screen. Oil return tube (no screen)

In the above picture, the oil return tube is in the foreground and the oil pickup tube is in the background.  Both tubes attach to the “crank girdle”.  In the next picture we’ll see the passageway that lies under the crank girdle, on the bottom side of the block, that transfers the oil from the pickup tube to the oil pump mounted on the front of the engine.  I have used red wire to highlight the passageway.  This picture is looking at the front of the block, without the oil pump mounted in position.

There are a couple of things to note about the above picture.  The oil pump is not mounted but if it were, it would suck the oil from the passageway and insert it under pressure into the hole in the front of the block where the red wire disappears into.  The green wires are stuck into the oil passageways for the crankshaft main bearings.  We’ll come back to them.

In the above picture the red wire enters the front of the block.  Near its entry point is a round hole.  This hole is normally plugged with a cap.  This hole passes through the length of the block and is the main “oil gallery”.  The following pictures show the cap.  There are caps at both ends of the block so if you remove them both, you can give the gallery a really good cleaning.  I highly recommend this.  The caps are destroyed upon their removal but new ones are like $2 a piece.

Oil gallery cap

Now we’ll look around the side of the block where the housing for the oil cooler (on a N/A engine) and the oil filter attach.  Its hard to show but I have used a red wire to show the hot uncooled oil and a green wire to show the cooled filtered oil.  It is important to note that the flow back into the block is pointing right at the #1 crankshaft bearing, with no turns or changes in direction required.  Conversely and very important, the oil for the rest of the engine has to go around the first of several 90 degree corners to get into the oil gallery.  The second picture shows the fitting for the oil filter.

So let’s go back to that photo of the block without the crank shaft or bearings installed.  There are 5 crankshaft bearings on a 944, a very robust design.  Note that each bearing has a passageway drilled from the bearing back to the oil gallery.  These are shown by the green wires in the photo.  If you could see the other end of the green wires, you would see them sticking into the oil gallery, at a right angle.

Green wires show the drilled passages for the crankshaft bearing oil supply.

Here is another picture with the bearing shell ready to be installed.  You can see how the oil is allowed to get through the bearing and flood the narrow space between the bearing and the “journal” on the crankshaft.  A “journal” is the smooth shiny very hard surface on the crankshaft that serves as 1/2 of the bearing surface.  The other half of the bearing surface is formed by the bearing shell, which is hopefully the wear item that can be replaced during an engine rebuild.

So the oil floods the space between the bearing and the crankshaft journal.  Where does it go from there?  The answer is revealed by a close examination of the crankshaft.  This is a little hard to show because all the passages are internal to the crank.  First, each crankshaft journal is drilled with a hole radially towards its mid-point.  This hole is intercepted by a diagonal hole that heads off towards the rod bearing journal.  There is a hole drilled radially through the rod bearing journal that intercepts the diagonal hole. Note that the diagonal hole is plugged at its end, once drilled, to blank it off and force the oil into the rod bearing journal area.  So the oil goes radially inward from the crankshaft journal, then diagonally outward towards the rod bearing journal, then radially outward to the surface of the rod bearing journal. Whew!  Here are a few pictures.

Note that the crankshaft journal surfaces are much bigger than those for the rod bearings.  From this you might deduce that the rod bearings are more highly loaded due to their smaller surface area.

In the above 2 pictures, I have just run the green wire externally to the internal passages in the crank, to give you an idea of how they run.  The actual passages are hidden.  If you look closely to the right side of the above picture, there is a machined hole.  This is where they inserted the drill to make the passage, then installed a plug.  Also note that if the rod bearing journal is “cross drilled” the hole shown in the above picture extends all the way through the rod bearing journal.  If it is not cross drilled, it only extends halfway through the rod bearing journal.  Regardless of either design, 100% of the oil coming out of the diagonal passageway reaches the bearing.  But being a zero sum game, I don’t see that cross drilling does anything to improve the oil supply, as there are so many passages before it that serve to limit the flow.  Think of it like the interstate leading into Charlotte.  Its 2 lanes north of town and then opens up to 4 lanes nearer to town.  But during the morning rush hour, the 2 lane section is stop and go.  Having the 4 lane section down the road really doesn’t help much because the 2 lane section is the restriction to flow.  Indeed, I have seen that early cranks (83, 84) were cross drilled by Porsche but later cranks are not.

FYI, near the rear end of the oil gallery there is yet another 90 degree branch that shoots straight up, goes through the headgasket, and enters the cam housing to lubricate the cam.  EDIT: A more detailed discussion regarding this flow path has been added at the end of the article.

So finally the oil has reached the rod bearing.  If all goes according to plan, the space between the rod bearing shell and its journal stays filled with a very thin layer of oil at all times and all is well.

Now a diagram would be in nice.  But in words, we see that the oil to the 1st rod bearing has no 90 degree turns and two 45 degree turns.  The oil to the #2 rod bearing has a 90 degree turn to get out of the oil gallery and then two 45 degree turns.  The same is true for the #3 and #4 bearings.  It is worth noting that the oil for all but the first crankshaft bearing also goes through a 90 degree turn.

So my theory (not original to me but I subscribe to it) is that the oil in the sump is “aerated” (air bubbles are introduced into the oil) by the crank beating the oil like an egg beater.  This mixture of oil and trapped air bubbles runs through the path described.  Every time it goes through a change in direction the air, being lighter, is happier to go around the corner than the oil is.  That is not of consequence until there is a split or branch in the passageway.  The air bubbles are willing to go around the corner, the oil not so much.  As we have seen above, the oil coming out of the filter/cooler goes straight on to the 1st crank bearing (and eventually to the 1st rod bearing).  It has to go around a 90 degree corner to enter the oil gallery so all the oil going to crank and rod bearings 2 through 4 has a higher concentration of air bubbles.  When it gets to the first branch off the oil gallery going to the 2nd crank and rod bearing, the air once again is selected preferentially as it is more willing to go around the 90 degree corner.  For the 3rd crank and rod bearing the phenomena is the same but the bulk of the air has been extracted.  By the time it gets to the passage for the 4th crank and rod bearing, the oil has been de-aerated pretty well.  Ditto for the oil going to the last crank bearing and up to the cam.

So I believe that the culprit is the air entrained in the oil and the design of the oil passageways.  And what are some remedial solutions?  Here is what I have come up with.

• Street use typically means lower rpms and less frothing of the oil.  Don’t rev your engine so much!
• Hell, I want to rev my engine!  So use a good quality oil that minimizes foaming.  There is a long and opinionated discussion about what that oil should be.  Check the archives or Bobs the Oil Guy. Personally I use Valvoline Racing oil.
• Use a synthetic oil.
• Use the correct viscosity.  More discussion and opinions abound.  I use 20W-50.
• Keep your oil filled to the mark but do not overfill.  Overfilling just increases foaming.
• For heavy track use, use aggressive oil change intervals.
• Get a crank scraper to help minimize the foaming.  Alternatively use an oil pan from a later model 944 that has a built in crank scraper.  See article here.

Except for the crank scraper, all of the above can be done by any and all of us that want to track our cars without too much effort.

As I discussed above, I think cross drilling the crank has no value.  I do think that oil pickup extenders and sump baffles are a good idea but more for preventing general oil starvation in long sweepers than anything else.

So there you go. The oiling system for the 944 is reasonably straight forward.  And a good oil supply is the life blood of a high performance engine.  Treat it well and it will hopefully treat you well.

OIL FLOW AT THE HEAD – SINGLE OVERHEAD CAM ENGINE

Edit- Feb 2018
After performing a recent full engine rebuild, I did an inspection and established the flow path for the oil in the head, specifically at the Cam Housing. There is technically no pressurized oil lubrication required in the cylinder head. Oil lubrication is required at the camshaft housing, which is the casting with PORSCHE imprinted on it that you see when you look down on the engine. This casting holds a single overhead cam, which is driven by the camshaft belt. The belt you need to change on a routine basis, before it breaks and trashes your engine! But that is another article.

Anyway, it turns out the camshaft needs lubrication for 2 reasons. The camshaft rotates in 5 bearings that require lubrication. There are also hydraulic valve tappets, commonly known as “hydraulic lifters” or “cam follower buckets” which require oil pressure for both lubrication and to establish sufficient hydraulic pressure to keep the tappet firmly in contact with the camshaft during all operating modes.

Previously in this article, we discussed that there is a large horizontal passageway in the block which has branches that feed the various crankshaft bearings. Towards the rear of this horizontal passageway there is an intersecting vertical passageway that leads straight upward, through the block, through the cylinder head, and into the camshaft housing. This passageway is not particularly exciting to look at but it does have one interesting feature. Embedded in the cylinder head, there is a small round ball. This ball is spring loaded and serves to only allow oil flow upward when the engine is operating. When the engine is shut off, this “check valve” closes and prevents all the oil in the camshaft housing from draining by gravity back into the oil sump. In this way, oil supply to the camshaft is supplied immediately upon engine startup.

The above picture gives a top view of the cylinder head, showing the passageway with the check valve ball. When the camshaft housing is fitted to the top of the cylinder head, this passageway aligns with a oil distribution network in the camshaft housing. This distribution network is what we will discuss next.

In the above picture, the camshaft housing has been flopped over from its normal position on the cylinder head. The 8 large holes hold the hydraulic tappets, one of which is shown in the picture. When the lobe of the camshaft presses on the tappet, it transfers this motion to the end of the valve stem, thus opening the valve. The smaller rectangular shaped holes are oil return passageways. They allow return oil to drain out of the camshaft housing back into the cylinder head. From there, return passageways in the cylinder head (not visible in this photo) allow oil to drain through the head and through the block, all the way back to the sump. This area in the head, around the valve springs, is not fed by any oil directly under pressure, rather it is a “splash” system where the motion of the moving parts tends to pick up the oil that is coming back from the camshaft housing and splash it on the moving parts, which in this case is the valve springs and the portion of the valve stem sticking above the valve guides embedded in the cylinder head.

Getting back to the camshaft housing, the long red wire on the left hand side of the above picture is an extremely exaggerated representation of the oil supply path from the cylinder head into the camshaft housing. When assembled, the oil passageway in the cylinder head lines up with the long horizontal passageway in the camshaft housing. This passageway fills with pressurized oil . The oil leaves the passageway in 2 important ways. The short green wire shows one of five side passages that feed the various bearings visible on the camshaft. Note that Porsche designed the camshaft bearings without replaceable bearing shells. The round cylinders visible in the photos engage matching round holes machined into the inside of the camshaft housing. The oil passageways emerge on the perimeter of the machined hole and provides pressurized lubricating oil to the bearing surface. The oil that exits the sides of the bearing is returned by gravity to the sump, as discussed above. In just a minute, we’ll look at how the two end bearings have special return passages but first let’s talk about the hydraulic tappets. In the above picture, the short red wire leads to a side passageway which leads directly to one of the 8 large bores that contain the tappets.

The tappets are difficult to look at and see how they function but in general, there is a top and a bottom to the tappets that is separated by an oil filled space. The volume of oil in the space determines the height or thickness of the tappet. If the tappet were machined out of a single solid piece of metal, they would be fixed in their height. My Jag E-Type has this type of system. The exact height of each tappet is dialed in using thin circular shims. Since there is no one correct height for all operating conditions, the tappets can be a little noisy, say upon cold startup, when all the parts have not come up to temperature. Except for very high performance engines, manufacturers long ago developed hydraulically actuated tappets. These tappets can compensate their operating height “on the fly” so to speak in response to changing conditions. The exact mechanics of this remain a mystery to me but suffice it to say the internal construction of the tappet allows it to “pump up” or grow taller when conditions demand and to also collapse a little or grow shorter, as required. This happens automatically as the engine is in operation. If it has been a very long time since you started your engine, you may hear a loud clatter for a short while after you start your engine. This is because during that time since the engine last ran, some of the oil has manged to drain out of the tappet, leading to a more collapsed position that provides slop in the camshaft actuation. The clatter occurs when the tappet takes up the slop and the parts come together. As soon as the proper oil volume inside the tappet is re-established, the tappet takes up its correct height, the slop if removed, and the parts engage silently.

Some excess oil emerges from the tappets, which will lubricate them in their bores and also lubricate the critical high contact stress surface between the tappet and the cam lobe. That contact surface between the cam lobe and the tappet is one of the toughest lubrication challenges in your engine. This gets into the science of Tribology. This situation is especially tough to keep lubricated because of the very narrow surface profile of the camshaft lobe as it engages the tappet. This involves science that is way beyond my comprehension but, hey, it works!

In the above picture you can see some faint grooves on the perimeter of the bearing on the camshaft and on the perimeter of the bearing surface machined into the housing. Pressurized oil is distributed around the perimeter of the bearing contact surface and provides a thin film of oil to keep everything running smoothly. Oil is constantly oozing out of the both sides of the bearing. The oil that comes out the front side of the front bearing, like all the oil from the bearings, needs to return to the sump. There is a small hole that leads from the front cam cover  back into the main body of the housing, allowing the oil to return.

The same situation occurs at the rear bearing. Assembly is fairly foolproof but you must make sure that your gasket has the appropriate cutout to expose the return hole.

And that is about it. Now you understand better the configuration of the overhead camshaft on an early 944 engine and how the lubricating oil is supplied and returned!

PS I hope you have noticed how nice and clean all the parts are in my pictures. I have spent much effort bead blasting the various parts to get rid of years of grime and baked on oil. Exterior surfaces are painted with Duplicolor Cast Aluminum High Temperature Engine paint.

OIL PRESSURE RELIEF VALVE

After my recent engine rebuild, I was disappointed to discover that the oil pressure gauge was showing about 3 bar at normal operating conditions. Note that 1 bar is equal to approximately 15 psi. I tried another gauge, to no effect. I also bought a new sending unit and installed it, again with no effect. Note you need a ground down (thin) 24 mm wrench to remove the sending unit in place, or I bought a set of inexpensive “crows feet”, which are already thin. I was certainly hoping that the low oil pressure wasn’t due to a mistake in my rebuild. So the last thing I changed, which fixed the problem, was the oil pressure relief valve. Now I’m getting 4-5 bar. Workshop manual spec is 4 bar at 3000 rpm. Following is some discussion about the oil pressure relief valve (OPRV).

First I would direct you to the following article by Dave in London, which had done a nice job explaining some key elements of the OPRV.

http://944foot2thefloor.blogspot.com/2010/07/how-to-remove-and-replace-oprv-oil.html

I had always known there was the newer 1 Piece valve and the older 3 Piece valve. What I didn’t know was that there were 2 versions of the 1 Piece valve. My engine is normally aspirated but is based on a used turbo block. It turns out after checking the engine code on the side of the block (just to the rear of the orange/black oil filler tower) that my block was from an 87 car and needed the later version of the 1 Piece OPRV. I had used a valve that I got along with a turbo oil cooler. I don’t really need the oil cooler so I used the oil/water cooler from an 85 N/A car. After reading Dave’s article, in particular the Porsche tech bulletin, I realized that I might have the wrong valve. Not only did I have the wrong valve but when I pulled it from the car, it came out in multiple pieces. And the o-ring on the end of the valve was missing. It’s amazing I had any oil pressure at all! Since I had the valve, in pieces, here is an attempt to show how it works.

My internet reading noted that if the system oil pressure goes too high, the weak link is the oil filter, which will blow out. The highest oil pressure might be expected when the oil is cold and thus very thick or viscous. In essence, when the oil pressure exceeds a certain “setpoint” value, the OPRV will open and relieve the excess pressure.

The OPRV resides just downstream of the oil filter. It’s end is exposed to a vertical oil passage in the side of the block. Here is a picture but it is hard to photograph.

When the OPRV opens beyond a certain point, it exposes a drain path. Interesting to me is that the drain path doesn’t go right to the sump but rather it goes back to the suction channel in the crank girdle for the oil pump.

The OPRV has a small piston that is held closed with a long spring. I checked the spring force, very crudely using a kitchen scale, and measured 5 pounds of force. The diameter of the piston is 0.4″. Doing the math, the force to move the piston is only 40 psi. This may explain why I was getting 3 bar! I neglected to measure the spring force on my new valve, which I bought from Travis at Rennbay, but as I said it fixed my problem.

The piston fits inside the outer sleeve, shown above on the right. When the piston moves back into the sleeve, it starts to uncover the 4 round holes. This opens up a flow path, as noted above, which leads to the oil pump suction. I have posted a couple of Youtube videos below. The longer one shows me depressing the piston with a nail, opening the round holes. The second one shows that the far end of the OPRV is slightly spring loaded. I think this mainly serves to allow for small amounts of differential axial expansion due to heating and cooling.

Many folks worry about alignment of the valve. More specifically, one must align the cooler housing correctly when installing it. There isn’t much wiggle room on the position of the cooler housing. And for the 1 piece valve, the external housing doesn’t move, just the piston inside. So I suspect you can get away with doing a visual alignment. But to be safe, I bought and used the alignment tool.

Folks talk about the valve sticking open and causing a total loss of oil pressure. I guess it could happen, for a valve in very poor condition, coked up with deposits, etc. But it would be a hard thing to prove unless you were to remove the valve and see the piston compressed with the little holes visible. Same story with it sticking closed.

The Porsche OEM valve is very expensive but then again it is a precision device with bad consequences if it does not work right. The one I got from Rennbay seems to be working fine and is less costly. It only fits the 87 and newer blocks.

So who knew there was so much to know about the 944 oiling system. Knowledge is power.

 

 

Update: Here is a tip from Michael Mount:

When rebuilding a 944 engine I always disassemble the oil pump to inspect the housing and gears. When reassembling, coat the gears and housing with white grease (it doesn’t take much). The grease closes up the clearances and maximizes the ‘suck’ to pull oil up the pickup tube. The result is oil pressure without the mess or headache when there is none.
MM
PS: You’ll need a hand held hammer type impact driver to disassemble the pump (cheap) and 574 to reseal the pump housing. You’ll see where the factory sealed it, so just duplicate what they did.

Update: This link was posted to Rennlist. It takes you to a Clarks-Garage discussion regarding resistance values for the oil pressure sending unit and a way to test the circuitry downstream of the sending unit. Brilliant!

http://www.clarks-garage.com/shop-ma…ect-19.htm#oil