bebbetufs

To be honest i don't know. Heavier weight oil may make it harder for the pump to draw the oil up from the sump and increase the chance of the pump cavitating. On the other hand the film should be stronger and provide a larger safety margin.

I don't want to theorize more than that, we really need testing to determine this properly. A sacrficial engine on a moveable engine dyno with looking glasses and pressure transducers on all oil feeds should do it....

bebbetufs

I just came across this. Could this be of value to us?

V2Rocket

hey that's kinda cool...

as far as air in the oil goes, could adding a second oil filter to system help in any way? oil having to negotiate another matrix before being sent through the block might catch bubbles, no?

bebbetufs

Unfortunately not. It would have to be under vacuum or vent to atmosphere or it would fill up with gas.

It would be really interesting to model our engine in CFD to see what happens to the oil flow. There must be some engineers in here who could whip up a working model in Creo or SolidWorks and who has access to some CFD software?

V2Rocket

i recall from previous discussions that the 944 oil pump starts to aerate/cavitate around 4500-5000 rpm, that it can put out more oil than it can suck up at higher RPM.

can the oil-pump feed passageway in the block be machined slightly larger, to allow more oil to go to the pump?

bebbetufs

Probably. But the real problem with cavitation is that the space between the gears on the gerotor pump cannot fill fast enough due to a combination of the weight of the oil, the depth of the teeth and the speed of the pump. The oil can't fill the voids between the teeth of the pum and is violently pulled apart causing gass bubbles to form.

Bubbles can also form downsteam of the pump if the pressure falls far enough, but I doubt this happens in our engines. Here is a youtube clip that as far as I can tell shows this form of cavitation.
Here's a video of a computer simulaiton showing where cavitation occurs on a gerotor pump.

As far as cavitation it is the pump that is the main problem. I assume it is optimized for cold starts and cruising, not sustained high RPMs. As far as I know these bubbles start to dissapear as soon as the pressure increases (upstream of the pump). The main problem is that cavitation seriously impairs the ability of the pump to move fluid effectively. Combine this with accelerating hard out of a corner and sucking up the oily foam that was whipped up by the crank under heavy breaking a moment before, and boom....

Still, you may have a valid point in that improving flow could help reduce cavitation or at least increase the RPM threshold for when it sets in slightly.

bebbetufs

Since I'm building my engine now I'm researching this subject and I thought I would add my thoughts here in to to hopefully get some ideas going and maybe help out a little.

I have been looking at the block and there are many 90 degree and 45 degree turns the oil has to make. Now all these turns have sharp edges. This is very bad, especially on the inside of the bend. Radiusing these edges and smoothing them as carefully as possible should really help flow.
At first I worried that radiusing the corners would draw even more air to certain bearings, but after thinking about it I've reached the conclusion that increasing flow as much as possible should be prioritized as high flow will provide higher pressures at the bearings and more oil, even if this oil is full of bubbles. I will radius and polish all sharp edges within reach in the oiling system and galleys in an effort to improve flow. It is imortant that all corner radii are made as smooth as possible. Rough surfaces cause turbulence which lets the flow detach. A small radius bend that is smooth is better than a large rough bend. It makes sense if one things about the oiling system in terms of intake and exhaust ports and headers. Of course the block will have to be thoroughly cleaned afterwards so this is probably only valid for complete rebuilds.

I have already ported the oil returns from the head. They are extremely restrictive and can be almost doubled in size. I don't think this will help the bearing failures, but it will definitely reduce the risk of blowing out the cam tower gasket due to pressure build up in the cam tower/head.

I will also tear drop the oil feed hole of no 2 and 4 main journals which feed rod journals 2 and 3.

Finally I will install a custom baffle system in the pan and crank scrapers to help reduce the windage cloud.

I really can't see what else can be done to reduce help improve oiling apart from reducing the diameter of the crank to keep it out of the oil as much as possible. However, since the rod pins cannot be reduced they will hit the oil anyway, so the effect may not be as great as one would hope. To reduce rotational mass I will use a lightened flywheel instead because this can be tuned more easily if I decide to change direction or sell the car.

bebbetufs

Here's a link to an oil-flow diagram for the 944/951.

kevin12973

HarveyF, Thanks for all your research and documentation. Its two years later and would love an update to your thoughts of worthwhile oil system upgrades, and what is unnecessary for our(racers) application. Have you seen anymore bearing failures due to aeration of oil? Thanks, Kevin

Ernie J

Hi All, just read what you of you have all stated. Mantissport has some more methods of oil control/better oiling for 944's. Since we race these cars. We first mod the block to stop the flow of bypassed oil back to oil pump, air in the oil. We eyebrow the rod journal, cross drill the crank, crank scraper used, , fabricate a 2L larger oil pan,remove the stock baffling, which is all wrong for racing , most of our tracks are clockwise, install 2 baffles in the pan,which seal right and left turns. More oil means more time for air to release from the oil. That's one of the big advanatges of dry sump systems. Baffle on non 3.L blocks, the bottoms of the cylinders, oil hangs up there, more oil in the pan of whats in the engine available to the oil pickup tube, oil tube pickup shrouded. Oil pans/blocks modified on exchange base. www.mantissport.ca

Ernie J

Gee forgot ultimate fix, includes changing the rod bearings to a size inwhich you can get "race bearings". this requires different rod big end sizing and rod journals. Or aftermarket rods which accept race bearings.

Humboldtgrin

I would like to agree with most all that has been stated and add too.

Aeration is the number one problem with all engine oils, not viscosity breakdown from heat like old oil commercials stated. A dry sump is nice but cost a lot and there is one with an AOS that LR sells and others without, but they can maintain high RPM's without cavitation. Another add to solution is to use a crank scraper with a screen so when the egg beater flings oil off it will hit the screen and form a drop, not hit the oil and create more aeration of the oil. A ring around the pick up tube is helpful so it only picks up oil from the bottom and not the sides of the pick up tube screen. Baffling the oil pan helps a lot for track duty and using a differanct baffle then the factory isn't a bad idea. Even cutting out the lower part of the stock baffle could help in regards to slowing down the oil from splashing back and forth in the pan. A baffle on the turbo oil return side of the pan is a good idea also to help the oil slow down and cool down before being sucked back up into the pickup tube. Oil brands make a huge difference in regarding everything about oils. Millers oil is the brand that I have been looking at for my 951 for track duty.
http://www.millersoils.co.uk/
http://www.crank-scrapers.com/Porsche_944.html
http://www.lindseyracing.com/LR/Parts/LRDSUMP.html

Also like Van pointed out years ago adding a third hole, second oil supply from the other main journals would increase oil pressure to the rod bearings. The thrust journal and bearing (middle journal on crankshaft) would need to be modified to allow oil to be supplied thru it to the #2 and #3 rod bearings as well. I also see that the rod journal on the crankshaft has its oil feed pointing in the direction of the crankshaft spinning. The oil has to fight centrifugal force (like being on the gravitron at the fair), so at high RPM the wieght of the oil has to fight its way out to the rod bearing. Cross drilling or perp drilling the rod journal would reduce that effect dramatically but only if there was enough oil pressure at the rod journal at those higher RPM's.

Regarding the stock oil pump, if someone was inclined to machine a different style gear set for the stock oil pump that may help, meaning use round tip lobes instead of sharp edge teeth to create the oil pressure needed. Maybe the oil wouldn't cavitate with round lobes vs sharp edges like the stock oil pump uses now. Just an idea.

If you used the proper oil and a LR dry sump pump w/OAS or baffle the stock oil pan, use a crankshaft scraper and screen, used a ring around the pickup tube, cross drill or perp drill the rod journals, use a second main bearing to aid in supplying oil to the rod journals like Van has mentioned then I think the rods bearings would have a better supply of non aerated oil with good oil pressure.

I had to edit: I was thinking the upper main bearings were not 180 degree groved but they are. I was thinking that would have created some problems but it's not the issue.

And that's my 2 cents.

harveyf

I don't have much to add. But a few thoughts and observations. The 1985 2.5L engine on my dedicated track car now has about 175 hours of solid track use with no problems. I do not have the "trap door" in the sump nor do I have the "ring" around the pickup. I am running a pan from a later car that has the Porsche designed crank scrapers built in. My crank is not cross drilled. So all in all I am just running a really clean engine using Valvoline 20W-50 synthetic racing oil with oil/filter changes after every other event. I have replaced the rod bearings twice in that timeframe. When examined, the used ones looked fine to me. I would also add that in the 2 years since I posted the original article I can't even think of 5 posts regarding oil related rod bearing failures involving cars that were in good condition and that did not suffer from other deferred maintenance issues. Not to say they aren't out there but I'm not sure it's a problem so prevalent that more dramatic measures are called for. Of course, when it happens on the 2nd run of your track weekend, that really sucks!!!

I would say, again, that cross drilling, tear drops, etc at the journal surface is a waste of time. Kind of like those cheesy Y pipes that they are putting on the exhaust pipes on large single pipe pickup trucks. This is not a problem in my mind with oil pressure or oil volume. The volume is established by the small passage sizes upstream of the journal surface. The pressure is established by the clearance between the rod journal and the bearing, and to some degree the side clearance at the edges of the rod to the crank. You are aren't going to appreciably affect those dynamics with a short, additional passageway through the rod journal or with a teardrop.

I do like what Michael Mount has done with the conversion to a NASCAR style rod bearing. I am going to use his services for that type of rod bearing on my next serious engine build.

odonnell

Has anyone modified the oil return, so that it returns the oil to a different area of the sump (i.e. a little farther away from the pickup)? Like, maybe braze an elbow on the end. Idea being that frothy oil from a hard pull or something would be less likely to be immediately drawn back in.

Humboldtgrin

I'm inclined to disagree. I've owned two engines with #2 rod bearing failures that were maintained. Replacing the rod bearings and the oil that often tells me that you don't trust the oil system as well. The mod that MM does has to due with using a nascar style H beam rod bearings with harder meterial vs the bearings avalible for the current rod journal size that are softer. However that doesn't take care of the aeration issue, or help with centrifical force playing on the oil in the crankshaft at higher RPM's. Or the sock oil pump from cavitating at higher RPM's. Or flow enough oil to the rod bearings at higher RPM's. If you keep the RPM's below 4500 then you should be fine with the stock setup. If above 4500 RPM's I would invest in the parts and machine service to help aid the oil to do its job as best as it can.