#2 Starvation

[jperine]

Can any one point me to a resource that describes and solves the #2 rod dry-out problem that has been experienced by some of you?

 

I have heard everything from modern tires inflicting more g’s then the design spec of the stock oilpan to sloppy engine rebuilds causing this failure.

 

My guess is someone builds a trick baffled pan that solves this. I am certain Porsche never lost an engine to this problem… twice.

 

I have been over every post on the board and cannot find the solution.

 

Thanks for the help.

 

Jeff

[Spec-944#70]

Jon Milledge and other believe an air/oil seperator will solve this problem. Debate is still out on this as other think cross drilling the crank will help also.

[greg f]

The problem with trying to research this problem is that once it happens, most owners throw the gamut of repairs at it so it won't happen again.

 

We have not had any failures doing the following mods.....

 

1. Lindsey racing style swinging baffle for the oil pan

2. Cross drill #2 and #3 rod bearing journals to 3/16"

3. Make certain that the cylinder head has the Turbo press relief valve

4. Use the 1 piece oil pressure relief valve, and use Mobil one oil.

 

I also use the plastic pan baffle from a 944S2 ( better, taller crank scraper) and a 88 style pan if budget allows. If I had to choose what makes the most difference it would be the cross drilling of 2,3 rod journals to better balance the oil flow to those bearings. We have an engine here with 128 racing hours on the rod bearings, ( they looked perfect when replaced) with these mods and the owner had also installed an Accusump and used it to pre-oil before starting. Greg F

[944-Spec#94]

I agree with Greg's Assessment.

 

I have suffer two such failures. First one I believe was due to broken oil pickup tube. It broke due to vibes and well pumped sucked air and well you get the idea.

 

 

I also suffer one on fresh rebuild. I thougth I did it all right, but I still don't have a definitive cause. Luckily the bearing just spun and did not exit block.

 

Unfortauntly there are not solid solutions, but alot of "well you can try this..." answers.

 

It unfortunatly is the ONLY weak spot on an otherwise bulletproof car.

 

 

PS... only other things likely to fail are due to old age and a rough life with previous owner(s). After those are addressed the cars run solid with maintenence being tires twice year, brake pads once a year, rotors every couple years. Oh and a supply of pump gas.

[Danno]

There's still no definite answer to the actual cause of this #2 rod bearing problem. There are a tonne of "fixes" and "solutions" that appear to alleviate the problem without ever really pinpointing what it is.

 

Oil and hydraulics is a tricky thing to model. As with all fluid-dynamics, it's basically is parallel flow, not serial. All four main journals are feed in parallel by the oil-pump, oil-pressure is the same on all surfaces that the oil-touches in the oil galleys. This oil is then squeezed through the main bearings into the crank's main journals. After the oil has completely filled the crank, less than 1-second after startup, pressure everywhere inside the crank is exactly the same. This oil is then squeezed out through the webbing into the rod-journals. I suspect that this last step is where the differences exist between the different journals. If the problem was anywhere earlier, every single rod-journal would be affected evenly.

 

So I'm going to do some flow-testing on each rod-journal to see if they all flow evenly, given identical pressures inside the crank. Then I'm going to cut one open to see if there are any dimensional differences internally...

[greg f]

I think that is pretty much what I did here. Not finding a definitive answer anywhere, I made a manifold that allowed me to supply 90psi water through the front of the block. I tested with no crank, crank and main bearings only, and with rods installed.

 

What I found was that while the pressure might be the same the flow rate and oil supply is not. #1 gets 50% of the flow, #4 gets 25%, and then #2,#3 share the remaining 25%. Water would gush out of the #1 area, but be berely a trickle from #2,3.

 

The first solution was a 30% restrictor installed in the crankcase under the #1 bearing oil supply hole. This balanced the flow better, but a difference could still be measured. Next I installed a fully crossdrilled crankshaft. No visible change to the pressure distribution pattern. I then took a crank down and had just the #2,3 rod journals crossdrilled to 3/16" size. This was the best yet, almost more water was flowing from the 2,3 than #1. The restrictor was drilled out to be only 10% smaller and the flow was the closest yet.

 

Lately, we have not been installing the restrictor, instead just drilling the crank, and have had great results. No #2 bearing failures in the 8 engines that are currently being campained. Hal Hiltons car had over 128 racing hours on the bearings when they were removed and found to be in excellent condition.

 

I think the thing to remember here is that a combination of oil volume and pressure keeps the bearings off of the crank. The higher the revs or the lower the pressure, the more likely a failure will be. For my cars the oil pressure must be between 4 and 5 bar @ 5000RPM to run the car at the track. If it falls under 2.5 bar at hot idle, it's time for bearings. Greg F

[jjgrow]

That's the most definitive work on this problem I've seen. Sounds like it might just be worth cross-drilling the crank if the engine is apart. I'd like to better understand the baffling too.

[greg f]

I just can't be sure how effective baffling the pan is. When I look at a 944 pan it seems like a good design. There is a large plastic windage tray that looks better than a lot of engines that do not have this severe of an oiling problem.

 

I install the Lindsey baffle. It welds into the pan on the drainplug side of the oil pick-up. It's purpose is to keep the oil that would be rushing across the pan, ( and up the sides ) closer to the pick-up. They also supply a weld on stainless ring that lowers the oil pick-up in the pan.

 

the 944S2 plactic windage tray is to help with the second big oiling problem with 944 series engines. The crankshaft tries to pick up any oil it touches. This can create a "ball" of oil around the crank at high RPM's. If you look closely at an 88 oil pan, you would notice that Porsche raised the longitudinal ribs that run parallel to the crank. They are actually sculpted to be closer to the crank and scrape the oil from the crank. The S2 also has much taller scrapers on it, and fits directly into a late or early pan. This phenomenon is why I think it is generally a bad idea to run the engine 1 QT high on oil. The crank just flings it around, aerating it and taking away HP while making more heat.

 

Milledge is a big fan of sealing or gluing the plastic windage tray in place to eliminate oil forcing its way past the edges in hard braking and cornering. You can't really use silicon or other soft sealants because of their tendency to come undone and clog the oil pick up. Epoxy works well, as does the new Porsche sealant they use on the new cars. It is a Dow / Corning product #732 and is a great way to solve nagging 944 oil leaks. I have also riveted small bits of aluminum to make the windage tray more effective at keeping the oil down in the pan.

 

My next project is to see if a 996 oil/air separator can be fitted to the 944 oil return pipe. The oil return pipe discharges right next to the oil suction pipe. Normally this would be good, but I suspect that the oil that is coming back from the lifters may have some air mixed in with it. The 996 uses a simple oil/air separator on the return side to only supply oil to the pick up, no air bubbles. Greg F

[Danno]

Greg, thanks for sharing your data with us! I was going to do a similar experiment with an electric pump flowing oil into the block. Then I was going to measure configurations:

 

1. measure oil-flow out of each main-bearing suppply hole (no crank in place)

2. measure oil-flow out of each rod-journal

3. measure oil-flow out of each rod-journal with con-rods in place

 

I was going to run the oil for 1-minute and collected what came out of each hole. On experiment #2, I haven't figured out how to capture the rod-journal yet as one hole would be aimed up when the next one over is aimed down.

 

Also on experiment #3 , I haven't figured out how to collect the samples cleanly.

 

Another thing to consider is that pressure and flow are not related. It depends upon the orifice you're flowing the fluid through. With new bearings, the rate-limiting step is the bearing-clearance. You can open up all the holes ahead of the bearings, but it won't change the flow-rate.

 

I wonder why the #3 bearing doesn't have teh same failure rates as the #2?

[greg f]

I have had quite a few engines through here with #3 failures. I almost never see a #3 failure without some similar damage to #2. If #3 goes it invariably pitches the rod through the block, making post failure analysis difficult.

 

I varied the water pressure from zero to 90psi to zero and noted that if there is a low pressure event the #2,3 holes would stop flowing first every time before crossdrilling.

 

When I started researching this problem, ( still in process!) I spoke with a few noted builders of big block Chevy race engines. They have virtually the same sort of failures as the 944 series engines. They also share oil system designs to a degree, and have a very large, heavy piston connected to a relatively small rod bearing. The failure of these engines is limited to the center fed rod bearings. The #1 or #8 bearings never fail. All of the failures relate to higher than stock rpm's and some sort of G-force. Interesting to note that Drag racers seemed to have as many or more failures than the circle track guys. This would seem to indicate rpm rather than G force for this engine. ( most of the circle track big block applications are dry sumped, but still could see the #2 failure)

 

Greg F

[skip]

...I almost never see a #3 failure without some similar damage to #2. If #3 goes it invariably pitches the rod through the block, making post failure analysis difficult.

 

Having just pulled my engine apart, add another to this statistic.

 

 

 

#2 was almost non-existent. #3 was nearly welded on the crank. Logic follows what Greg has reported. #2 failed well before #3 caused engine to lock up. Increased flow thru #2, or, simply reduced oil pressure appears to have affected #3 as the next weakest link. #1 and #4 looked fine. All of this happened within a few hundred feet. I hesitated when I thought I felt a loss of power coming off a corner pretty hard (lefty, #2 at Pacific) - high in 3rd - looked at the OP gauge and immediately let off the gas and turned off the engine. Coasted to a stop, then just tried to bump the starter to see if it would still turn - no luck, locked up hard. Luckily did not throw a rod off. My excuse is that I was chasing a turbo - thou shalt not use na's to chase turbos... bad things happen.

[Danno]

I just realized something else after staring at this engine on my stand for the past year... Take a look at what happens to oil in the head that's been beaten up by the cam and foamed up real nice by the valves & valve-springs. Then take a look at where that oil is dumped into the pan... what's right next to the spot where it dumps???

[944-Spec#94]

Danno,

I am not really good with quizes so... what are you talkign about?

 

Are you saying the drain back and pickup tubes are too close?

[greg f]

Danno is refering to the oil return tube returning the foamy oil from the lifters right next to the oil pickup. I weld a 3/4" stainless ring around the pick-up to improve scavenging, at the same time I cut a 1/2" half circular relief in the side of the return tube that faces away from the pick-up. This allows any air to escape to the rear first, before being picked up by the inlet tube.

My next plan is to try to incorporate a 996 oil de-aerator into a 944 engine. The 996 has an oil pump in each cylinder head to force oil back to the sump. At the end of the return tube there is a device that helps separate the air from the oil before it is returned to the sump.

Side note: Danno, the last chip you sent is a definite keeper, I stuck it in a bone stock 86 with our exhaust and it was impressively quick last weekend. Greg F.

[skip]

I weld a 3/4" stainless ring around the pick-up to improve scavenging, at the same time I cut a 1/2" half circular relief in the side of the return tube that faces away from the pick-up.

 

regarding the 3/4" ring, I think I understand you are able to do this without modifying the pan - as long as it's a late deep pan (smooth bottom)? Why wouldn't the pickup be as close to the bottom of the pan from Porsche? At this point, the pickup screen is basically touching the bottom of the pan, right?

 

My next plan is to try to incorporate a 996 oil de-aerator into a 944 engine.

 

Is this in addition to a catch can? Or, is the catch can not needed with a better air/oil sep?

 

Also, is the crank cross-drill (#2,3) and pickup tube mod a service yet, or are you still playing around with it? I need something soon - engine is basically ready to go in. I was going to risk it again by doing nothing by just running out the $500/engine clock - maybe it's time for a bit of prevention.

[greg f]

Skip, I can have the crank done right away for around $200. It will be crossdrilled, magnufluxed, checked for true and trued, polished and balanced.

The problem I have had with just running a catch can is that at high RPM the engine wants to push out all of the seals without the negative pressure from the intake connection. What we do now is fabricate a de-aerator can that is sealed, and still use the intake pipe connection to create crankcase vacuum. ( keeps the oil leaks to a minimum)

The ring can be installed on a late or early pan, the problem with the factory design is that the intake is curved the wrong way. The screen may well touch the pan, but the oil follows the contour of the inlet. By lowering the sides of the inlet you decrease the possibility of sucking air. The ring also helps prevent the aerated oil from the return tube from being drawn directly into the oil pump. The intended purpose of the ring is to stop the oil that is rushing back and forth across the pan from drawing oil away from the suction pipe. Greg F

[skip]

sounds good. Can you do the pickup mod as well? All you need is the crank and the pickup tube? Should I source the S2 baffle myself or do you have used?

 

Need to keep everything ITS legal - I finally have a 88 2.5l going in!

 

I should be able to drop it by your shop next week sometime. Thanks!

[greg f]

I can do the pickup tube mod as well. I just buy the S2 windage tray new, it must work ok since there arent any dead S2 engines here to get the part from used......

I also experimented recently with the application of an ant-friction coating on the rod bearings. This has been popular in Winston cup for years so I thought I would try it out. The company is Calico coatings and I will say I am impressed! We accidentally tested the coating by installing the wrong thrust washer under the crank bolt on a new engine. The engine ran for 45-60 seconds with no oil pressure. ( the washer was an 83 style and provided no thrust on the oil pump drive.) I dissasembled the new engine and found marks on the main bearings, but zero wear on the coated rod bearings. Greg F

[skip]

okay, sump insert 944.107.389.03 is on its way - not as pricey as I had expected.

 

Every time you post, it's costing me more money! I'm trying to find that sweet spot of not paying 3 times the cost of an engine on endurance mods.

 

That said, I have heard from the local drag car folks that the coatings are really worth it. How does Calico compare to Swain ($41.50 for rods and mains)? What about Performance Coatings in Auburn ($35 for rods and mains)? For under $50, it seems worthwhile.

 

I'd really like to leave out the Accusump if possible - that's another $250... or, nearly a set of tires. Is the Accusump really that useful when all these other issues are dealt with?

 

Any guesses as to how many track days/hours we can expect from this setup (baffle, pickup, crossdrill, coating)? I promise not to hold you to it

[greg f]

I have no data to support the advantage of an Accusump except for the pre-oiling advantage. Hal Hilton uses his religously and I have never seen a better looking set of engine bearings considering the car had over 128 hard racing hours on the engine hour meter.

I do not run one in my car, which has 5 YEARS of hard racing on a stock 88 motor with just the crank drilled. I have it out now and will advise as to the condition of the bearings. Greg F

[skip]

...over 128 hard racing hours... ...has 5 YEARS of hard racing on a stock 88 motor...

 

LOL - I guess that'll do for a testimonial. That would equate to about 6 years given my activity level!