Smudo

Thanks, will do the most of the mod anyway However, don't want to make any mistakes, thus any insightful comments are welcomed! I always remember about the 944 cup cars that were run on the track - I wonder how many of those did have rod bearing issues. Most probably not lot of them as they used racing oil and changed it frequently and most probably they did change also the bearings more often than each 100k miles Don't want to do it on a semi street car, however, I am using Millers race oil and change it one a season or any 3k miles.

refresh951

I did not read through this whole thread but here is my take on the #2 bearing:

The oiling system has an inherent flaw in that the input side of the oil pump is over-driven at high RPM (some analysis says over 5500 rpm) resulting in cavitation. The cavitation can be minimized as follows:

1) Use high viscosity oil such as 20W-50. Jet951 has several detailed threads about issues with modern oils and additives and the like. He has tons of race hours and I am not aware of his having any #2 failures. Many modern synthetic oils such as Mobil One should be avoided. I run Castrol GTX 20W-50 non synthetic. Many will disagree with me on this selection but whats new. I change the oil after every track event. Many options here but oil selection is critical.

2) Keep oil temps below 220F (by definition this means you monitor oil temp). Many issues with not properly cooling the oil but hot oil cavitates more easily which is critical on the 951.

3) Do your best not to over-rev. Try to keep RPM below 6400 rpm. The stock intake limits power above 6K so no real reason to intentionally go beyond 6K.

4) If you do a rebuild, do the Michael Mount mod or go Hybrid Stroker so you can use modern bearings (Cleavite or equivalent) which are much more robust than the original rod bearings.

All the other mods are second order IMO. Some may do more harm than good (cross drilling for example).

rlm328

see above in red

951and944S

Automatic transmissions use positive displacement gear type pumps of almost identical design and have to pull suction from cloth, fine mesh filters placed before the pump through a suction feed as small or smaller than the 944 pickup tube and can hold 270 psi to past 6000 rpms.

I have a friend that ran at the front end of national PCA races in an S2 for years, not someone beating around the back end of the field and now in a 968. Uses Mobil 1 exclusively with no engine modifications except an oil pan baffle and has never had an engine failure.


On some tracks.....the gearing warrants staying at high revs in a given gear whether you "want to" or not....it's unavoidable. Momentary blips up to and over 6400 are of less a concern as sustained high rpms through a constant sweeper or flat footed through an esses complex. 6000+ a few 100 rpms is totally safe for these engine. Click the link of an actual race video of my son in his SP2 at NOLA and pay attention from 1:10 up to the downshift for the slow right hander.
You can clearly see the tach and the oil gauge. He is at and over 6000 rpms for 25+ seconds and the oil gauge never budges except for a second or two where he feathered a couple hundred rpms which is even more evidence that the pump is still responsive to a 200 rpms change with oil as hot as it will get in his engine. Swepco 20/50 btw. No engine mods except oil pan baffle kit and an external air/oil cooler. https://vimeo.com/96543775


A beneficial modification...agreed. Had this done on a brand new set of Pauter rods for current race engine. That said.....the engine running in the video above is an all original 1987 engine that has only had a cylinder head refresh and glyco rod bearings a couple of times over 7 years of heavy track use. That engine is wrapped up on a roller cart as a backup and is still 100% functional.
Again.....like the S2/968 racer friend I mentioned above.....my son aint beatin 'round the bush.....he's a top 944 wherever he goes....., always near the front and has several track records. These engine are beat on as hard as anybody beats on a 944 engine.

T

333pg333

^^ Always good to read of success with these cars. My only question would be, are those cars and classes exclusively n/a cars? What weight and tyre sizes are allowed? Possibly lighter cars with smaller rubber and lower power not generating the same heat and loads as a higher powered turbo car on wider stickier rubber?

333pg333

Also, I asked my engine builder his thoughts about x-drilling cranks. I knew he had opposed this modification. I wasn't sure of the actual reason though. Was it a possible loss of strength or oil flow?
His reply:

"Oil flow. All you do is pull the oil away from the loaded side of the
bearing because of centrifugal forces. There is a lot more complex reason but
that is it in a nut shell. "

Smudo

I posted the same question also on another forum and it seems that there are guys that are racing these engines for many years without any major modifications, but all of them mentioned that they are using high (edited) viscosity oils (usually 20w50) and have additional oil coolers to keep the temperatures of the oil down. So it seems that if you fallow those two rules, you significantly reduce a risk of sized #2 bearing. Of course, this might not be true regarding highly modified 951.

333pg333

High viscosity, si.

Thom

A point that I do not think I have ever seen raised on this board is the accuracy (or rather lack thereof) of the tachometer, which from my experience can vary from one car to another.
As I upgraded to a standalone ECU one of the first thing I noticed was that an indicated 7000 rpm on the speedo was an actual engine speed of 6400 rpm, and checking with other cars I have access to, taking account actual speed, trans ratios and tyre sizes returned that the tachometer was not always reflecting the actual engine speed and sometimes showing a +-10% error.
Those 6000+ rpm you are seeing there may actually be some actual 5500rpm if the tachometer has never been recalibrated.

951and944S

2600#, 245/16, Hoosier R7/Hankook Z214/BFG R.

There isn't a 944T I have seen yet at 8 NASA and 3-4 PCA races here that is faster through this particular complex.

In the video posted....the cars he is racing around should be minutes up the track.

That Black/Tan 70's 911 is a very fast PCA D car and was national champ in class previously.
The guy that passes in the dark blue Cayman that obviously disappears due to power but didn't get away in the turns and was repassed at the right hander was the 944 SP2 national champion a couple of years ago.

T

951and944S

Possibly....but not in this case.

In the car I race personally, I have redundant tachs. The stock one and an aftermarket with a shift light (we do a once a year 400 mile enduro partially at night) with maybe 50 rpms variance.

The car in question has the now mandatory PCA "data" plug which allows PCA to attach a data acquisition device which measures G forces, speed and tachometer. This tach is verified +/- 50 rpms.

T

refresh951

Apples and oranges. A Honda engineer did some specific analysis. The details matter.

I have researched a ton of failures. Using Mobile 1 is a common point in turbo failures.

My point is avoid high rev if possible. It needs to be a strategy consideration. Otherwise use a dry sump (which has its own complexities).

Agreed, there are many exception But if you research the failures, several empirical common points come into view.

951and944S

One more....ok, maybe two more notes on this topic...

Short of a dry sump (read - costly and also not legal in the classes we compete in) in attempts to get a reliable supply of engine oil to the pump, I had an idea that seemed to have merit on several fronts.
The oil pan could be bored and tapped at the lowest point on the passenger side (just makes more sense logistically)....a stainless line of at least equal cross section as the girdle oil suction passage run along the pan rail towards oil pump and tapped directly into the suction side of the pump itself. The circuitry allows for the oil pressure relief valve and oil filter housing location to remain as factory.

I have 30 years transmission background and we used to do this to IMCA dirt track cars (the class that goes through turns sideways with the big wings on the top.. ) that ran automatics. They would ease drive engagement by controlling the pump pressure through a hand operated flow valve.

On the 944....., it would allow much more creativity in sump baffling and also eliminate the notoriously problematic oil pickup tube. The existing hole in the girdle where the sometimes questionable seal locates, would be plugged and permanently eliminated.

I ended up not making an attempt to prove the theory out due to time constraints and my urgency in needing that particular block.
The hurdle would be making the turn to the suction face on front of the pump as the space between that surface and the crank pulley is small....but I think a banjo bolt would do the job.

Refresh mentioned the rod modification for a more durable bearing. Some things just make sense....I agree with the theory behind that modification and have had it done myself.

There is talk of venting the blocks through the crankcase webbing, that too I agree with as Kolbenschmidt casts blocks this way currently so there is obviously a reliability benefit as well as potential for a performance enhancement as a byproduct. 944 family blocks are already vented from cylinder to cylinder through the oil drain back passages in the girdle that can theoretically use the cylinder head and camshaft housing as a pressure equalizing conduit. Unless you believe that each cylinder inside the crankcase has an air tight seal....

T

V2Rocket

928 guys have luck with a 3/8" spacer that moves the oil pan down.
they then space the oil pickup tube down to compensate.
their reasoning is to add air-gap between the crank whirlwind and the oil...
maybe we could use something like that to add more oil capacity (more oil to completely submerge the pickup tube, less likely to find air if it is "deeper")

951and944S

I have seen some of the research. Not saying there is no merit there, just that IMO, the topic is no further "settled" that man made global warming... And no, I disagree, we are not talking apples and oranges in relation to the transmission pump of identical design. The transmission has thinner oil (a point you were trying to argue as a factor to avoid re, heat = thinner oil), has an inter-mesh gear pump that is of the same design and is further restricted by having to pull suction from the filter first through a passage more limited in size and routing turns than the 944 girdle.



I have no data here to argue this point, only that there could definitely be other conditions that lead to specific turbo engine failures, maintenance, engine clearances and general condition of these engines. I am only offering examples of highly competitive cars that compete directly with the 944T in E and SP3 class PCA with an engine of same design that use M1 exclusively without any problems.



That "strategy" is a more viable consideration with the turbo cars as they make way more torque....the reason that they are weight penalized compared to an S2...but in NA 8V cars, momentum is everything. You make a shift that takes you down the rev range unnecessarily and you are backing up down the order. I would argue that a competitive front running 8V NA car's engine takes at least an equal beating (revs and constant high rpm) as the turbo engined cars. Probably not on the heat soak aspect but this can be addressed through proper cooling.



Could be.....but data, theory, potential fixes have been exchanged since 944s have been racing and here, at least as long as I have been around (2001) and the issue remains unresolved.

If you think 100% sure that the pump is limited, the answer would be to increase pump efficiency by investigating a modification to the pump itself for conversion to a vane/rotor type like found in modern automatics.
At a GM training facility in New Orleans years ago, to demonstrate the differences in vane/rotor and positive displacement gear pumps (prior technology), the instructor had us route a 5/16 thread fitted hose to the line pressure tap (external fitting on case used for diagnosing pump issues) routed back into the dipstick tube. Basically creating a 5/16 cross section leak that would be terminal to a previous design era TH350/250/400 (as was also demonstrated). The hose flow was controlled by a ball valve, vehicle run up to speed and line pressure (pumps viability) was measured to be 100% in specification. The ball valve opened, the pump recovered after a 1 second drop off of spec and recovered to 100%. We are talking about a hole in the 944 girdle the size of a pencil. The positive displacement pump example during the same test fell 30 psi (30% of spec) and never recovered.
Transmission pump components are inserted into a cavity called a "slide" that is very much like the insert found in 944 oil pumps. Converting this pump to vane/rotor is definitely doable.

T