944CS

Mahle motorsport pistons are going to be more durable than stock. You have an option for anodizing on the ring lands, and they come with grafal coated skirts.

cartpaulo

I'll add my 2 cents as I've failed many connecting rods in the M44 NA engine and I have the line-up of wasted crankshafts and blocks to show for it. This engine suffers from some design flaws in the connecting rod big end as well as the lubrication supply to rod bearings, particularly rod #2. I fully agree with the point made that it is the top of the stroke reversal on the ex / int overlap that creates the worst case loading mode on a connecting rod as any engineer with background in IC reciprocating engines will confirm. The loads can be tremendous and are typically distributed in the big-end area / bolts. After analyzing several failures in detail it appears the failure mode goes something like this. The bolts stretch and the bearing support bore goes out of round, pinching the bearing against the rod journal at the split. The bearing shell then acts as a squegee, scraping the oil film from the crankshft journal and eliminating the hydro component of the hydrostatic bearing. At 6,000 rpm it takes but a microsecond in this dry mode to melt off the bearing babbit and leave the steel backing of the bearing riding the journal with enough extra clearance to now allow the piston to contact the cylincer head. I have caught this failure at all of these stages and have as evidence streched rod bolts (intact), rods with out of round bearing bores big in the direction of the length of the rod by .002-.003, bearings that "fall out" of the connecting rods at disassembly no longer being a slide-in fit, and of course the more catastrophic failures mentioned previously. All this on a stock N/A engine (158 hp) on a stock DME with a 6300 rpm limiter. This is under racing conditions where the engine is living between 4800 and 6200 rpm. All failures have occurred in an off throttle condition up around 6000 rpm. The stock rods came in both forged (up through the model 83) and a cast steel afterwards. Both weight approx. the same, and fail in a similar manner. The piston / pin assemblies are a bit heavy for the size and are most likely the biggest culprit in the problem. Removing mass from that assembly will greatly improve the reliability in these conditions. The problem area in any connecting rod is big end integrity. Although the rods appear robust, they are not quite up to the continuous punishment of hours of racing in the 6000 rpm range. Some improvement can be found using a better bolt /nut combination, and by replacing the bolts after a number of racing hours.
An interesting item I came across in my search to remedy this situation is what the factory did in response. They were obviously aware of the failure mode, and they made some changes in the final iteration of this engine. The 3.0L S2 and 968 use a differernt rod, piston, pin combination. I haven't gotten my hands on a piston / pin set to examine it, but I do have a set of the rods. They are considerably lighter, by a couple hundred grams! They returned to a forging which resembles the aftermarket Pauter rods which seem to have worked well. The big end configuration appears no stronger, but it is pulling on considerably less mass. The rod pn is 944-103-111-1 R55. I'm betting that the engineers there analyzed the problem and, considering the greater output rating and loading of the 3.0L determined it was time to add some safety margin.

blade7

We get along very well, it's just if they start a fight we finish it.