By Ron LaDow
Special thanks for contributions by Alan Klingen of The Stable and Neil Fennessey, Ph.D.
First published 356 Registry Magazine, Vol 28. No. 4, Nov/Dec 2004
Modern engine design, lubrication and filtration systems provide amazing engine life and no sacrifice in performance; that is a development of the last 10 years or so. We've got the stock 356 engine design and the lube system is fine, but the filtration system leaves a lot to be desired and can be improved. Several shops and parts vendors (I own one) offer modern oil filtration.
Along about mid-20th century, engine speeds were increasing, bearing clearances were dropping to about .0013 per inch of bearing diameter, oil pumps were becoming well-developed and the engine designers saw that providing pressurized oil directly to the bearings drastically improved engine life. But to be effective under vehicle accelerations, and to keep the height of the engine low enough for modern cars, the pump pickups were placed at the lowest part of the sump; exactly where the dirt settled without harm in the former 'splash' lubrication systems. They became bottom feeders.
Filter technology was still pretty immature, so the best that could be done within the available space and technology was to strain all of the oil and filter some of it. That was the state of technology when the 356 engine was designed, and it's been pointed out that 356 engines do have full-flow oil filtration; it's that screen around the oil pickup. At the time, oil changes were required at 3,000 miles and the factory handed out badges for cars which had run 60,000 miles without rebuilds. Long-lived engines enjoyed 1,500 mile oil change intervals.
During an engine rebuild, it was (and is) common to grind the crank along with renewing the bearings, a result of contaminates into the lube system. Or worse, a spun bearing means align-boring the case and trying to get the proper bearings. When parts were commonly available and relatively cheap, this was acceptable. But filtration technology changed drastically in the late 20th century, driven largely by the cleanliness required to produce semiconductors. Specialized paper filter elements now trap very small particles from both liquids and gases and do so in small packages and small pressure drops. A 60,000 mile engine is now just broken-in.
My recent interest in better filtration was sparked when rebuilding a fairly fresh 356 engine (~50,000 miles) and at the same time inspecting a disassembled 911 engine with some 250,000 miles on it. The 356 bearings were trash and the crank needed polishing, only just staying within spec without a grind. The 911 crank looked brand new and in fact, the bearings were re-used. So the practical results of better filtration were obvious, but the data on the bypass filtration remained unknown. The improvement in wear we can expect should roughly approximate the improvement in oil filtration we can effect, but specific numbers were unknown.
It i s commonly stated (me, too) that 15% of the oil in a 356 engine goes through the stock filter rather than directly to the bearings. But checking the Factory Shop Manual, the Owner' Manual, and both Elfrink and Clymer finds neither this nor any other number stated. Rough geometrical estimates suggested less than that, perhaps as low as 5%, but Neil Fennessey (who did the calculations) was able to authoritatively state that the fluid dynamics complicated the matter. The only way to establish the amount was through experimentation.
Time on the engine dynamometer at The Stable had already been scheduled for some other investigations, so it was a matter of designing and fabricating the parts required for this experiment (shown at right). Some scheming showed it not to be difficult; nothing more than a 'total-loss' lube system, measuring the input and two outputs. The only 'accurate' measurement required would be the output of the filter, all others could be within a couple of ounces and serve our purposes quite well.
An old sump plate was modified to accept a supply hose to to the oil pick up and to exhaust the sump oil into a pan for measurement. An up-ended pit-lane fuel jug with the bottom cut off was enlisted as a calibrated supply reservoir. The filter output was simplicity itself; a stock hose was cut across the bottom 'tube' section and fitted with some vinyl hose, the other end of which got which got clipped inside a calibrated beaker. Finally, various valves and fittings were included in the hose runs to manage the flow so I needn't later mop Alan' floor. The photo shows the parts as- installed.
The procedure was also simple: Pull the spark plugs, run the engine on the starter until 2 quarts were pumped from the reservoir and measure the results. It was run twice since the first run showed the various engine cavities still filling; that is to say, the total amount from the filter and the sump did not add up to the 2 quarts pumped. The second run got within acceptable total amounts (less than 2 ounces difference) and that data was recorded as the results.
To put it still another way, how long does some contaminate circulate before being captured by the bypass filter? To answer that, we need to know how long it takes to filter some percentage of the sump oil. Neil Fennessey had earlier estimated the time it takes to filter those amounts and graciously offered to complete the calcs based on the data now available. Briefly, Neil finds it takes approximately 20 minutes of operation to filter 50% of the oil in the sump, while 99.9% filtration requires 3-1/2 hours (see chart). Time is of interest only with bypass filtration systems, since we want to know how long contaminates pass through the engine before they find their way to the filter. Full-flow technology makes time irrelevant; none of the oil passes to the engine working surfaces without being filtered so there is zero time for contaminates to circulate.
Craig Richter, writing in How to Make an Old Porsche Fly commented: "This (full-flow filtration) is the single most important thing you can do for your Porsche if you drive it at all." I agree.