http://www.langracing.com/finding-a-...aring-failure/
September 10, 2013
When we first began looking at the S54 knowing its history of rod bearing failure we immediately noticed how small the rod bearings were compared to other engines we’d built. To be more specific, the width of them is relatively narrow compared to other engines that are designed to make that much power. When you look at the rest of the engine you see similar ideas in other areas. BMW made every attempt to eliminate friction surfaces in this engine to increase efficiency and power. This is, of course, is every automotive engineer’s goal. The expense of doing this of course can mean that engine reliability can suffer. Every engine will have a weak point at which it will fail first, or at least most commonly. For the S54 this is the rod bearing. For another engine it might be the rod or the cylinder wall, or something else. Whether its a design failure or intentional, the S54 has a very well documented history of consuming rod bearings at a rate faster than desired.
I’ll do my best to illustrate this with words but really the easiest way to explain it is by simply looking at the engine internals on an S54. Here at Lang Racing Development we have engineered racing engines for some time now, not just the S54. That experience gives us a broad perspective when we look at every engine.
To understand why the rod bearing fails on this engine you first have to understand how the bearings on the crankshaft work. The bearing is a wearable surface, i.e. softer than the material of the crankshaft and the rod itself which are both, typically 4340 steel, one of the hardest/strongest steels available. If those two were to contact each other we’d be in a world of hurt as they would wear each other out. The bearings job is to prevent that from happening. The bearing is designed to be a relatively soft metal alloy so that when it does touch the crankshaft journal it wears away, rather than wearing the crankshaft. Now ideally we keep an oil film between the journal of the crankshaft and the bearing. We choose an oil viscosity based on the bearing clearance we decided on when we assemble our engine.
When BMW built this engine they decided some numbers they thought would work really well. We found out though in 2003 that they made a mistake. They made their recall explaining that there was a problem with the rod bearings they manufactured in the early years. The explanation they gave was that the rod bearings were manufactured incorrectly resulting in too tight of a rod bearing clearance when the engines were assembled. This resulted in inadequate oil film between the crank journal and the rod bearing.
My personal suspicion is that this is merely an excuse and that that the recall was actually an update to a bearing with larger clearance. BMW could never admit openly “our original engine design had too small of a rod bearing clearance and made the bearings spin”. At the time of the recall we noticed that the rod bolts/entire rod changed design as well, but of course there wouldn’t be a recall on rods themselves, that would be a full rebuild of an engine. And finally, the most vital component that couldn’t be recalled in a cost effective manner would be the crankshaft. Does the rod bearing recall solve the problem? My answer is no, and I see this time and time again that even engines that have had the rod bearing recall done fairly recently still exhibit a large amount of rod bearing wear.
Let’s get down to what I consider the main problem: when the piston is on its compression stroke and the mixture is ignited a tremendous pressure is imparted on the top of the piston, down the wrist pin, through the rod, and into the rod bearing. Some engines have trouble or weakness upwards of this chain but what the S54 does is squeeze out all the oil causing the bearing to wear on the top side. This is a simple conclusion to reach when you dissect S54′s day in and day out, the bearing wear tells the story. This situation is also what makes me a bit uneasy about the common suggestion that a coated rod bearing is a great solution on the S54. If a coated bearing is adding material to a rod bearing then, obviously, it must also be reducing the bearing clearance on the crankshaft journal. Some rod bearing treatments are designed to allow the bearing to retain oil better, which, if they do actually do that would be helpful. Still, I don’t believe that these bearings adequately address the original engine design flaw.
There are a couple ways to solve this problem, BMW tried the simplest. Increase the oil viscosity and increase the bearing clearance. This provides a higher volume of viscous oil for the combustion pressure to act upon. Does this help the problem? Definitely. Another option would be to increase oil pressure, this is one of the reasons that some people say all S54 race cars should have a dry sump oil system. This helps not so much because it prevents oil starvation but also because you are able to change gear sizes to obtain the oil pressure you desire.
To really get down to properly fixing the engine we have to go back and re-engineer it. The best way would be to make a new billet crankshaft (like we do for our stroker kits). A cheap alternative is to have the stock journal reground to accept a wider bearing. To do this we also have to choose a bearing that has a smaller diameter. Conveniently this also means that we can fix some crankshafts that have already spun bearings. We grind down the journal and widen it to accept the bearing for our new rod. For economics we choose a rod bearing that is common and easily available. This also means we have a wide variety of rods to choose from that are available in different lengths. Once we choose the rod, which can vary depending on customers preferences, we then measure the compression height we desire and order our custom piston from one of the various suppliers we use. Obviously what I’m describing takes quite a bit more thought than simply reassembling a stock engine and as such the labor cost is higher, and that might scare some people away.
I like to let them think about the cost for a new set of OEM rings, or pistons, or rods from BMW and compare those costs to the cost of some aftermarket rods/bearings/rings from aftermarket sources. Often times the quality is better and the cost is significantly less. The end result is an engine that costs very close to what an OEM rebuild would cost and ends up with lighter weight components that are fully balanced and engineered. More over some of the compromises that BMW made on their production engines don’t need to be made when building custom engines for race applications.
Some might say that its best just to replace the S54 rod bearings as part of regular maintenance, perhaps every 50k miles or so. I think this is a good suggestion but it also carries an expense and is a bandaid on a permanent problem. If a person is in the position to do a complete rebuild and intends to drive the car aggressively I think modifying the crankshaft for future reliability is a wise decision to minimize future maintenance and reduce the chances of catastrophic engine failure.
Even more so I think it makes the most sense for those in the position that their engine has already spun a bearing and has damaged a crankshaft. At this point they are left with the decision to either purchase a used crankshaft or try to repair their current one. The cost of modifying the crankshaft journal for this wider bearing is only slightly more than a standard crankshaft repair when a rod bearing has spun.
If a rebuild is in your future on an S54 I encourage you to investigate the costs of a standard rebuild and compare those costs to our rod bearing modification kit.