View Single Post
      09-23-2013, 10:15 PM   #2
regular guy
Lieutenant Colonel
427
Rep
1,947
Posts

Drives: Sprint car
Join Date: May 2013
Location: California

iTrader: (0)

Historical Background of S65 Bearing Issue

Three truths of S65 bearing/side clearance
There are three things we seem to know for a fact:
  1. There is a long standing clearance-to-journal ratio best practice rule that factory and racing engine builders alike have followed for 50+ years. This best practice clearance ratio is well documented in many of web sites, and recommended by Clevite, the maker of three years worth of S65 engine bearings. The S65 clearance ratio for 2008 - ~2010 engines is smaller than this minimum recommended best practice value. (1, 2)
  2. Most of the bearing wear patterns we're seeing all seem to match the Clevite online/interactive bearing failure diagnosis web site, example #12 "Oil Starvation / Marginal Oil Film Thickness." The first of many causes mentioned by Clevite for this type of failure is "too little bearing oil clearance." (3)
  3. There is also a long standing rod side clearance best practice rule that factory and racing engine builders alike have followed for quite some time. Best practice clearance ratio is well documented in many of web sites. The S65 engine cuts that clearance significantly.

Notes:
1. Artitle: Clevite Bearing Clearance White Paper, Pages 16-18
2. Article: Geometrical parameters of engine bearings
3. Interactive: Clevite online/interactive bearing failure diagnosis web site

Historical Background of S65 Bearing Issue
It was very easy to ignore the very first blown motors because most seemed to originate from highly modified cars. Very early there was one blown NA motor from a guy in Thailand "custom tuned" by a local tuner. His motor blew and it cost a lot of money to replace it. Tuning by a local, unknown tuner was another reason to ignore this blown motor.

But about the same time, the first one or two blown motors showed up on bone stock M3's as well. These cars were totally stock with very low mileage. As time went on, more and more blown motors showed up, again NA motors with low miles. It seemed if you made it past 25,000 miles, then your motor would likely survive the warranty period.

Various theories were discussed to explain this phenomena. The theories ranged from inferior oil pump design to bearing clearance issues. But nobody had any data to support one argument over the other.

Discovering S65 Bearing Issue
1
2
3
4
5
6
7
8


The bearings shown above came from a supercharged stroker motor. The engine was 4.6L and ran a very modest 6.0 PSI boost. After 24,000 miles, the engine was disassembled in July 2010 to rebuild as a low comperssion 4.7L motor. The bearings were photographed and stashed away. I always thought the bearings looked pretty "ratty" but at the time, nobody thought anything else about it. Today these photos might be ground-zero, exhibit-one in the discussion of rod bearing clearance.

By September 2011, the engine was ready to be reassembled when the race shop made the discovery. Van Dyne Engineering in Huntington Beach called to explain that the engine couldn't be reassembled because the bearing clearance was too small. They measured as little as 0.0011 inch rod bearing clearance, which Van Dyne said would lead to oil starvation and potentially catastrophic engine failure. To confirm these measurements we gathered one more factory crankshaft, and samples of two sets of connecting rods from different engines. Van Dyne confirmed the measurements all matched, and the factory BMW clearance was dangerously too small.

Van Dyne explained that the industry standard clearance should be 0.001" clearance per inch of journal diameter. The BMW factory measurements were approximately half those standard values. Van Dyne proposed a solution to send the crankshaft out for machining to have the journals resized to allow proper clearance. By November 2011, the machining had been completed, and the engine was ready for reassembly. During the engine assembly process, the clearances were all checked and verified to ensure the clearance problem was fixed.

Connecting Rod Side Clearance Issues

Discovering rod bearing clearance issues wasn't the only adverse discovery we made while assembling this motor. In addition to measuring bearing clearances, Van Dyne also measures connecting rod side play, crankshaft end play, valve opening/closing angles, valve spring seat pressure, and many other things. Connecting rod side clearance is measured by sticking a feeler gauge between the connecting rods to see how much space is between them. Engine builders like to see the connecting rod side clearance in a very specific measurement range. The measurement is taken when the bottom end is mostly assembled: the crankshaft is in place, and the pistons and connecting rods are attached and torqued down. If the side clearance is too little, then oil cannot escape and extra heat is generated by the friction of the connecting rods colliding against each other.

The crankshaft itself can be damaged as well. When the rod side clearance is too tight, as the engine heats up, the rods will swell and the crank will shrink. The tight clearance gets smaller and smaller until the rods collide and start riding against the crankshaft journal fillet. The fillet can be worn down, and damaged by this process, and the metal shavings it produces will damage the connecting rod bearings and other engine parts as well.



To measure the connecting rod side clearance, Van Dyne instinctively grabbed the feeler gauge for the normal, industry-accepted side clearance measurement. The feeler gauge wouldn't fit. Van Dyne grabbed the next feeler gauge half the thickness of the first: it still wouldn't fit. He then grabbed the next feeler gauge half that size: it still wouldn't fit. After four different tries, Van Dyne finally found the feeler gauge that fit. The side clearance was approximately 1/4 the size of industry expected clearance. The clearance was the thickness of two human hairs.

Without hesitation, Van Dyne grabbed our spare set of Carrillo rods that had been in the motor previously. He grabbed them as if he expected to find evidence of possible damage. Within a split second, Van Dyne pointed to the side of the Carrillo rod and said:
See that tiny blue spot? That's caused by heat from these two rods rubbing together because they are so tight. See that scrape mark? That's also caused by these two rods rubbing together.
As Van Dyne (and kawasaki00) explained, proper rod side clearance is essential to allow oil to escape. As the engine heats up, the connecting rods grow and the crankshaft shrinks until the connecting rods are touching each other. The oil can't escape, and extra heat is generated by the friction of the connecting rod collisions. The side of the crankshaft journal could also be damaged by the collisions as well. To me, the situation sounded pretty serious, and I thought it was going to mean another multi-week setback for the project.

Van Dyne deemed the situation so serious that he asked if we could go 60-minutes across town and bring back a factory BMW crankshaft (ours was a billet "stroker" crankshaft) and some factory BMW connecting rods (we were working with two sets of Carrillos). Van Dyne wanted to measure the factory journal width, and factory connecting rod thickness to see if the stroker crankshaft or Carrillo connecting rods were to blame.


Measuring Connecting Rod Side Clearance

Almost in disbelief, Van Dyne started taking measurements. First he measured with a set of calipers. Then when he realized the measurement didn't lie, Van Dyne knew he had to take accurate measurements with micrometers so he could make the necessary calculations to fix the problem.



To make sure the measurements are accurate, first the micrometers were calibrated to a known good thickness.



Van Dyne then started to measure both crankshafts: our billet crankshaft, and the factory crankshaft. The journal thickness measurements both matched. Next, we need to check the connecting rods to see if the Carrillo's were machined thicker than factory rods.



Following the crankshaft measurements, Van Dyne measured the connecting rods to see how they compared. Van Dyne measured both Carrillo and factory connecting rods.



The measurements showed the Carrillo and factory connecting rods are the same thickness. This is positive proof that the factory BMW crankshafts are machined without enough rod side clearance AND without enough rod bearing clearance. It's a double-whammy against proper engine oiling and operation.



Finally, Van Dyne set out to "fix" the clearance issue. I was relieved when Van Dyne said he could fix the issue immediately – as I feared project would suffer another long delay. Without hesitation, Van Dyne pulled out the bottle of DYKEM and started color coating one side of the connecting rods. This would be the side he planned to machine, and the DYKEM would serve as a visual reminder which connecting rods were finished, and which ones weren't.




Van Dyne's machine shop was able to grind down each connecting rod to the exact specification he wanted. When he reassembled the bottom end a few hours later, the measurements all agreed. The rod side clearance issue was now fixed.

What does this mean to you?

To get second opinion of the seriousness of this issue, I emailed kawasaki00 and I shared the actual specs with him (I'm not sharing them here). I asked if this was a big issue or not. Without hesitation, kawasaki00 agreed with Van Dyne that this was indeed a big issue, and he thought I was lucky to have an engine shop capable enough to measure and fix this problem. As he said, we really dodged a bullet.

I'm not an engine builder -- so I'll let kawasaki00 and BMRLVR chime in to answer questions on this topic. But I can put the pieces of the puzzle together and tell you what they mean to me. BMW gave you the thickness of two human hairs to let the oil squeeze out of the connecting rods. I don't think it takes an engine builder to realize that oil can't escape with side clearance that small. It makes me wonder how heated that trapped oil might get and how fast it might wear out my bearings -- bearings that already suffer from too little clearance of their own. Then with the connecting rods scraping against each other after they swell and the crankshaft shrinks, I can also see how metal shavings could get generated to destory the bearings and other engine components.

Here's some photos of the micro "blue marks" and abrasive scuff marks left on my Carrillo connecting rods from the previous engine build with improper side clearance.

1
1
2
2
3
3
4
4
5
5
6
6
7
7
8
8


A few of the photos above are recreations because I didn't take actual photos of the calibration process, nor the actual scuff marks on the day of the engine build. I took those photos today to show what the process looks like. The scuff marks are the actual scuff marks on the same Carrillo connecting rods.

Complete photo library on this issue:

More photos are available at the following links.
Instrument Calibration and Connecting Rod Measurements
Side Play Clearance Measurements
Side Play Clearance Adjustments
Side Play Scuffing Evidence

Mahle-Clevite White Paper on Bearing Clearance Issues

This quote comes directly from a Mahle-Clevite white paper on bearing clearance, then I'll summarize at the end.
For most applications .00075 to .0010” (three quarters to one thousandth of an inch) of clearance per inch of shaft diameter is a reasonable starting point.

...

Using this formula will provide a safe starting point for most applications. For High Performance engines it is recommended that .0005” be added to the maximum value determined by the above calculation. The recommendation for our 2.000” shaft would be .0025” of clearance.

...

High Performance engines on the other hand, typically employ greater bearing clearances for a number of reasons. Their higher operating speeds result in considerably higher oil temperatures and an accompanying loss in oil viscosity due to fluid film friction that increases with shaft speed. Increased clearance provides less sensitivity to shaft, block, and connecting rod deflections and the resulting misalignments that result from the higher levels of loading in these engines. Use of synthetic oils with their better flow properties
can help to reduce fluid film friction.

...

Use of these coated bearings may result in slightly less clearance than the uncoated CLEVITE 77® high performance parts for the same application. This will typically be in the range of .0005.” This is because the coating, although expected to remain in place during service, is considered to be somewhat of a sacrificial layer. Some amount of the coating will be removed during break-in and operation, resulting in a slight increase in clearance.
Here's what this means to you. If you run high horsepower and high RPM, then you need extra bearing clearance, not less of it. The coated bearings are great, but you have to size your journals for them; and as Alekshop mentioned in a previous thread, you shouldn't have coated parting lines. If you have coated parting lines you must remove the coating in this area before using these bearings.

Mahle-Clevite recommends adding an extra 0.0005 for good measure for high horsepower, high RPM, and coated bearing applications. To see some numbers in real life, please see the following examples.

Factory clearance: 0.00125
Factory clearance with Calico coated bearings: 0.00085 - 0.00105 (16-32% smaller).
Factory clearance with TriArmor coated bearigns: 0.00075 - 0.00095 (24-40% smaller).

Recommended clearance for mains (60mm journal):
0.00100/inch + 0.0005 = 0.00286
0.00075/inch + 0.0005 = 0.00182

Recommended clearance for rods (52mm journal):
0.00100/inch + 0.0005 = 0.00205
0.00075/inch + 0.0005 = 0.00153

So there you have it: ideal mains should be 0.0018 - 0.0029 and ideal rods should be 0.00153 - 0.00205. You're given 0.00125 clearance from the factory and some may be thinking of reducing it another 16-40% with coated bearings without sizing the journals because they believe it gives them extra protection. The extra protection is true if you kept all things the same including the bearing clearance; but reducing the bearing clearance by 16-40% to get the extra protection of the coating is not recommended.

Some Photo Evidence of Oil Starvation

Time to add some pictures to the discussion to see what a bearing looks like due to lack of clearance. The first few pictures come straight from Mahle-Clevite. Thanks to BMRLVR for posting the B&W version of the Clevite bearing failure guide. I found the color version of the same thing. It looks like the color version has better pictures, but the B&W version might have some better illustrations and explanations. So here's both versions for reference.

B&W: http://www.wilmink.nl/Clevite/Clevit..._tech_info.pdf
Color: http://catalog.mahleclevite.com/bearing

Using the color bearing failure guide, see example #12 "Oil Starvation / Marginal Oil Film Thickness." The two pictures below come from this example. According to Mahle-Clevite, the number-one cause of failure of this type is too little oil clearance.





Next, let's compare these reference photos to some actual S65 bearings from various engines. Where possible, the mileage and circumstances will be mentioned. These photos are all found in other threads on this forum.


Engine-1: Modifications unknown, stock internals, Mileage: 24,000.




Engine-2: Bone Stock, stock internals. Mileage: ~4,000

This bone stock engine suffered from main bearing failure. Notice the wear spot on the main bearing and the appearance of a channel the oil made to get around it. The connecting rod bearings aren't much better. Many of the connecting rod bearings show signs of excessive wear due to oil starvation (too little clearance).













Engine-3: Upgraded internals, supercharged, Mileage: 24,000






Engine-4: Stock internals, supercharged, Mileage: 72,000




Engine-5: Stock internals, supercharged, Mileage: 90,000



The connection between these pictures and the reference Mahle-Clevite pictures is obvious. Even though there can always be multiple causes of oil starvation and bearing failure, the number one cause of the damage shown in these pictures is listed by Mahle-Clevite as too little bearing clearance.

I have more photos that I will post in the Bearing Photo Database. I've got at least one set, maybe two sets of photos of broken (snapped) S65 connecting rods with the big end of the connecting rod turned blue due to excessive heat (oil starvation due possible bearing clearance).

I hope these photos have been helpful to see with your own eyes what this problem looks like, and why it's happening.

Last edited by regular guy; 06-19-2015 at 08:49 PM..
Appreciate 4
DrFerry6722.50
kafandi13.00