“You have an oil leak,” my friend Mike remarked last July, looking down at my engine. The bike was on its sidestand, so you could get a good angled view into the skid plate, where a glistening smear foretold almost a year of diagnostic troubleshooting.
I quickly localized it to the front lower portion of the engine. It would drip down the front and collect in the skid plate at the bottom. Seeing the oil wasn’t the problem; determining its source was.
Oil can travel a great distance on a dirty engine. The dirt soaks up the oil and transports it across the bike so that diagnosis is like a shell game. Is it soaking up from the bottom or dripping down from the top? All you’ve got to go on is a blow pattern like a gruesome crime scene.
I decided to approach this like Hercule Poirot.
- Sump plug sealing washer
- Crankcase gasket
- Starter motor O-ring
- Base gasket
Sump plug sealing washer
Following the principle of Occam’s Razor, I started with the simplest explanation. I had just changed the oil and cheated and re-used the sump plug washer, so that was my first guess. You can re-use those copper washers a few times, especially if you braze them between uses, but it’s a crapshoot. Yes, there’s oil up the front of the engine, but like I said above, oil can defy gravity on a dirty engine.
I replaced the washer at the next oil change, but the leak persisted. It wasn’t the sump plug washer.
Someone on a forum said the crankcase on these bikes is prone to leaking. This theory was supported by the fact that the leak only occurs at high revs under pressure. So last autumn, while doing some other maintenance on the bike, I torqued all the crankcase bolts. Lo and behold! A few in the front of the engine were under-torqued. Surely I’d found the source this time.
I hit the highway the following weekend and at my first rest stop I took a look in the skidplate: that familiar smear of liquid gold. It wasn’t the crankcase gasket.
The starter motor O-ring.
This is when things got interesting. The leak seemed to be coming from somewhere under the starter motor. There’s a single O-ring that prevents oil in the crankcase escaping at the interface with the starter. Again, I’d had the starter off for another job and didn’t replace the O-ring, which my shop manual recommends. Since the starter connects with the crankcase, the pressure at high revs could blow the oil past the the O-ring.
I replaced the O-ring, waiting the requisite two weeks for this $5.14 part to arrive from Germany. The leak persisted. I decided to get serious.
The best way to source an oil leak is to clean up the engine, take the bike for a short ride, then look. In theory, you should see a trickle coming from the source. So that’s what I did. When I got home, I took the starter off and this is what I saw.
I said in theory. There’s definitely a trickle coming from that front corner of the engine, but there’re also drops near the front??? Could it be getting past the O-ring, dripping along the underside of the starter motor, and dropping onto the crankcase?
I was down to two suspects—the starter motor O-ring and the base gasket. I was nervous about torquing the gasket. Maybe it’s not a big deal, but those big, crucial gaskets scare me. Mess that up and over-tighten and you are pulling the engine. And in all my research, I had come across only two instances of a weeping base gasket. I decided to stick to the O-ring theory a little longer before interrogating the big boy.
I picked up some Optimum Black gasket maker. The packaging said it resists vibration (important on a thumper) and is “one of the most advanced, maximum flex, maximum oil resistant RTV silicone gasket makers available.” I added a smear on the O-ring, let it cure 24 hours, and tried again.
Still no luck. I added a bead around the neck of the starter next to the O-ring. No luck.
I put some Optimum Black in the channel for the O-ring, let it cure, added the O-ring, and added a bead at the neck for good measure. Nyet! I found a slightly bigger O-ring in plumbing and tried that. Nope.
I was beginning to think it wasn’t the O-ring.
I asked on my bike forum and somebody suggested the oil could be coming from inside the starter. I hadn’t thought of that. There’s an oil seal and bearing inside. If the seal is finished, oil could be getting inside and then dripping out of the starter. If the bearing is worn, it could be causing vibration that is preventing the external O-ring from completely sealing. I had noticed slight pitting on the underside of the starter neck, which supported the theory of a worn bearing.
So the starter came apart. I pulled the old bearing using a valve puller and tapped a new one on, adding a new, greased oil seal. I was hopeful this time. There were a few symptoms supporting this theory, not the least that my bike now has over 90K on it, and with the crank revving at 5K rpm, I’d be surprised if that bearing weren’t worn.
Once all back together, I took Bigby for another short test ride. The bike was still leaking. Now I know why people say fixing an oil leak is hard.
This is when my cursing and ruminating and otherwise surly or downcast behaviour elicits from my wife the question, “Why don’t you take it to a professional mechanic?” Actually, others were beginning to say the same, including a club member. My response: “What can he do that I can’t?” (And yes, it’s almost always a “he.”) And the amount of time spent diagnosing this kind of problem would cost a fortune. It was time to concede the inevitable: I’d have to tighten the base gasket.
The base gasket is Item 4 in the diagram below. Tightening it sounds easy—and probably would be on any other bike—but it isn’t on a BMW. The little M6 bolts on the left side of the bike (Item 3) are nested into cutouts in the corners of the engine block, so accessible only with a torx key. I’d have to buy a set of keys and torque those bolts by hand. They are 10 Nm and I think my feel is getting pretty good, so that was the easy part. I nudged them, that’s all.
The other side has two M10 x 223mm bolts that go all the way through the head, the block, and into the base (see Item 1 above), and they are only accessible beneath the rocker (valve) cover. I’d have to open up my engine.
So off came the plastics, the battery, the airbox, the oil tank, the battery tray, the starter relay, the electrical box cover, the heat shield, the intake manifold, the throttle cable, the spark plug coils, the coils holder and, finally, the valve cover.
With the rocker cover off, I could now tighten those long bolts. I was nervous about tightening those babies! While the little M6 bolts were only 10 Nm, these were 60. I’d need my big 1/2″ torque wrench. Someone on the forum had suggested I back off the bolts a bit first, just in case they are seized, then retighten. I decided to aim for 62 Nm.
The one at the back went smoothly enough; I got an accurate reading from the wrench so moved on to the crucial front one. I could not get an accurate read with the wrench on this one but I did move it a few times, as much as I dared. My experience told me that I’d better stop there or risk over-tightening if I continued. Sometimes you just can’t trust a torque wrench; you have to trust your gut first. I hoped I’d moved it enough or I’d be taking the top end of the bike off again.
That was the tough part done. I took a break, made a tea, and slowly reassembled the bike.
When I replaced the starter, I added another smear of Optimum Black on it for good measure. I also torqued all the crankcase bolts again, replaced the oil filter and added a smear of gasket to the cover sealing ring, and torqued all the engine mountings. Basically I torqued every bolt on the right side of the bike! Desperate times call for desperate measures.
I was so worried about this test I actually didn’t ride for another whole day. I was out of leads, and if the leak were still there, I’d be at a dead end.
The next day, my wife and I headed off for an afternoon excursion into Ontario. We packed a picnic lunch and headed west along the St. Lawrence River. After about an hour and a half of riding, my wife spotted a church overlooking the river where we could stop. I pulled in, turned off the bike, climbed off, and with God and my wife present for emotional support, I looked into the skid plate.
No oil. I looked under the starter motor. No oil.
I’m not a superstitious man, but I dared not celebrate just yet. I didn’t want to jinx myself. But on the way home I intentionally kept the revs up into the 5K+ range to stress the engine. Once back on the driveway, I looked again: no oil!
Now we did celebrate. We poured two beers and toasted the end of the oil leak. This was one of the hardest jobs I’ve ever done. It took me almost a year off and on to diagnose the problem and over a week of full time work to get to the bottom of it. A big thank you to the guys at The Chain Gang for their help. Bigby is now finally ready for the 2020 season.
The finishing touch is a new custom decal designed by my friend Brian Chu at Brian Chu Design and Illustration, Inc. in Calgary. Thanks, Brian.
4 thoughts on “The Oil Leak”
WoW that was a big job.
You must know every part of that bike now.
No but I’m learning A LOT!
Nice writup! I am currently going through something very similar with a thruxton. I am convinced the leak is coming from the starter. But now I’ve replaced the starter shaft seal and oring and the leak is still there…
Good luck! Persistence pays off.