Replacing Rear Main Seal – 1 (Garcia)

Part One

On Wednesday I decided to start a project I have put off for months. My 1978 BMW R80/7 had been leaking oil, and I suspected the rear main seal to be the cause. At first the leak was barely noticeable, but grew worse as the months progressed. I called several different shops to find out how much it would cost to replace the seal, and was quoted on average three-hundred and fifty dollars. I had worked as an apprentice mechanic at a motorcycle repair shop in Santa Barbara, CA., under the expert tutelage of master mechanic John Ireland, but I had never tackled anything so extensive on my own. I realized, though, that if I did not deal with the oil leak soon, the clutch could be ruined.

I did not want to pay for a new clutch as well as a new oil seal, so with tremendous amounts of anxiety and trepidation, I began the project by reading about the procedure in my Haynes workshop manual. I concentrated on the parts that did I did not understand, and attempted to visualize the sequence from beginning to end. Next I cleaned the garage and made sure all of my tools were in proper order. I found lots of newspaper, rags, and some cardboard boxes for parts storage. All of this preparation was crucial, since I had recently moved, and was not used to my new garage.

I missed the workshop I had built at my old house, where I had the luxury of space and light. Now I was in a cramped garage with poor lighting, so I would have to adapt. I discovered that I would have to disassemble many of the major components of the bike, as well as buy or fabricate some of the special tools that would be necessary for the job. This seemed daunting, especially since I do not have a compressor, a grinder, or a bench on which to work on the bike. Nonetheless, I started with the basics. I knew that if I kept focused, patient, and creative, the job could be successful.

The bike was placed on the center stand. I disconnected the battery, and removed the seat, fuel tank and side covers. These I placed away from the work area where they would not be disturbed. Next I removed the battery and battery tray, which was incredibly difficult as there is very little room in the frame to accommodate them. The carbs, the air intake ducts, and the air cleaner assembly were all disconnected and removed after the fuel was drained from the carb bowls and fuel lines. The fuel crossover lines were also removed.

The mufflers came next. At this point, I decided to drain all of the motor oil (which was probably not necessary, but would have to be done anyway as the bike was due for an oil change,) and the drive shaft oil. I left the oil in the rear bevel drive (rear differential,) as the entire swing arm would be pulled out and the oil would not be disturbed unless it was tipped forward. (I made sure to place the swing arm and rear drive on a level support after I removed it.)

Following the directions of the manual, I removed the forward clamp and peeled back the drive shaft rubber gaiter. I could now see the four bolts which connected the drive shaft to the gearbox. I tried using my six-sided 10mm box-end wrench and found that it would not fit on the 10 mm, 12-sided stretch bolts. I jumped on my Vespa and rode to Sears. After searching through the various wrenches and consulting with one of the employees, I had the correct wrench. The bolts were tight, but by using the rear brake to prevent the drive shaft from turning, I was able to remove the bolts and put them aside. BMW insists that these bolts cannot be reused, and since they only cost a couple of bucks, there is no need to.

I placed a piece of two-by-six lumber underneath the center stand and removed the rear wheel and tire. I also removed the rear brake pedal, brake rod, and right foot pedal (which I also found to be unnecessary.) Before removing the lower bolts on the shocks, I placed two boxes underneath the swing arm to support it’s weight. With the shocks and the rear rack out of the way, I was able to remove the swing arm. Using a flat head screwdriver, I removed the dust covers and exposed the swing arm adjusting bolts and locking nuts. I noticed the locking nuts were larger than any socket I had in my tool box, so I called a BMW mechanic who said the socket I would need was 27mm. He also mentioned that the socket would have to be machined in order to fit into the frame tube. I rode back to Sears to buy the socket. Unfortunately, I only have 3/8″ and 1/4″ drive ratchets. The 27mm socket only came in 1/2″ drive. I bought the socket anyway and as an afterthought I bought a 3/8″ drive swivel-head breaker bar that I had been eyeing for some time. I knew it would come in handy. On the way home, I raided my fathers tool box and found a 1/2″ ratchet. I got the locknut off and removed the Allen head bolts, which freed the swing arm. I placed it on a Volkswagen spare tire, and this held it perfectly horizontal and prevented the rear drive oil from leaking.

I now had access to the gearbox. Although not necessary, I drained the gearbox oil and removed the left foot pedal and the gear change mechanism. I disconnected the clutch cable after slackening the adjustment bolt, and removed the circlip, pin, spring, and clutch operating lever. I noticed that there were a total of four bolts connecting the gear box to the engine, three of which were Allen head and one of which was protruding from the case and had a 12mm nut on it. I found that my Allen wrench set would not fit on one of the bolts, so I had to go to the store a third time in order to get a 6mm Allen wrench-specifically, one that was “L” shaped. About eight hours after I began, I finally succeeded in removing the gearbox from the bike. I was exhausted and quit for the day.

Part Two

The next day I woke up early and drank lots of coffee before going out to the garage. I knew I had two distinct problems confronting me. First, I would have to purchase an impact tool in order to loosen the six bolts holding the clutch together. Also, I would have to purchase at least three 8mm by 50mm bolts and nuts in order to prevent the clutch from exploding when the six bolts were unscrewed. For the third time in two days, I went to Sears and asked the salesman, (who I now I knew as Art,) where I could find an impact tool. He did not know what it was I was looking for, but he asked another employee and soon I had the tool as well as a 6mm Allen socket. Unfortunately, the Allen socket only came in 3/8″ drive, and the impact tool only came in 1/2″ drive. Thinking quickly, I purchased a 1/2″ to 3/8″ adapter, as well as a 3lb. dead-blow hammer. On the way home, I stopped at a Chief Auto Parts to buy the bolts I would need. The closest size they had were 8mm by 40mm, which was nearly 10mm too short. I was unsure whether the clutch cover bolts were coarse thread or fine thread, so I bought three of each, along with corresponding nuts.

Finally, around 11 o’clock in the morning, I was back at the garage ready to work. I prepared the impact tool and placed the adapter and the Allen socket on it. I began to hammer on the end of the tool, trying to loosen each of the six bolts. None of them seemed to move, even though it seemed the impact tool was turning. Finally, I put the Allen socket on the breaker bar and found that all of the bolts had actually come loose. The tension from the clutch spring made it nearly impossible to see, however. I removed three of the bolts in a triangular pattern. Into these empty holes I threaded the fine-thread 40mm bolts with nuts attached and tightened them evenly. I was now able to remove the remaining Allen bolts. By slowly turning and loosening the nuts evenly on the 40mm bolts, I relieved the pressure from the clutch spring until the cover plate was free. I removed the cover plate, the clutch, the pressure plate, and finally the clutch spring. I had already marked each of the elements with Whiteout to insure that they would fit together the same way on installation. Apparently, this is essential as the flywheel could be rendered out-of-balance if the clutch components are not installed correctly.

The flywheel itself was now exposed. I could see the five bolts that attached it to the crankshaft. I now used a tool I had fabricated. Although the manual describes two possible tools that can be fabricated, I found a piece of metal that resembled the tool and decided to use it. Unfortunately it was not strong enough and broke. I turned and looked at my workbench, and noticed a bracket which I had previously fabricated for mounting a mirror on my Vespa. The piece of steel was extremely strong and was already pre-drilled with correct-sized holes. It fit perfectly, so I placed it over the exposed bolt which protruded from the case, and placed the other end on one of the 40mm bolts, which I then bolted into the flywheel itself. I used the breaker bar again and removed the five bolts. The flywheel came loose after inserting two more 40mm bolts and tugging on them evenly.

I could finally see the rear main seal and the oil pump cover, and to my surprise, the oil seal was in fair condition. The oil was actually leaking from the oil pump cover. I began loosening the four bolts on the oil pump cover, and found the bottom two to be extremely loose. In fact, I was able to remove these bolts by hand. I noticed that the “O” ring was smashed flat and was essentially useless. I removed the cover and cleaned it along with the “O” ring, as I planned to take them with me when I bought replacement parts the next day. Four hours after starting, I cleaned my tools and put them away. My work day was finished.

Part Three

I woke up early again and waited for morning rush hour traffic to die down. I drove about forty minutes to Brown Motor Works in Pomona, CA. I had spent the past two days tearing apart my bike, and had called them from time-to-time to inquire about tools, parts, and procedures. Everyone had been extremely helpful and courteous. I brought along my clutch and the oil pump cover, which I showed to the service manager, Gerry. I asked him about the condition of the clutch and he said it was fine, but suggested some parts that I hadn’t realized I would need, like the “O” ring inside the flywheel. In addition, I was told that BMW recommends replacing the flywheel bolts as well as the drive shaft bolts. He may have even mentioned replacing the oil cover bolts. I went inside the showroom and walked up to the parts counter. I ordered all of the parts I could think of, although I had the sense that I was forgetting something. I was on the freeway when I remembered that I had forgotten to buy a new oil filter. However, I realized, this was not crucial since I had just replaced it only a few hundred miles before.

I got home and cleaned all of the clutch parts as well as the oil pump cover. I replaced the “O” rings inside the flywheel and inside the oil pump cover. I prepared to pull out the old rear main seal. But first I compared it to the one I had just bought. There were significant differences. For example, the old seal was off-white while the new seal was black. Also, the old seal had a spring inside of it while the new seal did not. I was mostly concerned about the outside and inside diameters, however, and these seemed to match. I made sure to check the installed depth of the old oil seal before I removed it. It was nearly 1mm countersunk with the case. Installing the new seal was a slow process, as I wanted the seal to go in perfectly flat and straight. When the seal was in, I placed the oil cover back in place, put a drop of Loctite Red on the bolts, and bolted the cover back on, and made sure to torque the cover to proper specifications. (I have since discovered that one should never use Loctite Red to assemble motorcycle parts, as it is too strong. Loctite Blue, Number 242, is recommended instead…) I didn’t want to over-tighten in fear of smashing the “O” ring, but I didn’t want the cover too loose for fear of having the cover leak and the bolts fall out.

When this was accomplished, I knew I was past the halfway point of the job. Now it would simply be a matter of reinstalling all of the parts that I had removed. (Except, of course, the flywheel bolts and the driveshaft bolts. These would be discarded.) The reassembly went smoothly, except that I had to go to the auto parts store twice for gear oil and a proper funnel. The transmission filler is extremely awkward as it is on the side of the gearbox, just below the gear-change mechanism. I ended up purchasing a pump which made filling the transmission, drive shaft, and rear bevel drive easier. Now that I had the proper tools, I could work quickly without interruption.

As the sun set, I bolted the last items back onto the bike. Although I was tired from two-and-a-half days of wrenching, I was also feeling euphoric now that the job was coming to a successful end. I walked around the bike tightening all of the bolts and filling all of the fluids. A light misty drizzle began just as I finished the job. The bike was now ready for a test drive, so I put the transmission in neutral and turned the fuel petcocks on to allow gas to fill the carb bowls. I noticed that the neutral light was not on, and looking down I noticed that I’d forgotten to reattach the neutral sensor. Since it happens to be tucked away underneath the gearbox, I knew I would not be able to reach it without long needlenose pliers. I did not feel like wrestling with it now, so I simply made sure that the bike was in neutral by rolling it along the floor.

The R80/7 roared to life on the third try, and I let the bike idle and warm up. Although I was exhausted, I nearly skipped with joy into the house to get my jacket and helmet. I walked the bike down the driveway, and sat on it. It felt great to be back on the bike, and to know that I had successfully repaired it. So I thought. I rode the bike around the neighborhood, but it did not feel right. The generator light came on at around 2000 rpm, which was strange. I decided to cut the ride short and headed home slowly and carefully. When I reached the driveway, my heart sank. There was a huge puddle of oil where I had sat on the bike in the driveway. A shiny trail of oil snaked it’s way from the puddle, across the sidewalk, and down the street as far as I could see in the dim, orange streetlights. I turned the ignition off and walked the bike back into the garage, where I once again put it up onto the centerstand. I was hoping that I’d forgotten to tighten a drain plug, but I could see that oil had gushed from underneath the engine. It continued leaking even with the engine off.

I have never felt so disappointed or shattered. I had spent dozens of hours and had followed the workshop manual flawlessly, but now the bike was in worse shape than before. I prayed that the clutch had not gotten slick with oil, but with this much oil flying around the clutch and flywheel, I saw little hope. Completely beaten and frustrated, I simply put newspapers under the bike, turned out the lights, and closed the garage door.

Part Four

That night, all I did was dream about the bike. I went over every procedure in my sleep, step-by-step, trying desperately to find out why the repair did not work. This went on for three nights. During this time, I kept wondering if I should give up and take the bike to a mechanic. I was not looking forward to doing the job again. I couldn’t imagine spending another two or three days working on the bike, but in the end, I decided that it would be best. I wanted to finish the job I had set out to do. If nothing else, I had to know what had gone wrong. I hoped it was some small error or omission that could easily be rectified. After obsessing about the problem, I decided that most likely the pump cover had not been tightened correctly. I had reused the bolts and they were loose when I had first removed the cover. Perhaps the “O” ring had not set properly, or perhaps the pump cover had been over-tightened and had smashed the “O” ring. I was beginning to think of a million theories which would account for the problem, but I knew I would never know the truth until I tore the bike apart again.

I was fortunate to find that the oil pump cover bolts had been slightly oily when I applied the Loctite Red. This had prevented the Loctite from sticking and the bolts came off easily. With the pump cover off, I was now able to see the reason why the oil had spewed past the cover. The “O” ring was completely smashed and flattened on one side. Apparently, it had slipped out of the indentation on the cover just as I had tightened it, which had not only smashed it, but actually sliced part of it off. Since the oil pressure at the cover is over 72 psi, the pressurized oil had literally gushed past the cover. I was still lucky. Although the entire flywheel and clutch cover plate were oily, the clutch itself was dry. It was salvageable. I went back to Brown Motor Works and bought yet another “O” ring, as well as new oil pump cover bolts, flywheel bolts and driveshaft bolts. I also bought Loctite Blue. This time, Gerry told me to clean the oil pump cover and threaded holes, apply Loctite Blue to the dry bolts, and to use a smear of grease to keep the “O” ring firmly in place. I made a concerted effort not to jostle the cover while I bolted it down. This time, the entire job, including driving over an hour for parts, was completed in just eight hours.

I desperately wanted to take the bike on a test drive, but I wondered if the Loctite Blue really needed twenty-four hours to set. I decided that if the bolts loosened and the Loctite didn’t hold, then it would be just as well. I’d rather this happen near my house then in the middle of the desert. Just like the week before, I started the bike and let it idle on it’s centerstand. However, this time I did not walk away but diligently observed the BMW for leaking oil. There were none this time. I took a test ride at sunset. The bike felt better and ran fine. I stopped constantly, checking for leaks, but the bike stayed dry.

Part Five

The next day I woke up late and donned my riding gear. It was a beautiful spring morning, especially for Los Angeles. I could clearly see the mountains, with a light frosting of snow on some of the higher peaks. I warmed up my bike a bit, and started out. I didn’t have a set route, but simply meandered down a long road heading in the general direction of Santa Anita Park. The bike felt okay, but I was still uneasy due to the all of the trouble I’d had with the bike in the past couple of weeks. Just as I began to relax and enjoy the ride, I detected a faint but noticeable squeal coming from the engine. I was at a stoplight, somewhat far from home, so I decided to turn around and head back. Unfortunately, I made a left turn that twisted away from home, instead of toward home. The squeal became louder, so I significantly reduced my speed. All of my senses, I noticed, were heightened as I strained to hear the noise, which came and went. I soon realized that the noise was only present when I pulled in the clutch. I pulled over to the side of the road and tested the clutch. It seemed to work fine, but when I pulled in the clutch lever the squeal was apparent. I limped the bike home and parked it, wondering what the problem could be. I formulated complex theories for this second setback.

When I finally returned home, I rode the bike into the garage and left it. “At least it isn’t leaking oil,” I thought. The next day I determined that indeed there was something wrong with the clutch. For the third time, I removed all of the components leading to the clutch, but this time it only took about an hour. I had figured out some shortcuts along the way, and I could do the procedure in my sleep. I took the clutch apart and checked for signs of damage. There did not seem to be any, but I took the clutch components to Marty’s Foreign Motors in Torrance, CA. I walked up to the parts counter and waited patiently while a woman talked on the telephone. She brusquely asked what I wanted, and just as I began to explain another employee interrupted and the two of them began to have a conversation. I waited again until I was eventually helped. I found the parts employee to be unfriendly and not very helpful. However, I asked one of the mechanics to diagnose the noise, and he said it was most likely caused by a failure to grease the clutch spring. He gave me some of the proper grease recommended by BMW, and I drove home. Although the woman at the parts counter had been rude, at least the mechanic at Marty’s had been helpful.

I lubricated everything correctly and reassembled everything once again. When I started the bike, I did not hear the noise, until perhaps two minutes later. I was genuinely and officially puzzled, and I finally decided that I would take the bike to Brown’s Motorworks when I got the chance. I vowed to forget about my motorcycle until I could deal with it. The problem continued to nag me. I called my old boss, John Ireland, to see if he could give any insight. He told me to look for a couple of things, but I assured him I had already looked at these items. I truly believed I had done everything correctly, so in the end I mentioned that I was going to give up and let the experts handle it. We said our good-byes and I hung up the phone.

For some reason, simply talking with him inspired me to have one last look at the bike. I started the bike and listened to it idle. It sounded fine, but when I pulled the clutch in, the bike would whine and squeal. I listened even more carefully, and actually got on my knees next to the gearbox. The sound was actually coming from behind the clutch — perhaps the gearbox itself or the operating mechanism. I turned off the ignition and removed as much of the clutch operating mechanism as possible. Instantly, I noticed something amiss. The cage-needle bearing, (the one I had carefully removed and lubricated,) was destroyed. For the fourth time, I removed the back end of the bike so that I could pull out every part of the clutch operating mechanism and inspect it. What I found made me feel slightly embarrassed. Apparently I had referred to the wrong picture when I had reassembled the operating mechanism after lubricating it, which resulted in the bearing literally exploding inside the gearbox. Part of the bearing cage had actually welded itself to the clutch push rod. I removed all the bits and determined that most of the parts could be salvaged. I went to Brown’s and they reconfirmed my suspicions. I bought the necessary parts and reinstalled them. The bike ran smoothly, and the clutch worked perfectly and quietly.

Part Six

I felt a sense of triumph and vindication, and I was glad that I had decided to repair my bike. Even though the experience was frustrating and draining at times, the fact that I exceeded and expanded my knowledge and experience made the whole effort worthwhile in the end. This morning I rode up the driveway past the puddle of oil from the second day’s failure. Underneath my helmet I wore a huge grin.

Member Login