Saturday, April 7, 2007
I guess it's a sign of owning an older BMW that I consider three weeks a "long time" since I last logged anything, but following the recent valve cover gasket work the car has been running beautifully. That's not to say, of course, that I don't have work to do on it.
The next mechanical work I plan to do involves cleaning the idle control valve as preventative maintenance. When I perform a cold start the engine normally revs to around 1100 RPM and slowly works its way down to the normal idle speed of 600 RPM as the engine and exhaust temperatures rise. That transition is supposed to be smooth and gradual, but on a few of the colder mornings lately I've noticed the RPM move down in a stepped fashion. The effect is very subtle, to be sure. Most people wouldn't even notice it, but I do remember seeing that chracteristic years ago...right before my ICV got stuck and turned my throttle into a deadman's switch. As my technician told me recently when I queried him about this -- "they just get lots of carbon built up in them...it's not like the unit itself fails", so it seems to me a little TLC now in the form of a can of carb cleaner will do wonders.
I'm also planning to replace the secondary air pump check valve, again as preventative maintenance. The last one failed at roughly 60K miles in service and allowed exhaust gases to work their way back into the pump and destroy it. Since the air pump isn't exactly cheap, spending $70 on a check valve that's easy to replace seems like the right thing to do.
Oil Service with Oil Analysis
Things were so quiet with the car lately that I briefly lapsed into complacency. I was reviewing my maintenance schedule spreadsheet this week when I suddenly realized I'd accumulated over 4500 miles and it was time for another oil service. I picked up the parts at my local dealer Friday and performed the work this morning. Everything went as planned and I had it done in less than 30 minutes: 10 minutes to do the actual work and 20 minutes to let it drain. In other words, it was a pretty standard oil service.
Given that I passed 140K miles recently, however, I decided to do something a little less "standard" and put the engine on a oil analysis program. I'm looking for all the usual wear metals, but in particular I'm looking for coolant in the oil, as this would be a sign of a deteriorating head gasket. While this is the first time I've ever done an oil analysis on any car I've owned, I'm not new to oil analysis since I've used it in my aviation pursuits for years to make sure we recognize engine problems before they affect our operations. As we pilots say, the worst thing you can do is turn off the fan that keeps the pilot cool. Do so and watch him sweat!
As many proponents of oil analysis will tell you it's not so much the absolute values of the numbers in the report that are important in as much as the TREND of those numbers. In fact, it was the trend of certain higher wear metals over a couple samples on one of the airplanes I used to fly that pointed to defective rocker arm bushings. We pulled the rocker boxes off and found a couple bushings on a recently-replaced cylinder loose and wearing excessively. I don't really like to think what would have happend if we weren't doing oil analysis and had just continued to blindly turn the key every time we went flying, but the point is oil analysis isn't marketing -- it really works and is an important part of any preventative maintenance program. Automotive engines have become so reliable that I doubt there's much need for analysis in the first 75K miles, but it could be argued that the analysis is pretty cheap insurance, and the earlier you start the analysis the more trend data you have to diagnose potential issues later in the engine's life.
While I don't plan to do it on a regular basis I ordered a Total Base Number (TBN) test for an additional $10 (total of $30). This will tell me how much life the oil has left in it after running in my engine for 4700 miles and it should satisfy my curiosity as to how well BMW oil stacks up against other oils in terms of longevity. It will probably be a week or so before I get the report, but I'll post it here when I get it.
With a 10% CCA discount and state sales tax of 7% essentially erasing that, I paid $41 for 6 quarts of BMW 5W-30 Synthetic and the filter kit. The oil analysis is normally $20 and the TBN test is $10, or $30 total. My dealer gets 0.6 hours to do an oil service, which at current labor rates & tax amounts to a labor savings of $77. Not much, but better in my pocket than theirs.
Mileage: 142035, Parts (including oil analysis kit) $70, Labor Saved: $77
Thursday, April 12, 2007
Oil Analysis Results
Just as the vendor (BlackStone Labs) promised, they emailed me a copy of the report in advance of receipt of the paper copy. The results are pretty self explanatory and indicate that the engine is running like a top at 142K miles. All the wear metals are well within reason and the oil reportedly contained no coolant, so I'd say it's good to go.
I find it interesting that the analyst pointed out that the viscosity of the BMW 5W-30 oil is slightly high for a 30W oil. And that's because it's supposed to be. When oils are classified as a 30 weight, that actually refers to a range of viscosities, and BMW 5W-30 synthetic oil is known to be at the high end of the 30 weight range. In this way it behaves more like a light 40 weight oil than a 30 weight and it is this quality that allows it to meet BMW's own LL-01 specification. That's why you can't necessarily use any generic 5W-30 oil and expect the same results. Lighter oils are known, for example, to cause the vanos to chatter as well as other top-end noise. While I think BMW 5W-30 is a great oil at a great price, if you're hell bent on finding something cheaper, just make sure you buy something that's LL-01 approved. For reference, Mobil 0W-40 meets the LL-01 specification but its 5W-30 does not.
The TBN value doesn't surprise me. There's a reason why BMW thinks it can get away with 15K oil changes. The oil may reach the minimum TBN just as the oil reaches 15K miles in service. But of course oil serves many purposes and the TBN number does not indicate the oil's ability to hold particulates in suspension, for example. Keep your oil in service until the TBN number says it's finished and it's likely that oil has already contributed to a build-up of sludge in the engine.
The fact that there was no water found (or at least it was below their measurable threshold) is likely because I drive it every day, at highway speeds, and get it up to normal operating temperature sufficient to burn off any water that may find its way into the oil due to the normal process of combustion or via a form of condensate. This explains one of the reasons why cars that drive short distances actually need their oil changed more often than those run at highway speeds every day.
All in all, I think the analysis was worth it. I'm not sure if I'm going to do the analysis at every oil change (something tells me I'll do it every other change just to develop a trend), but I do know that I won't bother with the TBN test again unless I change oil brands (very unlikely) as the results prove the BMW 5W-30 oil holds up well in my application.
If you have any questions about oil analysis, check out the BlackStone Laboratories site.
Thursday, April 26, 2007
As I often do, I filled up the tank on the way home yesterday. The attendant managed to top off the tank with a total of 13.5 gallons, or about a gallon higher than I expected, but it didn't surprise me since the guy did the classic "oversqueeze" at $39 and for some unexplained reason decided to proceed to $40 in spite of the fact that I was paying by credit card. Aside from pondering why the idiotic New Jersey state legislature considers it illegal for an average citizen to pump our own gas when virtually every other state in the union couldn't care less, I didn't think much of it and continued home.
I arrived at my car this morning to go to work and stopped just short of opening the door when I smelled the distinct aroma of gasoline. I did my best impression of a bloodhound and walked around the car, then around the garage, all in an effort to exonerate my baby. I quickly realized that the smell was indeed coming from the car and was strongest near the driver's door so I took a quick peek under the vehicle only to find the bottom of the fuel tank wet with gasoline. It wasn't a steady leak and a quick turn of the key in the ignition showed I still had a full tank, but given that I had no idea of the real extent of the leak or where it was coming from, and given its proximity to the exhaust, my piloting judgment kicked in and I stopped the "preflight" right there and declared the car "unairworthy". I didn't have a flashlight, tools, or the time to dink around with it so I ran back inside to get the keys to the E46, as I concluded I'd be driving that to work today. As I pulled the E46 out of its lair for the first time in a week or more, I looked back at my E36 and felt sorry for her. She was built to drive, but she was getting the day off whether she liked it or not.
I drove over to the dealer to talk with my tech. I found him with his head wedged under a mostly dismantled dashboard of a new 7 series. So as to not startle him, I gave him the usual, polite greeting "hey, quit laying down on the job, you slacker!" When I asked what he had gotten himself into he rolled his eyes and admitted "heater core". Hmmm. That looked familiar. Anyway, I told him what happened. When I told him that I had recently topped off the tank, he immediately said it likely had to do with the sealing ring around the suction unit. I related that I had replaced the pump, suction unit and sealing rings over a year ago and never had a problem before. But then I remembered....the attendant really oversqueezed this time...more than they usually do. Could it really be that I didn't tighten the locking ring enough to create a perfect seal? I'll save you the suspense. The answer is yes.
This evening I pulled the bottom seat cushion out of the car and removed the inspection plate over the suction unit to find an even stronger smell of gas. A bunch of dirt and leaves had collected in the center of the suction unit so I had to vacuum that out before I could see that the leak was in fact due to a poor seal between the tank and the unit's sealing gasket. I pulled a large flat-blade screwdriver out of my toolkit and rapidly and lightly tapped the locking ring clockwise. I was surprised that it rotated a solid half turn before snugging up nicely. It certainly wasn't loose like that when I installed it. I suppose it's possible that the "dryness" of the rubber prevented me from achieving the correct torque, but I'm not exactly sure what cause this to loosen up. All I can recommend is to use some fuel lube or other rubber-safe lubricant during assembly so the locking ring won't bind up and give the false appearance of being fully tightened.
I cleaned up the area adjacent to the locking ring and under the car as well and took the car out for a spirited drive to burn off some gas. When I returned home I grabbed my hideously bright white LED flashlight and my telescoping mirror to look for leaks, particularly around the hardlines where they convert to rubber fuel hose. The supply line is always under pressure, so a leak at the fittings would be most evident there, but everything appeared dry now. The hoses appeared to be in relatively good shape, but I've added replacement of those hoses to the plans for later this summer. Mike Miller of BMW CCA has advised more than one owner of a 10 year old BMW to replace the fuel hoses, so I don't think I'm overreacting when I say "it's time".
Sunday, April 29, 2007
Door Pull Trim Replacement
When the car was in the body shop earlier this year I asked the shop to remove the driver's side door pull trim prior to painting the door so it wouldn't look like ass when it was done. Since it takes the same amount of effort to replace it, I asked the shop to buy a new trim and install it.
Several months back I ordered the door pull trim for the passenger side because the original rubber gasket was cracked and warped in several locations and generally looked like crap. The skies were relatively sunny and the temperatures tolerable yesterday so I figured I'd replace the trim.
I'll start by confirming that the E36 coupes indeed have a piece of metal built into the door structure that blocks access to the slider from the top edge of the door (under the rubber trim), so if you want to replace the door pull trim you definitely have to pull the door panel and peel back the vapor barrier to approach it from the inside. The bummer is, even if you do it this way, it's still a pain in the ass because the slider is located on the top of the door pull assembly and is not easily accessible. The rear vertical window slide rail, the door pull itself (the piece you grab and pull up to open the door) and lock assembly all conspire to make this harder than it should be.
The slider is a flat plate of metal about 1/2" tall that runs the length of the door pull assembly. It has a couple of notches cut into it that grab the pins moulded into the back of the trim piece. Various online references indicated that the slider had a hook at the rear (just above the lock assembly), but I couldn't find that. Granted, I was doing this all by feel since it's impossible to see what you're doing in there, but I just couldn't find it. However, I did manage to find a slot about 1/2" long by 1/8" wide at the opposite (front) edge of the slider into which I was able to insert a long (8" shank) flat blade screwdriver in order to gain the leverage required to move the slider. Keep in mind that you must keep a steady pressure on the new trim to keep it flush with the door as you move the slider rearward to lock the new trim in place, or the slider will bind.
Before you mate the trim piece with the door make sure you examine its rubber gasket and ensure that the lip of the gasket reaches around the sides of the plastic as intended. If the lip of the gasket remains behind the trim it will be possible to install the trim but it will not seat properly. If this occurs, the trim will sit high on the door and the door pull will appear recessed relative to the trim. I now see that the body shop didn't take care of this properly and the driver's side trim piece is incorrectly installed. This means that I'll need to pull the driver's door panel off again, but I have plans to replace the lock cylinder in a few months so I'll just fix it then. And if you're thinking "oh, I can just stick a micro screwdriver or maybe a feeler gauge in there and pull the gasket out from under", don't bother. If you do that you'll stretch the gasket, it will not reform to its original shape, and you'll need to buy another trim piece.
By the way, you would be well advised to wear old clothes when you do this and have some GooGone, WD-40, terpentine, or kerosene handy as you will need a solvent to clean up if you get any of the gooey vapor barrier sealant on anything. It's very tenacious stuff. I got a ton of it on my arm that did not come off with ANY of the various soaps (including citris-based cleaners) I tried.
I'm pleased with the end result, in spite of the fact that I had to repair the door panel for the third time (yea, that stupid upper plastic strip broke free of the door panel substrate again...errr). It was well worth doing myself as the dealer would have charged 0.8 in labor (about $90) for the job. As usual, that's better in my pocket than theirs.
Mileage: 143000, Parts $11, Labor Saved: $90.