OB99W

Are you suggesting that there should be? The NHTSA site itself doesn't list any either, although it comes up with mention of 2 TSBs for the '94 Legacy. For more info, see my post in this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=48997 I've also found that this data isn't always easy to find, due to the Legacy/Outback - Impreza/Outback(Sport) naming confusion. For example, my own '99 Legacy Outback Wagon was subject to a recall for brake master cylinder replacement. If I look for the notice under '99 Outback, I can only find info about rusting front springs, but if I check for the '99 Legacy, there are 3 recalls listed, including the master cylinder one.

Apparently, it's not the first time that Subaru has had problems with Purolator oil filters. Although I've used lots of their filters in the past on other vehicles, I'm reluctant to do so on the Subaru. It seems that SoA is taking an "if it first you don't succeed,..." approach with Purolator. What follows is from NHTSA; check out the notification date: ENGINE AND ENGINE COOLING:ENGINE NHTSA Campaign No:98V127000 Manufacturer's Campaign No: Manufactured From:01/21/1998 - 04/07/1998 Units Affected:20,306 Owner Notification Date:09/15/1998 Description of Defect VEHICLE DESCRIPTION: PASSENGER VEHICLES, ALSO SERVICE PARTS SUPPLIED TO SUBARU DEALERS ARE INVOLVED (LOT NUMBERS F01 J21D2, F01 J30D2, F02 J09D1, F02 J09D2, AND F02 J20D1). THE PUROLATOR OIL FILTER CANS USED ON THESE VEHICLES CAN FRACTURE CAUSING A VAPORIZED OIL SPRAY AND SUBSEQUENT OIL LEAK. Consequence of Defect THIS VAPORIZED OIL SPRAY CAN RESULT IN OIL VAPOR IGNITION BY THE HOT EXHAUST SYSTEM RESULTING IN A POSSIBLE UNDERHOOD FIRE. Corrective Action DEALERS WILL INSPECT AND REPLACE THESE ENGINE OIL FILTERS.

That might actually work, as it seems not much else did: http://www.ultimatesubaru.org/forum/showthread.php?t=15552 http://forums.nasioc.com/forums/archive/index.php/t-434985.html I do have a low-tech idea. I wonder if restricting the flow in the tubing a bit might help. That could be done with a plastic wire tie around the tube, gradually cinching it down until the output is sufficiently lessened. Of course, if you go too far, since they don't release very readily, you'd have to cut the tie off (carefully, avoiding nicking the tubing) and start again. Or, go for higher viscosity fluid; try adding Jello ? --OB99W

Yes, they're different, and perhaps not for the better. See: http://www.ultimatesubaru.org/forum/showthread.php?t=48543 If you have the time/patience, read the thread at the NASIOC site. Under any circumstance, check for oil leaks and level for a while to make sure that you don't have one of the defective filters. --OB99W

It's a good thing I share your apparent thirst for knowledge ... A little background; both zirconium- and titanium-based O2 sensors are actually ceramic semiconductors (zirconium dioxide = zirconia, titanium dioxide = titania). However, the zirconia type is (mostly) a voltage-generative type, while the titania resistance-changing. ( I said "mostly" because all semiconductors change resistance and generate voltage to some degree with external influences like heat, pressure, and yes, even O2 .) The zirconia is the common one, and as you questioned, the titania rarer. The normal operation voltage of the zirconia units is a max of 1 volt or so, and some ohmmeters can damage the semiconductor. The titania units actually have voltage applied to them during normal operation, since they aren't self-generative, and as long as it isn't excessive, they aren't damaged by it. I'll admit that I'm not that familiar with the exact applications, so without info I just play it safe and use voltage rather than resistance tests of O2 sensors during diagnosis. Of course, having the manufacturer's procedures and specs would really help! --OB99W

Since one of your two transmitters seemingly works correctly (I'm assuming not only that a single button push is sufficient, but that you can be 30 feet away and still have it operate), that would seem to eliminate receiver/antenna problems, as well as a low-voltage car battery, as causes. You mentioned replacing the battery in the uncooperative transmitter twice; just on the off chance that you're getting weak batteries, could you try swapping them between the working and problematical transmitters? Obviously, if the good unit is still good and the bad one still bad after the swap, that eliminates battery problems as the cause. Gentle cleaning of the battery contacts with a bit of alcohol on a cotton swab might help; be sure to remove any alcohol residue before reassembling. Although intermittent operation wouldn't typically be cured by this, have you tried reprogramming for your transmitters? The procedure is in the owner's manual. It's a bit tedious, but if you follow it carefully there's a [bad pun] "remote" [/bad pun] chance the problem will be resolved. Otherwise, I can only think that there's some circuit damage in the reluctant transmitter. Has it been dropped, dunked, etc.? --OB99W

Thanks for the links, they reminded me of another problem with an excessively lean mixture that I didn't mention. Not only will NOx go up with a too-lean mix, but HCs can too, due to misfire; that would help explain the "lean mix, but overheating cat" scenario. All of this is why it's often not easy to diagnose A/F problems. I knew about not using an ohmmeter on zirconium-type O2 sensors; I've only used a voltmeter to obtain a rough indication, or a 'scope for the details. However, I've never tried heating an O2 sensor (off the vehicle) with a propane torch as suggested in the linked article, but it seems that should work. --OB99W

Hmmm, the OP mentioned " '03 Legacy ", but this is hardly worth debating, especially since the ABS operation is likely normal. --OB99W

No links offhand, Carl; doing a Google search on terms like "3-way catalytic converter" ought to turn lots up. Your response made me realize that I did a poor job of explaining the lean/NOx aspect of cat operation. Of course, there's more to it. HCs and CO need to be oxidized, but the NOx reaction is one of reduction (okay, all are redox reactions, but I'm not getting into the chem that deeply now ). It may seem strange, but the catalyst for NOx reduction can't work unless it's in a rich enough environment. Therefore, engine controls are designed to cause the mixture to oscillate between the rich and lean states, at a fairly rapid pace; that's where O2 sensors come in. Anything that causes that oscillation to significantly change pace or amplitude makes the emissions go up and the cat unhappy. If the O2 sensor signal gets bad enough, the computer is supposed to default to "limp" mode (somewhat rich condition); that's why pulling the O2 sensor connector helps diagnose its condition. By the way, I didn't mean to suggest that a constant lean condition would cause the cat to overheat. In fact, it shouldn't, since lean running would normally suggest low HCs and CO, and even though the NOx would be up, the catalyst for NOx reduction wouldn't work well in that environment. --OB99

I used to think that was the only cause, too. Back when cats were the "2-way" type (converting only HCs and CO), it was true. Since about '77 or so, cats also need to deal with NOx (nitrogen oxides), and so are now "3-way". Modern cats can get too hot with an excessively lean mixture because that's when NOx is especially prevalent. --OB99W

Not that it's important, but are you sure that model doesn't have a Bosch? My info says the 97-03 Legacys have Bosch 5.3i ABS, the motor of which isn't inaudible. --OB99W

The short trips don't help, but idling (getting 0 mpg) a lot certainly doesn't improve the average . --OB99W

I guess that's a possibility, Steve. Since the transmitters look different from one system to the next, it seems humantag could tell whether the new one is correct just by verifying that it looks like the one that was lost. Mine appears to be the Alpine type, with gray buttons, and the info I gave is for that system. --OB99W

Now I know where humantag got what was quoted ; looks just like the info at cars101. Unfortunately, it seems to leave some details out that the owner's manual includes, which hopefully I clarified above. --OB99W

I assume you paraphrased what your manual says, because mine ('99 Legacy OB) is worded differently from the above. The two main differences are things accented by bold print in my manual. The first is that you have to complete the ten cycles within 15 seconds. A second is that you should stop at the LOCK position and leave the key in the ignition. It also states that the driver's door has to be opened (etc) within 45 seconds of first completing the key cycles. After you press a button on the transmitter, if the programming is successful there'll be two horn beeps. Something else to consider: Even assuming that you succeed in programming for the new transmitter, the code for your old one(s) will still be stored. If you want to prevent the old transmitter(s) from being used, you need to store not one, but four new codes in order to overwrite the old one(s). If you just have one new transmitter, to accomplish that you have to go through the programming procedure four times; with two new transmitters, twice each, etc. --OB99W

My '99 OBW only gets about 20 (19 in winter) mpg with my usual driving, and I don't "get on it" very often. However, let me provide some details. I don't do a lot of highway driving. The region is somewhat hilly. When I do get on the highway for a prolonged time, the mileage has been about 26 mpg. Overall, I think the mileage is a tiny bit lower than I'd expect/like for mixed driving, but not "terrible" based on the EPA 21/26 rating. --OB99W

That's an interesting observation. Assuming we equate "up north" to "colder", there are a lot of factors to consider that could cause fuel usage to go either way... 1) The colder the air, the denser it is; affects A/F ratio, and therefore power and mileage. Also, it provides more resistance to driving through it -- go fast enough, and the result is noticeable. 2) If the driving conditions are bad (snow, whatever), drivers slow down, there's wheelspin, etc. If they slow down from otherwise high speeds, mileage improves. If they're slipping and sliding, mileage goes down. 3) During engine warm-up, which of course takes longer the colder it is, fuel consumption is usually up due to inefficiencies. Lubricants can get very sluggish in the cold, also adding to fuel usage. 4) Gas expands with temperature. You literally get a bit more from a cold storage tank (when the samples are equalized to the same temp). So gas bought cold improves apparent mileage. ...just some thoughts. I'm sure I left things out. --OB99W

Not that the tire pressure discussion hasn't been interesting, but I had a couple of other thoughts concerning gas mileage. The first is that San Diego tends to be on the warm side, even this time of year, so A/C use could be a factor. The other is speed; boxy (not "boxer" ) vehicles aren't exactly aerodynamic, and going significantly above "normal" highway speed will have more of an impact on them than those with "nice" shapes and ground-hugging. I don't know if either of these come into play for lytnin88. --OB99W

Partially true. The more air pressure, the greater the circumference and the less the rolling resistance; rolling resistance is something to reduce if better gas mileage is the goal. However, inflating to maximum pressure (when checked cold) may cause the center of the tread to wear faster than that at the shoulder area. Of course, if you tend to scrub tires during high-speed turns, that could even the wear out, but you'll go through tires fast. Really ? Ever notice that a tire with no air in it looks perfectly round as long as the weight of a vehicle isn't on it? The load certainly does "flatten" the tire at the road-contact area, reducing the effective radius (distance from axle center to road). The greater the load, the more that happens. It would be interesting to get an "official" response from a tire manufacturer, but some applied physics and observation go a long way . Vehicle manufacturers sometimes (it used to be more prevalent years ago) suggest a slightly lower-than-optimum tire pressure because it gives a "softer" ride. For those of us more concerned with handling and other factors, a couple of psi or so more than spec usually does the trick. Overdoing it can actually cause drivability to worsen. I once rented a car which had very obviously underinflated tires. When I pointed it out to the agency, they "fixed" the problem" for me. I began driving, and the car was basically uncontrollable; it wandered all over the road. A check of pressure later (cold) found they had inflated to 44 psi, the max rating for the particular tires . I adjusted the pressure to a bit over spec, and the handling was fine for the rest of my trip. --OB99W

1)yes2)yes, I bought it new 3)not that I measured 4)pretty flat 5)5W-30 6) I think that's a reasonable assumption. Well, your answer to my 3rd Q is a bit ambiguous, but I'll make some assumptions. Since you've had the Forester from new, if it never got better gas mileage, then recently-defective parts isn't too likely. Because you haven't mentioned any drivability issues, I'm assuming that the car wasn't ever in an accident that could have caused some suspension misalignment, and that it isn't dragging tires somewhat sideways. Even if you've just hit some curbs hard, it might be worth having the alignment checked if it hasn't been done recently. On the other hand, perhaps you only began checking mileage when gas prices went up. A car running California gas (usually oxygenated more than most) tends to get poorer mileage than average, so 22 mpg may be normal for your driving style. California brings up another question - was the Forester bought there? O2 sensors, as you might expect , sense O2; if the gas is heavily oxygenated, they could see a lean enough exhaust to tell the computer to enrich the mixture (send more gas). Even if the O2 sensors are fine, under these circumstances (and especially if the computer is calibrated for a less-oxygenated fuel), mileage goes down somewhat. --OB99W

Here in New York State, they go a bit further. To quote the DMV: "For NYVIP inspection, the NYS-licensed inspection station uses new equipment to monitor the on-board diagnostic system (OBDII) of your vehicle. The station uses the new equipment, a modem, and a dedicated telephone line to report the inspection results to the DMV. The dedicated telephone line must be an analog telephone line; it cannot be a digital telephone line." I suppose that's just in case the problem is intermittent, or someone "accidentally" rigged the CEL. We probably have a bit more pollution than you do in Colorado, or maybe some of yours just blows over here; like Denver, perhaps there's an advantage to being "a mile high" . --OB99W

Yeah, but the instructions always make it sound easy. Read Setright's info, and believe it. After 8-9 years, those block drains may not want to come out very willingly. A trick that sometimes (and I do mean sometimes) helps is to have the block warm, but cool the plug with a shot of aerosol penetrant or the like. The idea is to shrink the plug slightly while the block is expanded from the heat, hopefully breaking any bond due to corrosion. Be careful not to hit the block with the spray, especially if you've already drained the coolant, because it will be less tolerant of a rapid temperature change. Maybe a block heater isn't that important . Of course, as has been noted here often, getting all the air out of the cooling system is critical. I've lost count of how many times I've heard or read "Everything was fine until the coolant was changed, but now its overheating.", or worse. --OB99W

First, make sure the gas tank cap and neck aren't damaged, and that you're tightening the cap fully (several "clicks"). Something as simple as an air leak there can cause the CEL to come on. If the cap is only loose, it will still probably take several ignition on/off cycles for it to reset unless a scanner is used. Even a partially plugged air filter or a slight vacuum leak can throw things off enough. I've recently heard that some MAF sensors from around MY99 may be bad, and that there is a recall for them. That could certainly cause a code to be set. Perhaps you should get your VIN, and call either the dealer or Subaru's toll-free national number and ask if yours is in that group. Since you didn't give any details (mileage, recent service, etc), it's hard to make further suggestions. Besides the CEL, anything else noticable? --OB99W

Don't feel too bad; I got the OB because of the snow, but here that leads to lots of salt, and the salt leads to lots of rust, and the rust leads to... ...never mind. Happy Holidays, anyway!:cool: --OB99W

Some people might call 22 mpg "somewhat below average" or even "average", not "lousy", and as you probably know, there are lots of things that can cause fuel mileage to suffer. But first... Those are interesting inflation pressures; I believe spec is 29f/28r. While the fronts are at spec, you might try another couple of pounds there, which shouldn't stiffen the ride too much or cause wear problems. Unless you're hauling very heavy loads (in which case, that might explain where the fuel is going), 36 psi at the rears is high. Lightly loaded, they should be a pound or so lower than the fronts; carrying more weight, equal to the fronts or very slightly higher than them. Pressure is, of course, measured with tires cold. Uneven tire wear or inflation pressures can cause drivetrain problems. With a bit more info, maybe we can offer other suggestions. Some questions: 1) Are you sure your tire gauge is accurate? 2) Did you buy the Forester new? 3) Did it ever get better mileage? 4) Is the terrain flat or hilly where you mostly drive? 5) What viscosity oil are you using? 6) Can we assume there are no driveability problems, unusual sounds, or other possible clues? --OB99W