griffenrider

According to this site: http://www.subaru-tpms.com/tpms-subaru-service-guide.html it appears that the TPMS is installed in a manner that should allow it to be moved from wheel to wheel, but will require unmount and then remount/rebalance of the wheels each time. According to the video they have posted, it looks like foresters also know the wheel location, so if you're going to actually use the TPMS to warn you about low pressure (instead of, you know, checking your wheels from time to time), then you have to relearn position every rotation. I wonder how much the stealership charges for that service? Anyway, short answer, IMHO you should be able to switch the TPMS over from wheel to wheel, but you may need to buy new valve stems to do it. If you plan on using both rims on the car (e.g. winter and summer tires), you'll end up paying the remount + rebalance fee twice a rear (80/year), which adds up pretty quickly.

Hey Hankosolder - thanks for the compliment, I'm definitely no expert on suspension systems, just a general mechE with a totally untested theory. To answer your thoughts: I don't believe that the problem is related to over (or under)steer. The problem, as far as I understand it, happens with the steering wheel straight, and is rather an issue with the stability of the rear end on surfaces of varying traction characteristics. If you do stiffen the rear end, stiffening the front is certainly a good idea, as is realigning the whole deal, I agree with that for sure. As far as the frequency issue, increasing speed would be equivalent to putting more energy into the system, which would make the oscillations worse, not better. The system behaves more like plucking a guitar string than swinging a pendulum - if you just touch the string, it vibrates, but you can't hear anything (less than 40mph), if you pluck in a little harder (40 mph), you can hear the sound, and the faster you drive (harder you pluck), the worse it gets until something catastrophic happens. I'd expect the problem to get worse at higher speeds as a result. I don't know nearly enough (or really much at all) about the geometry of the toe-angle adjustment on the rear suspension of '05 and later subies, If anyone could put an angle strain gage on the bearing hat and reference off the frame, it would be interesting to see what kind of data we get back.

Folks experiencing this problem might consider adding a much stiffer sway bar or stiffer springs to the rear suspension before spending several 10s of thousands on a different vehicle. Here's my theory: This swaying is not a traction issue, but a resonant-frequency issue in the suspension. Below about 40mph, the shock absorbers in the system can suck up enough energy to keep the oscillations below a noticeable threshold. If that's true, the problem should get exponentially worse the faster you go, which it sounds like it does. Varying traction on the back wheels will cause slight changes in the direction of force at the wheels compared to the center of mass of the car, which is effectively the same as shifting the center of mass of the car from side to side. If the surface is varying traction and the wheels aren't completely 100% solidly attached to the ground, like when you pass over a patch of black ice with one wheel, then the friction coefficient you have to worry about with the ground is no longer static friction, it's the kinetic friction coefficient. That force you get from that wheel, therefore, depends on the instantaneous load of the wheel, which, with a shifting CM, will vary. In the interests of explanation to folks who don't spend their lives on this forum, car suspension is designed to do a few basic things: 1) absorb bumps by moving up and down, 2) remain stiff (not move) during turns and other "slow forcing" events (like starting from a stop, etc.), and 3) support the mass of the car on the wheels. The forcing I just described would fall somewhere between category 1 and 2, it's wouldn't be nearly as fast as hitting a bump, but would be faster than the jerk (time rate-of-change of acceleration) of going around a turn or starting from a stop (I'm ignoring race-style driving. . .). The length and stiffness of suspension springs are adjusted so that the car at "expected normal load" is 1/3 or 1/2 compressed. The lockout pressure of the shock is adjusted to differentiate between cases (1) and (2), with the tradeoff that higher lockout pressures (stiffer in turns, etc.), also makes the shock less able to respond as quickly to bumps in the road. Little adjustments here make big differences in road feel. With the 05-09 model, Subaru increased the mass of the car, raised the car, and changed the wheelbase of the car, all of which change the natural frequencies in the car. They also, in an effort to capture some of the SUV/crossover market, softened the shocks to attract the hummer-in-suburbia-terrified-of-potholes market, who don't go around corners fast pretty much ever, so wouldn't notice a little body roll in day-to-day driving. However, like any damped-spring system, there are always frequencies that aren't damped out well. In case anyone's thinking that this couldn't possibly happen, recall your childhood when you and your 3 friends sitting in the back of mom's car could get it sway back and forth a lot by jumping side to side at red lights. Also see: washboarding on dirt roads, and that annoying buzz in basically every car with a V engine at certain RPMs. Obviously, the suspension is a lot more complicated than a simple spring & piston assembly, but the general principles still apply, including the resonant frequency one. So, with that lockout ratio adjustment, the increased mass, and the shifting CM issue described earlier, the varying load on the suspension when the car is just driving down the lane under certain very specific conditions hits a resonant frequency with the suspension system and you start getting sideways oscillations. Changing the wheel toe angle a little will cause the wheels to pull/push sideways slightly more (or less, depending on which direction you toe it), The amount of pull is related to the speed you're driving, and would function to increases the static lateral load on the shocks at higher speeds. Unfortunately, it also increases the rate of tire wear, so they can't make too much of a change. With a higher lateral load, the shifting CM/force vector issue is diminished (but not eliminated). The adjustment should allow you to put a bit more weight in the back, or drive a bit faster, but the problem would still surface at some point. I imagine the subaru engineers adjusted it as much as they could without causing tire wear issues, but the problem still exists at a lower speed than Joe Average with the average load feels comfortable driving in mixed conditions. Now, the big assumption: IF IF IF IF IF IF the above is correct (that's a judgement you'll have to make for yourself), the correction to the problem should be to modify the suspension characteristics of the rear shock. Adding a stiffer sway bar should do it, so should any number of aftermarket shocks, or adding a strut bar. If you're super mechanically inclined and have a welding shop, moving one or more knuckle joints should also do it. A slightly stiffer spring should have the same effect as the toe-angle increase (and combined with the toe angle increase might push the problem well above normal driving speeds). BE WARNED: Keep in mind that the suspension system is a highly tuned subcomponent, and the subaru engineers are some of the best engineers in the automobile industry. It sounds like they made a mistake and didn't run the particular (very specific) model that would have caught this. Modifying your vehicle suspension is going to fundamentally alter how your car behaves on the road. It's going to roll in corners slightly more or less, it's going to bounce a little more or less over potholes and bumps, and it's going to respond with slightly differently to steering corrections. You probably wouldn't notice it consciously unless you're paying attention specifically, but do us all a favor and drive your car as if you aren't familiar with it (like you would a rental) for a bit after you make this fix.

interesting point fairtax, I haven't owned (or looked into buying) anything besides a subaru pretty much ever, so the comparison with other brands is interesting news. Do you have some numbers from subarus you could compare to that formula?

ln is the natural log function. it outputs a value y such that e (the natural number) to the power y equals x.

I've been really curious about how depreciation rates change with time, so I did a little data-gathering experiment: 10 years, my (and the lady's) past 3 cars (aka very small sample size). It looks like the value at of a car at year X is given pretty accurately by (price of car + scrap price) - [ 0.34 * (price of car + scrap price) * ln(x) ]. And scrap price seems to be approximately 1/10 * the mass of the car. Does anyone else find this interesting? Have numbers to compare?

I wouldn't think you'd need to replace the entire wiring harness, you should be able to splice the wire together. Knock sensor signals should be pretty clean though, so make sure you get a nice tight splice and solder it cleanly, don't just wire-wrap and electrical tape. Not sure how far up the wire the break occurred, but if it isn't enough to do a good splice: once you get male end off your old sensor, you should be able to open up the plug and solder a new wire to the contact directly. You'll get as much slack as you need to make the repair. Make sure you disconnect the battery AND remove fuses before playing with naked wires though!

thanks guys! It sounds like the overwhelming majority (i.e. all of you) are voting repair, which is good news because a) I like the car, a lot. and nobody seems to have had the cat go into another 8k worth of repairs immediately afterwards. As far as the roof, sealant options are silicone, pvc tape, and electrical tape. I'd like to use something removable so that I can actually repair this thing come summer - are any of those permanent? I'd also like it not to crack this winter (expecting temperatures down to at least -20F), experience with that? @mdjdc: check your pm, and THANKS! And as far as the cat - 800-900 is for the part only. The low end gets me an OEM 49-state cat (1-year, 10k warranty), the high end gets me 50-state cat with 5Y, 50k warranty. I'm thinking the extra warranty (and the extra clean) is worth the extra cash. I'm going to let the shop put it in because by the time I get back from the ER with amputated frost-bit fingers, it'll be worth the trouble. specific replies to folks below: 1) @WRX2FFU: very good point, especially since people will likely quote the dealer-repair on the cat, which is $1300 parts and labor. 2) @grossgary: it's a P0420 (cat efficiency), not an 02. I reflashed the ECU with the recall a while back - WWF89, I think? - it cost me $30 at Van Bortel Subaru (only dealer I've been to in 4 states that I wouldn't call a stealership) and haven't had an 02 error since. I did play C.E.K.D. with the 02 error for a little over a year before I did the reflash though.

900 is the high-end, I've been quoted 800 - 900 for the cat, the issue is that I would rather pay someone with a shop and a lift rather than work under (or on top of) my car while it's on ramps and 13 degrees outside. Were it summer, I'd do it myself, but my timescale just isn't going to permit that. As far as the sunroof, I'm sure it's not the drains for two reasons: 1) I've cleaned them very, very carefully, and 2) because I can see the crack in the runners. The leaking is the reason I need to fix it, but the original problem is that it's, well, broken. In either case, it sounds like you're voting for repair because the car still has some major life left?

Oh no! an opinion question. Here's to hoping it gets interesting. My (2002 outback ltd) car needs a new cat to pass emissions in NY. I'm getting parts estimates of around 900. The moonroof chassis also should be replaces (so that my roof stops leaking - and no, it's not just the drain tubes), which is going to cost me 2200 in parts. It's winter and I don't have an indoor garage, so labor for the two is going to be around 400, assuming I drop the headliner and put it back in myself. The car is worth (according to an average of KBB and Edmunds) 5100. It has 130k and is in otherwise damn good condition. So, do I repair or replace? I'm arguing myself in circles, so please (or the sake of me learning something) justify your vote. Thanks in advance!

Is there a way to tell when the oil pan crush washer is worn out (and before all my nice fresh oil ends up on the driveway :-O ). I know I *should* go to the parts store and get a fresh washer every time, but I usually realize I need to change oil, grab a new filter from the stack I bought when a local Advance went out of business, and pull quarts from the bulk oil in my shop (I use Castrol GTX 5W-30, nobody panic!). I also know that the washer is like $0.80, or somewhere around there - but I also know that if "get crush washer" goes on the to-do list, "change oil" will take 2-3 extra weeks to come off the same list. Luke

Unless you're getting it repainted in the next few days, IMHO, you should put something over the scratch to prevent car-cancer. Once rust starts under the paint, there's really no way to get rid of it without either sanding everything off and starting over, or replacing the panel. put a line of pastewax in there like its caulk - it'll keep the water out and is easy to remove when you go for the actual repair. HF is a good chemical, but I'm not sure if it works fast enough for you - whats the glory in knowing the guy'll be gone in a few months when you don't get to see it happen? If you know who did it, let us know and I'm sure we can come up with some fantastic (and totally justified) retributions.

thanks for advice and links: Perhaps its just the 110k parts-wear-out mark, but it seems like I've been putting a lot of money into it recently - hesitation (fixed by 02 sensor) at the beginning of the summer, clutch, cat, a bunch of front-end work after getting run off the road by some fool doing 80 on a mountain road, 2 new wheels and tires after a sudden blowout on a gravel road that sent me into the ditch, and just recently a really bad hesitation and stall-on-idle issue that took 300 at my local shop to track down and (possibly) resolve - not sure if its fixed yet since I just got the car back. On the one hand, yeah, I just put all the money in and I'd like to drive it for a while before I let it go, on the other hand, all this has happened in the last year, and I'd rather put the money towards a less-problematic car than keep sinking it in here. All comments welcome. . .

Hey all - I'm considering trading in one of my subies for a newer one. I don't want to pay the brand-new-car premium. The car I'm trading in is an '02 obw ltd. I'd like to stick with the obw line (I need the space for skis and such, and I need the weight for snow) Which model years are the best 03-08? I know they changed some stuff around in the last 6 years - anyone have a history of that? Any mys to avoid?

Engine grounds are good, I put my multimeter on the most sensitive setting, stuck one lead against the engine block and the other against battery neg, no movement of the needle (with engine running)

Alright - the update. Took a long time to go through a tank, which is a good thing with today's prices. 4.29 at the cheapest station. Anyway: still happens at all fuel levels, but more frequently at lower levels, but not substantially so. - not (at least if my understanding of Nipper's posts are correct) the knock sensor - still no codes - not (if my gradual-throttle test of the TPS is correct) the TPS - Still happens with a new tank of gas, at full - so not bad gas (at least almost certainly, I use on-brand only) - cracking gas cap didn't appear to help, though the problem is so intermittent its difficult to tell. I don't want to drive with the cap off to find out! Other thoughts? Thanks, Luke

Vac lines have good suction and no leaks. I reved in neutral at various rates of increase and could not reproduce the hesitation - I take it to mean TPS is ok? Also disconnected the battery again and went for a 30-mile drive on and off the highway, problem still exists. Keep the suggestions coming! This thing's got me stumped.

thanks to forward-thinking at subaru, my tank came standard with a drain plug. Stuff like this is why subies rule. The problem *does* exist even at full, its just much much less noticable will do on the TPS, it only happens at low throttle, so this could be the answer. I don't believe I have a MAF (or a MAP) on the 02 obw?

I have the thule Ridgeline-4 bike rack (trailer hitch mount) and the Cascade XT (17 cf, I think) roof box. Running both at the same time I get about 26mpg on the highway (28 normally); doesn't seem to affect city driving at all, I get 22 either way. Running just the bike rack doesn't make a bit of difference, even on 1000 mile road trips.

I keep a 300amp inverter in the car instead, that way you can charge your phone on the regular charger, or charge your laptop, or power your girlfriend's hairdryer, or run a chop saw (yes, I've done that, though you have to tie it directly to the battery to do it). The 300 or 600W model by Husky (from Home Depot) works well and fits nicely in one of the side pockets in the trunk. I still maintain that anything that can't be fixed with a leatherman, duct tape, a paperclip, and or spit probably can't be fixed without a spare part. Get a real one, not an imitation. (swiss army and gerber both make imitations, I was given both as gifts, broke both in less than a year. Had the leatherman for 10 now, still going strong) EXTREMELY good point. Small and can potentially save your life. +1, if you aren't disciplined enough to check once a month (like me) carry a bike pump instead. true, use the spare if at all possible. That said, I've gotten 2 flats in quick (less than 30 seconds) succession on gravel roads 50 miles from the closest town. Spare on the worst flat, fix-a-flat (plus bike pump) on the other. If you're going out of the usual "help" area, it doesn't take up much room, and can save a very long hike. Don't use it unless you absolutely have to though -stopping every half hour to refill a slow leak is better than using the fix-a-flat. And one last thing I forgot on my earlier list: Haynes or FSM for your model year. Invaluable when trying to diagnose a problem you aren't familiar with.

That makes sense, thanks Nipper. I'll pull it off when I take lunch at work today (yeah, I work Sundays :-\ ) Though I'm still not entirely convinced it's bad - mostly because the hesitation only shows up well below half-a-tank, and only occurs under light acceleration, when even a loose knock sensor shouldn't be getting knocked around much. EDIT: Pulled it, wiped it, ran a piece of cotton around the threads, put it back in. No new CEL (which I take to mean I didn't break it putting it back in, always good news to an amateur like me). No cracks, corrosion, etc. Does this mean its good? I'll let you know in a couple weeks how it all works out.

sounds fairly obvious, I'll check when the tank gets down below 3/8ths (its full at the moment) As for the knock sensor, no, I'm not certain, but I read somewhere on USMB that if there's not a CEL, its almost certainly not the knock sensor. Is this incorrect?

I don't know much about gasoline quality, what am I looking for?

Always a tough one - what are you going to be doing? where do you live? how far from help do you expect to be on a regular basis? FYI, there is a large tray and a lot of storage under the floor in the trunk, you're entire kit can probably fit in there if you pack carefully (mine does) I live in upstate NY, transplant from VT, where I lived in the mountains on a gravel road that was bottom of the plow-list, with that in mind: long answer: -flares: good idea, I have 3 -I don't have a caution triangle, but I do have a roll of flourescent tape -flashlight or headlamp is a very good idea, keep the batteries separate and in an airtight package (original box or ziplock+rubber band works well); don't forget a spare bulb if the lamp isn't an LED (which is a good idea if you think you might be stuck for a while or have to walk a day to get out) -cell phone and credit card: keep these with you, not in the trunk :-) -I have a small bike pump and a pressure guage, much smaller, much lighter, works just as well (even if it takes some elbow grease). Bike pumps are usually rated to 120psi, so they can definitely fill any high-pressure doughnut thrown at them. That said, if you're worried about being far away, ditch the doughnut and get a full-size spare, especially for the winter, doughnuts are crap on snow, and they can't go far. Rotate your "spare" with the others for even wear, and rotate often (every 10 tanks of gas is my rule) -work gloves = good -I also have a set of warm gloves, a wool hat, and a wool blanket for winter emergencies (and also as padding in the packing process!) - first aid kit is good, make sure its on top since if you need it you'll need it quickly. - tool kit: duct tape; leatherman, paperclip, spit (not to be stored in the kit)!beyond that, a torque wrench and a set of sockets, a fuse puller w/spare fuses, a pair of pliers, a cat's claw, D clamps in the 3-4" range, spare belts, a non-folding knife, a set of screwdrivers, and a rope/tow strap make up my kit - you only really need a portable battery pack (which are heavy) if you got an automatic, you can always push-start a manual - I don't know why you'd need diesel filters (anyone?) - I carry a compass, not a GPS (magnets don't run out of batteries), map book is a good idea - when you get work done to your car and they replace a belt or a pump before it fails, ask to keep the spare parts and use those to fill out your inventory, I only carry belts and fuses on a regular basis - I carry a bottle of 5W30, but go warmer if you don't live up north - coolant: whatevers in your car currently - I carry a water filter, but not water itself - quarters, etc. optional: I just put all my change in the pocket in the driver's seat armrest, never had an issue - I keep 40 bucks stashed in the glove compartment for "emergencies," though any problem that can be solved by money alone isn't really a problem - wire and wire ends: I have a small roll of tie-wire (14 ga), comes in super useful since it won't melt - chains: I keep 20 feet of loose chain in addition to tire chains, a lot easier to play with, and sometimes all you need is a little traction over the course of a few feet, and if you're stuck, you can't put chains on without a jack anyway. - since I'm out in the woods, I also carry a small axe and sharpening stone, a small box of nails, 3 MREs, a signal mirror, a coil of 200lb test rope, 10 sq. ft. of plastic, a shovel, and a box of candles. one last comment: if you're not going out into the country much, you can skip most of this stuff, it adds weight and reduces gas mileage, and if you're within a cell phone call of help, it makes 90% of the above redundant. I'm pretty frequently in areas where there is no cell signal and no-one else will be for at least a day or two, so I'm pretty careful.

Time to put a question out there - I've read just about every thread with the words "hesitation" in it, and here's the conundrum: 02 OBW ltd, 110k, manual did the 105k sevice (timing belt, water pump, belts) new plugs/wires new air filter and fuel filter fixed dead front oxy sensor w/ new P0420 cat eff. error, turned out my doughnut was rusted through and leaking exhaust, and the cat was fried from driving with a bad 02 sensor for so long (expensive whoops), so: new cat, seals, etc. new clutch kit, as it was starting to slip in 5th on the highway no errors pending or in memory (so its not a knock sensor issue?) good gas mileage: 22 mpg city/28highway (better than "sticker") BUT: when accelerating gently there is a significant hesitation between 2k and 2500 rpm - almost identical to the hesitation with a bad 02 sensor, but not quite as bad. Happens mostly in 2nd and 3rd gears, but only because 90% of my city driving is in those gears. Happens more when I have 3/8 of a tank of gas or less. I'm currently thinking water in the tank? I've heard some horrer stories about what "dry gas" (ethanol) does to your gas mileage and the deposits that result. Any other way to get water out short of dropping the tank and pouring it out? Any other theories?