OB99W

The objective is to reduce resistance. That requires tight, clean connections (free of rust, paint, etc.) having good contact area. 1) A voltmeter can be used to check for voltage drop across a connection. Lower is better; 0.1 volt or less under operating conditions, if possible. 2) If the vehicle has seen enough miles (read "vibration") and/or salt (read "corrosion"), it probably isn't worth checking for bad connections with a meter. Start with any that look obviously bad, or just work on all that can be found, even if they appear to be tight and clean. 3) The smoother/flatter the surfaces, the better. Only the "high" points touch, so rough surfaces minimize contact area, increasing resistance and making it easier for corrosion to get into the connection again. Even if contact-enhancing "glop" (that's a technical term ) is used, a smooth surface to begin with is preferable. 4) To help with 3), a metal file, wire brush, or scraping with a razor blade is preferable to grinding. If grinding is necessary due to severe corrosion, file as flat as possible afterward. 5) Try to avoid "sandpaper" (and grindstones); the abrasive is typically a metal oxide (which is a semiconductor), and some is left behind on the surface being "cleaned". If you must use one of these, a fine-bristled wire-brushing afterward can remove most of the sanding dust before remaking the connection. 6) When cleaning terminals, don't overdo it; most are plated, and the base metal under the plating is usually not as corrosion-resistant.

If the vehicle has ever had bodywork done, that may be primer overspray. In my experience, "Bon Ami Cleanser" on a dampened rag will usually remove things like that with a bit of effort. I've never had Bon Ami scratch glass, but you should probably try it in a small inconspicuous spot first, just to make sure. You might even have luck using just a non-abrasive dish-cleaning pad and some dishwashing liquid.

The pic certainly provides a new perspective on "breaker". Yeah, that bar seems to be allergic to your "torque multiplier".

There's no "magic number" when it comes to resistance in ground circuits, other than "as-low-as-reasonable ohms" . That's because the real concern is voltage drop (you want as little as possible), and that's determined by the product of the resistance of the connection and the current going through it. For example, one amp through 0.5 ohm drops 0.5 volts, but 50 amps through only 0.1 ohm will cause a much more significant 5 volt drop. A voltmeter can be used to check the drop across connections under operating conditions, where that's practical (it isn't always practical). Under 0.1 volt of drop is a good connection no matter what the current, since a loss of that magnitude is fairly insignificant relative to the nominal "12 volt" system. Many circuits can tolerate more loss, but if you can keep it low, it's obviously better.

Sorry, 3rd is 1:1 . However, it won't go into 4th until warmed up.

My '99 OB has a 4EAT, with 4th being a 1:1 drive ratio (no OD). It won't go into that final gear until it's warm, just as Subaru states. --OB99W

The ABS light normally comes on when there is a system fault. Under any circumstance, it should at least come on for a second or two when the engine is first started, as a test. Does it? --OB99W

A weak signal from a wheel sensor might cause this; at slow wheel rotation speeds, the output from the sensor/tone-wheel is not only at a low frequency, but lower in amplitude. If too low, the ABS may think a wheel has locked, and kick in. Naturally, the wheels are going slowest near the end of a stop. Some Qs: 1) Has the ABS light been on (other than a second or two when first starting the engine)? 2) Is the problem recent? Weather conditions in the Northeast have caused unusual icing conditions in some areas. If the OB hasn't been in a warm garage lately, it might just need a defrosting. If you get back with As to the Qs, I might have some troubleshooting ideas.

If you haven't seen these articles already, they might be worth your attention:http://endwrench.com/pdf/feb2004pdf/4EAT.pdf http://endwrench.com/images/pdfs/4EAatPh2Win04.pdf http://endwrench.com/images/pdfs/4EAT.pdf The above indicate that a duty C solenoid failure will result in maximum drive to the rear wheels; by "failure", I assume it's meant that the solenoid can't be actuated (due to an open circuit, etc), or is stuck in the un-powered position. However, that would imply for the solenoid to be responsible for no rear drive, it would either have to be powered at all times, or stuck in the powered position. I've never had to deal with the duty C solenoid and so I don't know if this is practical, but if you can determine whether it is being powered all the time, it might help with the diagnosis.

Thanks for that info; whenever someone mentions 4EAT troubles, I first suggest checking the fluid level, or considering a flush if it's been a lot of miles since the last change. I also remind people what the "E" stands for, and it's good to hear that sometimes the "electronic" fix is simple.

Okay, thanks; I much prefer a discussion based on technical info . It's obvious that at the temps you've provided data for, the 5W-50 is more viscous. But I didn't say otherwise (and in fact that's why I suggested all along sticking with the recommended weight). Even without seeing the data, I'll give you that the 5W-50 is likely somewhat more viscous "further down" as well. However, it shouldn't be so viscous at cold temps that it couldn't legtimately be called "5W", or somebody is cheating. We seem to have gotten somewhat "deep", but I think the practical aspect for most drivers would be if at cold temps an oil sufficiently slowed cranking or decreased pumping such that the engine was difficult to start or was poorly lubricated. With full synthetic, as long as we're talking a reasonable temperature above the pour point, either 5W-50 or 5W-30 shouldn't present a problem in that respect. For conventional oils, it might be a bit more of a concern. Can we compromise on the above, because now I think it really is coming down to semantics? Peace .

Setright, I think your info on how to avoid problems when filling a Subaru cooling system is excellent. I don't mean to pick on you, but we apparently have differing viewpoints on oil viscosity, and it seems to go beyond semantics. Perhaps I can offer something when it comes to oil viscosity, and in particular multi-vis. VI improvers are long-chain polymers that are designed to "coil up" and affect viscosity very little at cold temps. At high temps, they "uncoil" (becoming longer), effectively "thickening" the oil. (Some full synthetics technically don't use what would be classified as VI improvers because their chemistry means they inherently don't thin as conventional oils do.) The "W" (winter) viscosity ratings are currently based on SAE J300. It actually doesn't allow for much range if an oil is to meet the "5W" standard. In fact, an approximately 5% shift in viscosity (with an attendant 5 degC change in temperature), depending on direction, moves a "5W" oil to either "0W" or "10W". In order to make this somewhat less lengthy, please refer to Table 1 from this web page: http://www.texacoxpresslube.com/carcare/article_viscosity.html A couple of points of reference: 1) The unit of viscosity "cP" is "centipoise", also equivalent to "mPa-s" or "millipascal-seconds". 2) To get a feel for cP, water's viscosity at 20 degC is very slightly over one (1), at 1.002 cP. Looking at the table, note that cold cranking viscosity for 5W should be less than 6600 cP at -30C, and that 6200 cP at -35 makes it 0W, while 7000 cP at -25C conforms to 10W. So, the spec is not all that loose at low temps, and the high-temp viscosity (of the "-xx" part of "xW-xx", as it was referred to by Setright) has no bearing on this. You can also see from the table that the high-temp viscosity is, of course, different for 30/40/50 weight oils. Taking into account the high-temp aspect of certain multi-viscosities, they're listed in the right-hand-most column. By the way, "cSt" is "centistokes". So, I'll stick with what I said before; 5W-50 and 5W-30 oils (assuming they aren't mislabled) should have fairly similar 5-weight viscosity characteristics at sufficiently cold temperatures. At high temps, they will take on the character of 50- and 30-weight oils, respectively. As I also said previously, I'd go with the manufacturers recommendations (unless the engine is being run outside the temp range that the manufacturer specs or it has looose-fitting parts due to excessive wear). Sorry for the length; hope it clarified things for some.

Although electrical systems will sometimes work (for a short time, anyway) when submerged, that they apparently didn't shouldn't be surprising. Fortunately I've never had the experience, but it's my understanding that the pressure of water on the outside of a vehicle, if the water's sufficiently deep, can make it very difficult or impossible to open a door, even if it's unlocked. In fact, it may be necessary to wait until the interior fills to about the same level as outside, equalizing the pressure, before opening the door becomes practical. Of course, all of that has to do with physical realities, and not the vehicle brand. The guy probably did panic, but how many of us could remain calm and wait while the vehicle filled to the level at which a door might be openable?

I've seen a lot of disparity on this as well as the subject of the "burn rate" of premium being slower than regular. I'm no fuel expert or chemist, but I'll share what little I've learned. Apparently it depends on how the octane is achieved. If it's done by addition of metallic salts (TEL, MMT, etc.) the ignitability and burn rate are interfered with somewhat. US premium currently uses neither, but instead blends more aromatics (tolene?). As far as I've been able to find out, it doesn't burn any slower or faster, but does resist self-ignition and does vaporize well so can aid cold-starting. Sheesh, I just can't seem to get away with using imprecise/less-technical language such as "resistant to burning" . Okay, what I really meant was "higher-octane gasoline burns with a more-controlled flame front, exhibiting less tendency for compression from the advancing pressure-wave to cause spontaneous ignition of the remaining vapor, as compared to lower-octane gas". Better ? Yes, results with premium vary, depending on formulation, which is why I said "it may actually be a worse performer when the engine is cold" (emphasis added). Although it varies with area and season, on average premium is slightly less volatile than regular. Additional toluene and other aromatics are used in some areas, but in California (where Cookie is apparently located), aromatics are limited in order to meet standards for CBG (Clean Burning Gas). Have a knock-free and high-mpg new year, everyone!

John, since you apparently have a DVM, you could look for bad connections using the "voltage drop" method. That's done by putting a probe of the meter on one side of a connection, and the remaining probe on the other side. For example, to check a connection to a battery terminal, put one probe on the post and the other on the cable clamp at the same terminal (+ or -). Any reading other than zero or very slightly more (less than 0.01V!) indicates a resistive connection that should be cleaned. In order to have sufficient current flowing for this type of test to work, you need to load the battery. That can be done by running headlights, preferably with high beam, and possibly accessories; with many Subarus you'd have to have the key "ON", but engine not started. Another way to check voltage drop at the battery terminals is to load the battery as above, then take voltage reading two ways; the first from positive post to negative post, the second from pos clamp to neg clamp. Compare the two readings; any difference beyond about 0.01V means there's a bad connection at one of the posts.

I also agree with the suggestions to clean terminals and other connections. There's a remote possibility that your battery has a damaged internal cell connector, but a bad connection is more likely, so you might as well start there. Don't be fooled by things appearing to be clean and tight; sometimes corrosion can form between metals where it isn't obvious. Of course, if you can see an obviously bad spot, begin with that. Another thing - if possible, use either a wire brush or file to do the cleaning. Some people use "sandpaper", which most of the time is actually some form of metallic-oxide type abrasive. The problem with that is unless you thoroughly clean off the connection before reassembling it, you can leave behind those oxides; they aren't good conductors, and will actually limit how much metal-to-metal contact you wind up with.

As long as an engine doesn't require fuel above a certain octane rating, that would be the expected outcome. Before engine management systems, you'd either give your car what it wanted, live with the knocking if you didn't (and possibly damage things), or manually retard the ignition timing (read:"turn the distributor a few degrees"). Now we have things like computers and knock sensors, and the system retards the timing for us if we try to "cheat". If a lower-grade gas causes the timing to get retarded, mileage and power are going to take a hit, and going up a grade or two helps. However, if the engine is already happy with the lower grade, timing won't get retarded, and a higher grade can't improve that. In fact, since a higher-octane gas is more resistant to burning (by definition), it may actually be a worse performer when the engine is cold.

Okay guys, I found another PDF covering the 4EAT in some detail. If you can get your head around this, maybe you can tell the dealer what's wrong or fix it yourself: http://endwrench.com/images/pdfs/4EAT.pdf

There are lots of possible causes, but I'd suggest that the first thing to do (if you haven't already) is check the fluid level. Make sure it's done hot, and using the proper procedure - see the owner's manual. Either under- or over-fill is problematical. Next is to consider that the 4EAT, like most auto transmissions, has some close-fitting parts that need to work in a very clean environment. Unfortunately, trans fluids can get gummy ("varnish"), and things begin to stick; if the trans hasn't seen new fluid in lots of miles it may be time for a flush. Sticking shift solenoids and/or duty solenoids, among other things, may loosen up and resolve the problem if nothing was damaged by waiting too long. Of course, since the "E" in 4EAT stands for "Electronic", there could be a problem related to that. But I'd think about the fluid first. I found two links that might be of interest; the first is for the phase-1 4EAT, the second the phase-2: http://endwrench.com/pdf/feb2004pdf/4EAT.pdf http://endwrench.com/images/pdfs/4EAatPh2Win04.pdf There are apparently "fail safe" modes that the trans can go into if it detects a problem, and might cause some strange-feeling operation. Also, a scanner might be able to find stored codes that could give someone (although maybe not your dealer ) a clue. I don't know off-hand what's applicable to phase-1 versus phase-2. My own '99 OBW developed a slow trans fluid leak in something Subaru basically says you might never need to change - the external spin-on filter. Road salt apparently got onto the filter can and ate a hole through it! I got lucky and noticed the problem in my driveway. A trip to the dealer courtesy of my brother, purchase of a new filter at $36 from the dealer, a filter wrench and fluid from an auto parts store, and time under the OB (on my back) and over it (with a long funnel) got me back on the road. After a short trip for warmup, I checked and adjusted the level. It's been okay since (well, the trans has, anyway ).

Not that these answer your questions, but you might find them interesting:http://endwrench.com/pdf/fuel/KnockSensorInfoF01.pdf http://www.geocities.com/hobiegary/hesfix.htm

Plato was pretty handy with double entendre, too.Happy new year to you and the other forum members!

I didn't want to get too deeply into this, which is why I just used the word "slows". However, since you seem somewhat reluctant to believe me, I'll explain a bit more. 186,282 mph is the speed of light in a vacuum. It's a common belief that electrons move at that speed in a wire, but it just isn't so. Even under ideal conditions, electron current moves slower than the speed of light; under certain circumstances (due to resistance, capacitance, inductance), dramatically so. Are you familiar with the term "rise time"? When speaking of voltage that is not constant, it's the time it takes for it to go from a certain low value to a certain higher value. Resistance in the secondary ignition circuit, in addition to slightly reducing the available voltage at the plug, causes the rise time of the spark voltage to decrease. The faster a pulse rises, the more its initial harmonic content, and the more likely it is to trigger parasitic oscillations and further harmonic generation; slowing the rise time therefore reduces EMI. At the same time, the slightly lengthened spark duration can be beneficial for ignition. I hope that helped.

Multigrade oils start off with a base stock, and then VI (viscosity index) improvers are added. An oil with a "5W" rating (the "W" is for "winter") has a 5-weight base, and will be free-flowing at low temperatures, whether the VI improvers make its high temp characteristics equivalent to a 30-weight or a 50-weight oil. So viscosity under cold conditions should not be an issue with 5W-50. What will be affected is a warm/hot engine; 30-weight is certainly going to be thinner than-50 weight. Manufacturers have typically specified 30 for the high-temp weight because nowadays they are concerned about gas mileage as well as lubrication. Using 5W-50 shouldn't affect cold weather starting, etc., but it could cause your mpg to decrease, and it might not as readily lubricate small spaces. In general, sticking with the manufacturer's recommendations makes sense, unless the engine is so worn that a higher-viscosity oil becomes beneficial.

That was probably for the best. As others have said, disconnecting the battery for a while should clear the stored code(s). Unfortunately, that also usually means the computer will have to relearn your car's engine management systems and your driving habits. While its doing that, it runs in a default state. Just don't be surprised/concerned if things seem a little "funny" for a few miles.

I tend to keep tabs on performance/mpg, and (besides the CEL) a change in either would motivate me to consider those things. So, absent bad mpg, misfiring, etc., I wouldn't worry about them at your current mileage. Now if a mouse has nibbled off the plug wire insulation, that's another thing . As to the resistance measurements, they seem in the ballpark. It's likely that anything from about 5-10 kohms wouldn't be problematical. Ignition voltage is high, but the current is not; a small increase in wire resistance won't make much of a difference at the plug. BTW, in addition to providing EMI (interference) suppression, the resistance also slightly "slows" the spark. That actually can be beneficial, giving a bit longer time for ignition; people who install low-resistance ignition wires thinking that performance will be improved may find the opposite, especially if the plugs are also non-resistance type.