OB99W

Are you saying that the CEL came on before the work was done, or after? If after, make sure that an OEM (Subaru) thermostat was used -- aftermarket ones often don't perform well.

That type of noise can be due to brake problems. In particular, check the condition of the inner pads and the side of the rotors they ride on. Make sure the pad ears are free in the calipers, and that in general the calipers aren't binding, etc.

Oxygen sensors don't function until they reach operating temperature of at least 600 deg F. As Gary explained, "newer" sensors have a circuit for an electric heater, besides the one for the O2 sensor itself. When the sensor is below operating temperature the engine management has to run "open loop", a time of high emissions. Although the exhaust alone can eventually get the sensor hot enough, the electric heater speeds that up, allowing the ECU to go "closed loop" sooner and therefore lessening emissions.

Yes, but Bigbusa was talking about New York City (stop-and-go) mileage. While 17.5 mpg could probably be improved on a bit, that's "Big Apple" mileage versus your highway "orange" 31.5 mpg.

Welcome to the forum. In addition to the items that Gary mentioned, I'd suggest that the brakes be given attention. You or a trusted mechanic should verify that the linings are still sufficiently thick, rotors are in good condition, the calipers aren't binding and causing the brakes to drag, etc. You might consider changing (flushing) the brake fluid. I'm assuming that things obvious with visual inspection (tire wear, wiper blades/refills, etc.) will be seen to.

The NGK plug numbering system can be found at: http://www.ngk.com/sparkplug411_ngk.asp The "-11" indicates a 1.1mm (.044") factory gap, but others can be had (although perhaps only by special order).

The two main factors that wheels/tires influence relevant to fuel mileage are inertia and rolling resistance. All other things being equal (assumes outer circumference of 15" and 16" tires would be the same, the 15" and 16" wheels have the same mass, etc.), the 16" wheels/tires could provide slightly higher gas mileage. That would happen because the 16" combination would have shorter (and therefore stiffer) tire sidewalls, resulting in less flex and rolling resistance. However, "all other things" often aren't equal. For one, the 16" ones may be more massive (have more inertia). While in theory the energy it takes to get wheels turning could be largely reclaimed by allowing the car to coast to a stop (rather than braking), that's not very practical in most situations. Therefore, to what degree the driver coasts when slowing, rather than having the throttle open and then braking at the last second, determines how much of that energy is recovered. Of course, that's true no matter what the wheel/tire mass, but the more massive, the more it matters. Stop-and-go driving tends to reduce or eliminate any advantage of the 16-inchers, while highway miles might show a couple percent improvement. Another consideration is that some tires (especially certain "high performance" ones) have higher rolling resistance than others. That could negate any advantage of the switch, and result in an actual decrease in gas mileage. It's hard to predict exactly what would happen for the reasons I gave above, and probably some others I didn't mention.

The Dexron-III is likely old stock, since GM stopped licensing it a while back; it also might list Mercon on the label, since the frictional and other characteristics allowed for that. What you got is probably "compatible", but as you mentioned who knows what was in there before. I didn't mention this previously, because I don't see a connection to an ATF change, but seemingly unrelated things can cause strange shifting in electronically-controlled transmissions. Somewhat-dragging brakes can confuse the ECU/TCU, since throttle angle that's detected would be higher than expected for the road speed; the electronics "thinks" you're on a steeper hill than you actually are and upshifting can be suppressed. If you have any reason to suspect the brakes aren't releasing fully, it might be worth looking into that.

Yes, the wasted spark configuration puts the paired cylinders' secondary ignition in series. However, only a few kilovolts are "wasted" on the plug that's not on the power stroke (it's on exhaust), and a little more would be if its gap were overly wide. So it's possible that #4 required just a bit more "juice" than #3 to fire properly, and sometimes wasn't getting quite enough, while #3 still managed with the lowered spark energy available. This happens more often than might be imagined. Certainly, if the gap or resistance became sufficiently excessive, both cylinders could be affected. However, sometimes one cylinder will need more spark than the other, and when it misfires the problem is noticed and gets corrected before it becomes bad enough for both to misfire. Another quirk of the wasted-spark arrangement is that the "companion" plugs receive spark of opposite polarity, so sometimes the one that's negative at the center electrode fires more readily.

Assuming a cause/effect relationship, perhaps the fresh fluid broke free some "crud" that's now sitting somewhere detrimental. Or maybe the trans isn't happy with characteristics of the new fluid (true Dexron-II or -III are no longer available). What fluid was used? OTOH, the problem arising and change of fluid might just be coincidence; possibly something else got disturbed while the work was being done. TPS, VSS, etc., difficulties could cause the symptom described. A recent somewhat-related thread: http://www.ultimatesubaru.org/forum/showthread.php?t=85096

I think you're referring to the Volvo B18 (1.8L) and B20 (2.0L) engines. They were both inline fours, and noted for their ability to keep going under hard usage (five main bearings helps). IIRC, their bores and strokes were closer to each other than the engines previously mentioned, although over-square by a small margin. BTW, most modern engines are over-square (larger bore than stroke). It tends to make for better fuel dispersion and higher-revving capability, at the cost of lowered torque.

Yes, and they were an inline slant-4; Pontiac cast the block with a wall where one of the cylinder banks was on the eight. Not really -- the 389 (and therefore the 194.5) had a bore of 4.06" and a stroke of 3.75", making it an "over-square" design. (Yeah, I used to own a '63 Tempest with one of the "195" engines.) For comparison (and to keep this Subaru-related ), a stock EJ25 has a bore of 3.92" and a stroke of 3.11" (also over-square), for 150 cubic inches.

You're welcome, I'm glad things got sorted out. You should see some increase in gas mileage as well.

Yes, assuming you're just trying to determine why the compression is low in that cylinder. Using the timing scale at the front of the engine, rotate the crank in the direction of normal running (clockwise, viewed from the front) until the mark is at 0 degrees. If that doesn't result in cylinder #1 being at TDC, then rotate one more full turn (360 deg). Since the firing order is 1-3-2-4, turning the crank in the direction of normal running an additional half turn (180 deg) will put cylinder #3 at TDC. If you intend to apply significant air pressure, the engine should be locked in that position during the test. For the testing you need to do, the "kit" doesn't have to be very sophisticated. In fact, only a means of applying air at low pressure (about 20 psi) to the plug hole is all that's required; keeping the pressure low lessens the need to lock the engine to prevent air pressure from rotating it. Gauges would give readings that permit comparisons, but all you have to do is determine where the air is leaking. Listen at the exhaust, crankcase (remove the oil filler cap), and intake to determine where the air is going. (If bubbles are formed in the coolant, that indicates a bad HG, but you'd have other symptoms if that was the case.) As long as the air is escaping only to either exhaust or intake, the work needed will be on the head, not the block.

"This posting has been deleted by its author."

That's an "industry standard" number; it should "translate" as follows: 275 = length in tenths of an inch = 27.5 inches K = cross-section, typical for automotive application (others are H, J, L, M) 4 = number of ribs For a good laugh, read the "Key Features" of the Prestone version at http://inpcars.com/serpentine1/242.html . I assume that's a botched translation from (perhaps) Chinese.

I can't imagine why you'd bother for "only" $1,000US/ounce.

A "slightly" modified EJ22? Obviously, the more power you require, the tighter the drive belt will have to be to prevent slippage, all other things being equal. Is that why the concern over bearings? Getting a fairly large degree of wrap will usually minimize slippage without the need for excessive tension. Around 120-150 degrees would likely be a good compromise -- 90 deg is often too little, and much over 150 deg can be hard to attain (and if the radius is small could unacceptably accelerate belt wear). Depending on the layout it might require idlers to accomplish the desired wrap.

Yes, above 160 psi would usually be expected for a warm engine with good ring sealing. However, unburned gas may have washed oil from the cylinder walls, somewhat lowering the reading. Also, was the throttle held open while cranking? Still, the lower than expected reading might be from mistiming... ...so rechecking would probably be a good idea at this point.

I assume that's a reference to Hilda, Rumpole's wife (of "Rumpole of the Bailey"), and not Ayesha, Queen of Death. To stay on topic -- yes, the "Low Fuel" light tends to be more trustworthy than the gauge, although not always.

I'd still suggest running a leak-down test, to verify it's valve-related. Valve adjustment might be called for, but a burned valve could explain the compression loss and adjustment isn't going to resolve that.

In my experience, the difference between 65 lbs and 80 lbs dry-vs-wet compression isn't enough to be certain of the cause, especially when run on a cold engine; there are things that can cause that much variation from test to test other than the added oil. A leak-down test on #3 with listening to determine where the air is going would be much more definitive. While head work isn't cheap, it would be considerably less than dealing with the block. A good used engine might be the answer if serious work is otherwise needed.

I'm fairly sure (although not positive) that the memory will clear if you just crank the engine with both (read memory and test mode) connectors coupled. A pulse that wide should be enough to start without problem, if properly timed -- it would normally narrow once the engine starts and is idling, especially when warm. Are you certain the fuel rails are pressurized? If so, then noid lights could be used to verify whether the ECU is switching the injectors.

What is the scan showing for injector pulse width? Have you verified that the reluctors on the crank sprocket are all okay, and that the crank position sensor tip isn't damaged? Otherwise, the ECU could be mis-interpreting the pulses. Info on pre-OBDII systems, including how to clear codes:http://endwrench.com/images/pdfs/EWPreOBDAug05.pdf Once they're properly cleared, see if any return.

Just in case... http://endwrench.com/images/pdfs/2.2SingleOverWin01.pdf http://endwrench.com/images/pdfs/2.2Liter.pdf http://endwrench.com/images/pdfs/TBeltEWWin05.pdf Motor Magazine series; starts with 2.2, then moves on to 2.5: http://www.motor.com/magazine/pdfs/072001_08.pdf http://www.motor.com/magazine/pdfs/082001_08.pdf http://www.motor.com/magazine/pdfs/092001_08.pdf http://www.motor.com/magazine/pdfs/102001_08.pdf http://www.motor.com/magazine/pdfs/112001_08.pdf