OB99W

That's not a good idea. At a minimum, if OBD is working correctly, you can get a CEL indicating an EVAP fault (since with the engine running, the tank shouldn't be at atmospheric pressure). Also, given that apparently the tank venting isn't working correctly, who knows what path the gas might take with the engine running. I won't get into possible safety or legality issues, which would probably just start a debate. Unfortunately, if the vent valve is indeed the problem, it will require lowering the tank to deal with that. Before 'going there', if you have the time and patience, it would seem verifying that all the EVAP and gas tank-related tubing is properly connected/routed and free of obstructions might be prudent.

Perhaps ferret will respond here, but if not, take a look at this: http://www.subaruforester.org/vbulletin/f88/car-not-filling-gas-pump-4002/

Below are links to info that might answer some of your questions. One includes an example of the drive cycle requirements to trigger the CEL. Another details usage of a piece of equipment dedicated to EVAP system diagnosis (which might make plain why they can be a PITA ). http://endwrench.com/images/pdfs/VehicleWont.pdf http://endwrench.com/images/pdfs/EvapSpring04.pdf http://www.endwrench.com/current/02SpringInsiderInfo.pdf By the way, specific model info, etc., might help.

Has anyone checked the fuel pressure regulator operation?

So if the engine doesn't start, how long before the ECU cries 'uncle' and automatically has OnStar call a flatbed tow truck?

I'm looking at a diagram of the ECU wiring for a '97 Legacy (I assume the '96 is similar). The R/Y wire from the inhibitor switch contacts that close in Park/Neutral is shown connecting to a pin on the ECU. On the starter circuit diagram, the same inhibitor switch contact (and R/Y wire) goes to the starter solenoid. It would seem that since one cause of a P1507 code is the ECU not knowing when it's in P/N (http://endwrench.com/images/pdfs/DiagTroubSum04.pdf), that wire might be going to the input that allows the determination. As to whether the 197 ohm measurement explains the 'no start' condition, I'm less inclined to think that. As Cougar suggested, the inhibitor switch might be the culprit. Also, bad contacts in the starter itself are pretty common.

The older style can be done horizontally using a vice, while the newer type should be done vertically. The key is to go slowly, since if compressed too rapidly either kind can be damaged.

If that's actually what's happening, then the 'correct' procedure is to hold the accelerator pedal to the floor while cranking (don't pump it). That will help clear any excess fuel more rapidly, although it will probably still take somewhat more cranking than 'normal' to start. If doing that doesn't help, then there's probably another cause.

As far as I know, the TCU doesn't control the trans when in reverse. That seems to suggest that a mechanical failure is more likely, based on your observation. Electrically-related (solenoids, sensors, wiring) trans problems can generate trouble codes, but mechanical problems (even if with the same components) typically don't.

I'm not sure that it's the only possible cause, but see the first sentence on page 12 of: http://endwrench.com/pdf/feb2004pdf/4EAT.pdf Was the car parked where there might have been 'critters'? Chewed wires can lead to all kinds of problems.

Neither Dexron-II nor Dexron-III is being made any more -- GM won't license them (the current spec is Dexron-VI). The 'replacement' ATF for Dexron-III/Mercon should work fine, especially in a PS system, where it isn't nearly as critical as in a trans.

A failed piston seal will allow brake fluid to get past it, but it might not be obvious if it's just starting to go bad, since the boot can retain some fluid. Neither the piston seal nor the boot is difficult to replace, although care has to be taken to avoid marring surfaces. However, scoring or pitting of the piston and/or bore (often due to contaminated brake fluid) can also cause leakage, and a new seal won't resolve that.

Cylinder numbering, facing car from front: 3 4 1 2 In other words, #4 is nearest driver.

See post #4 of this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=53485 EDIT: If that isn't clear, 20330AC320 is the right-side one. Most of the time, you can go to one of the online Subaru OEM parts supplier's sites, search on a part number, and get a description.

What happens if the trans is manually downshifted immediately before or simultaneously with flooring it?

As long as the excessive parasitic drain stops when the interior lamp fuse is pulled, then a good approach is to put an ammeter in circuit by connecting it across the open fuse contacts. That way, you're measuring the current in that circuit alone. Since your car has non-stock accessories and wiring, it could be harder to diagnose without being there. Visual inspection might be a key to finding the problem. By the way, many trailer connectors have an 'always hot' circuit connection, and sometimes it gets corroded and conducts. If you have one, it's worth looking at.

If the dealer has a used '06 on the lot, ask to take it on a test drive. If its performance is significantly better than yours, consider asking for your $50 back and finding another servicer. Hesitation can be caused by several things. It can be due to ignition misfire, intake airflow obstruction or vacuum leaks, fuel delivery problems, etc. Those faults often don't generate OBD trouble codes until they become severe, but assuming you're willing to pay for the diagnostic time, the cause can be found anyway. What diagnostics were performed on your car, beyond scanning for codes? Does the response lag occur at higher speeds (such as attempting to pass another car), or just from a stop, etc.? When were air filter, fuel filter, etc., last replaced?

Further info on 'normal' 4EAT Phase-II operation, beginning on page 23 of: http://www.endwrench.com/current/Current6/03/InsiderInfo.pdf Personally, I'm not very happy with some of the design decisions made on the 99.

Some of the operation you've described is 'normal', or not too significant. See: http://endwrench.com/images/pdfs/1999Info.pdf http://endwrench.com/images/pdfs/ExhaustInfo.pdf http://endwrench.com/images/pdfs/LegacyIdleSum04.pdf Short trips, especially in the cold on vehicles with AWD, tend to lead to poor gas mileage. Lubricants are more viscous when cold, and the fuel/air mixture is significantly enriched while the engine management system is running 'open loop'. Slow engagement into Drive or Reverse is a known problem with the 1999 auto trans (4EAT) -- it's caused by a bad internal seal. I'm surprised that Rooster2 hasn't mentioned a possible 'cure' for it. See: http://www.ultimatesubaru.org/forum/showthread.php?t=87708 What maintenance (preventative and otherwise) has the car had? By the way, sometimes a car that consistently sees short runs can benefit from a long, hard run, perhaps with Techron added to the gas tank. I do find your #8 'interesting'; I'll have to give that one more thought.

According to the circuit diagram I have, the fuse you pulled is for the 'always hot' circuit, powering not only 'room' lights and clock, but also the security/keyless-entry. If the excessive parasitic drain began after you installed the map lights, then of course look over your own work. If not, and there is no sign of problems with the room/luggage lights, consider the possibility that the radio (preset memory keep-alive and clock circuits) or security module might be the cause -- disconnecting each of them and rechecking current draw would be the ticket.

There is no part known as the 'idle control system'. The word 'system' is used in the P1507 code definition because it's comprised of more than one part. A major component is the IACV (Idle Air Control [solenoid] Valve), which is perhaps what crash123 is talking about. However, before assuming it's the culprit, looking for intake vacuum leaks and improper connections as previously mentioned might be prudent. If no leaks, etc., are found, then cleaning of the IACV and throttle body, perhaps using SeaFoam, might do the trick before getting into replacement of expensive parts. While you're at it, cleaning the PCV could be worthwhile as well.

If the neutral switch and wiring seems okay, I'd suggest next looking for intake vacuum leaks. Check around the IACV, intake manifold, and throttle body, including gaskets (something not snugged down?) and hoses (cracked/disconnected).

There are a few possibilities, but before we get into some of them, check the neutral switch and wiring to it. See: http://endwrench.com/images/pdfs/DiagTroubSum04.pdf

You're welcome, and good luck.

Is there a particular pattern to the blinking (x number of blinks, pause, etc.)? See: http://www.cars101.com/subaru/keyless.html