89Ru

Yes it sure has. Today started it and had the low oil pressure knock (my guess), topped it off with a quart and the noise went away next start. Drove around the neighborhood, nothing newsworthy. Have to admit I'm a bit too familiar with the knacka knacka on older subes but this is the first I've heard it on on ej's. One fix on ea's was to tear down to the oil pump, same for ej's? Wouldn't think I'd have low oil pres from only down a quart. Sometime in the (near) future will replace the oil pump o-ring and tighten loose screws if any inside oil pump... Wasn't leaking oil through crank seal. More trivia: the crank timing sprocket doesn't have the guard above it to prevent belt slippage during periods where the tensioner is compressed and catching up to tighten a slack belt, somewhere I read this was added back on later models in 1997, might have been removed for a period of time (I think my '95 ej22 has the tbelt slip guard). May explain the tbelt slip, but weird thing is that only the right side exhaust cam appeared to be slipped one notch (late). All other cams were timed ok...could have measured wrong, some of the work was done by flashlight and mirror plus squinting

Thx for production year clarification... I have no special insight into when tensioners were changed other than this Endwrench article, page 16 "The new style cambelt tensioner was first installed during mid-1997 vehicle production" http://endwrench.com/images/pdfs/2.5Timing.pdf

FWIW, my EJ25 DOHC OBW has the old style two-piece tensioner (build date 1/97), was switched mid-year 1997 to the new style.

Took a while (mostly reading and re-reading dohc-lore, shopping for ej22 replacements shhhhh) but I finally got the dohc re-timed, slept on it, re-measured timing, handspun it around a few more times (delay tactics), spoke the mantra (embrace the interference engine...), convinced myself to be content with a 3000 lb lawn ornament, gave it up to Father, cranked it, and it started! Having thought this engine was doomed, I stared at it running in disbelief for a while, expecting sudden loud noises. Still might be in for surprises, so far it has only been up to temp twice, burping, hasn't left the driveway, but no codes . Here's a few notes, one noob to another. Driver side cams: Not having the special tool, a rectangular piece of wood cut to fit into the cams on the crank side gets pinched nicely when the belt is taken off. Top cam wants to spin ccw, bottom cam spin tendency is cw. They move off the marks though, so wrenches are needed for the tbelt installation to make small adjustments. The left side (drivers) cams are prone to snap at any moment when not held, some balancing of 17mm wrenches to get both to line up is necessary, a third arm would help here. Both snapped anyway mostly because I wasn't expecting the amount of force required to keep position. A right-angle mirror helps line up the exhaust cam notches if doing on-car. When in doubt, count teeth especially if re-installing a belt without (worn off) marks. When the last (lower smooth) idler is put on, left cams rotate off marks due to belt tension but when the (old style) tensioner is wedged left, its bolts tightened, and its pin pulled, marks come back into line after the crank is spun to take up tbelt slack. Never mind the double marks on the cams, exactness is desired and these aren't the marks to use unless one enjoys endless re-measuring.

ebay timing belt kits http://cgi.ebay.com/ebaymotors/96-97-SUBARU-LEGACY-2-5L-DOHC-TIMING-BELT-KIT-SET-EJ25_W0QQcmdZViewItemQQ_trksidZp4634Q2ec0Q2em14Q2el1262QQhashZitem5d22df0b01QQitemZ400017001217QQptZMotorsQ5fCarQ5fTruckQ5fPartsQ5fAccessories http://www.ultimatesubaru.org/forum/showthread.php?t=95187 which tensioner do you have? old or new? seems like only the new styles are packaged into ebay kits

First thing, stop cranking it. If its interference with a broken tbelt you're likely to be bending valves depending on where the cams landed. Second, some tbelt covers have small access ports in them for probing. Or pull the cooling fans so you can get access to remove the outer tbelt covers. Go easy on the bolts for the covers as some have captive nuts that spin out if siezed.

Pardon the questions but how do you know its burning oil? Isn't there a plastic plate inside the bellhousing that cracks and leaks oil? 123K on an ej22 is just getting broken in.

Of course, my job is to fabricate wild ideas to balance the common sense of the rest. By the way welcome to the board

No problem at all, like Cougar said, the high beam filament and the relay coil are in parallel. Those electrons won't get mixed up between the parallel circuits, they'll find there way back to the battery (-) sharing that white wire. Just make sure you wire in the switch somewhere in series with the relay coil loop or the driving lights will be on all the time (when your high beams are on).

This is all quite true, however, unless I'm mistaken, the measurements were made open circuit (looking into the two-wire mating connector for the high beam light so the filament was out of the circuit). Regardless, I believe Cougar has the solution, using both high beam green and white wires across the relay trigger contacts to energize the coil (of your bosch normally open relay) should result in the driving lights being on with the high beams.

Yes, white wire is the return to ground. The circuit puts voltage on the white wire to turn off the high beam bulb (i.e. both sides of the high beam bulb are at voltage, so no current flows). A connection to ground via the white wire is made when only the low beams are on. Not sure if both high and low bulbs are on in high beam mode. Pull the low beam connector to find out and make the same measurements as with the high beam connector. If the low bulb is off in high beam mode, you could use its return wire to ground to trigger your normally open relay.

I see that you need the logical 'inverse' of the white wire (voltage at high beams and no voltage at low beams). Maybe you can still use that pesky white wire. If you can find a 30 amp normally closed relay (swapped from the normally open version you have now) you can use the white wire to trigger it. Keep wiring as is. The Bosch relay doesn't have a set of normally closed contacts built in does it? Maybe another trip to Montpelier before the leaf peepers start swarming

If I can picture it, essentially you are belting (old belt) the cam gear and the crank sprocket together and clamping the two overlapping ends of the belt in the middle of the two so the belt doesn't slip. Now you can 'drive' the cam from a wrench on the crank sprocket and wrench the cam bolt loose. So you don't need a chain wrench on the cam with this cleverness. hmmm a lot more words, someone else could say it more elegantly?

One tooth off is enough to affect performance. No special tools are necessary, if you are replacing cam seals then a chain wrench is helpful, search around the board, there's lots of cleverness to discover. If taking it to a shop, expect to pay $350 or so - maybe Tom and Ray can do it for you if you're near Cambridge although beware, not all shops know subarus and you might end up with problems later. Leave a copy of the procedure laying on the passenger seat for the tech to discover it, or better yet do it yourself and make your own mistakes

Sounds like your onto something, get those subie's running before small woodland creatures discover them

I have a '95 outback 2.2 wagon sort-of, its stock low to the ground and with 195/60R15's gets 30 highway on a good day, fwiw.

Disconnect the plug wires, you don't want this thing firing up with a breaker bar on it

For a quick shield, use aluminum foil wrapped around the hot wire insulation, tie the foil to the (-) battery or some clean connection to body with some alligator clips. If this works, a more robust shield can be made using heavy gauge ground braid, either the wrap method or thread the hot lead through the birdcaged center of the braid. Maybe you have an electrical supply store in your area?

In addition to above suggestions, to snuff RF that may be picked up by this wire, you could add 1 uF capacitors between the hot lead and body. Experiment with this, at the battery, across the relay terminals, and at the lights. In all cases, the capacitor should be across the hot (+12V) and return (body) or shield if you go that route. I kind of doubt this, since your car's wiring is just a big antenna anyway, and just driving from place to place changes the fields that it sees from time to time...so the signature comparison would be like chasing a moving target... Here's a thought, maybe the anti-theft has a time domain reflectometer (puts out a narrow pulse and then listens for the reflection, like an echo detector). Kind of like if you tie a string tight between two trees and then strum it, you can see the echo 'wave' bouncing back and forth between the tie points. Any changes in the stored signature echo and bingo, your thief is caught. Yes, the existing wiring puts out a chorus of echoes like frogs in a pond, but any new frog, you get it.... If this is the case, a series power resistor (10-30 ohms) placed at the take-off point of the new hot wire (i.e. battery...) will dampen the reflection (in effect hiding the wire somewhat from the wire-sniffer), but it will lower the brightness of the lights, so not a great solution. Any amount of tinkering may not defeat the anti-theft, if it is really sensitive. How about contacting SOA and seeing if your hunch is correct? Or any friendly anti-theft installer? If you are right, then the anti-theft should have a mode where it can store the new signature, and off you go.

Tried free rotating the cams with wrench? See if one turns noticeably harder than the other. Woodruff key intact? sidenote to casual readers: Try this ONLY on non-interference engines please!

Try a new tensioner from the dealer (repeating someone else's wisdom here)

Original post mentioned pumping the gas pedal, does this do anything helpful in a fuel injected car?

Checked the timing marks (w/o crank mark yet, still covered) and with the driver side dohc cam double marks lined up top/bottom, the pass side marks are one tooth offset. Will begin searching whether this can cause all the symptoms in this verbose post...still unknown how the cams skipped a tooth but I'm betting on a partial cam seizure, hoping not! Could have skipped during the last hard start (haven't cranked it since), sounded like chj-jeeeeeeer chj-jeeeeeeer (low compression?) Need to save this car from its worst enemy (me!) Storm last night knocked a branch down that clobbered and spider-webbed the windshield, insult to injury! The house is divided whether this car will survive the blunder, and my future handiwork :-\

Some clarifiying questions. What you are doing with the return (-) lead? i.e where you make the connection to battery (-), back through the body? When you connect a wire to the source (battery or alternator) am I correct to say you have just a single wire (flying lead), or in other words, an antenna? How long is this single wire? By chance this single wire (antenna) is radiating some interference that is being picked up by your car's electronics, try twisting the wires together, this assumes there is a second wire (ground, back to battery) that you can in fact twist with the 12V (+) wire. If you are going through the body, this won't work. If I read correctly, it sounds like the lights come on as soon as you make contact with the battery, i.e. the relay is closed all the time. Is this true?

The obdii format started in subarus in '95, my guess is that earlier models were not readable via serial cable, so the flashing led is the only way to get codes. Scratch that, just reread your post re: OBD1, there are a few websites that support it. I use 'OBD Gauge' software (free) maybe they have a version for OBD1. With OBDII, you need a converter box to convert to RS232 for the laptop. Can't speak for OBD1 though.