Subaru Outback Forums banner

'02 OBW 2.5L coolant loss

1848 Views 11 Replies 5 Participants Last post by  plain OM
I drive an '02 2.5L OBW, with about 275,000km on the clock.

This weekend, shortly after our car was shut down, I noticed steam from the engine bay. Coolant was leaking from around the radiator cap. The temp gauge has never passed normal, just below half-way.

I later topped up with a couple litres of coolant, and replaced the radiator cap. I've begun checking the coolant each morning, and the radiator is typically down about 1/2 to 1 cup each morning.

I see no signs of obvious leaks. The level doesn't change if the car sits, it seems to require a heat/cool cycle.


1 - 12 of 12 Posts
The below assumes you're adding coolant to the overflow tank and this is all recent, you may have needed to 'burp' air from the system since it seems it was quite low.

At this point, you could just keep an eye on the level for the next few days. If it doesn't soon level out and stop requiring fluid, you may have a different issue.

confirm that the radiator fans are operating and watch the gauge - particularly while idling waiting for long lights/trains and under other conditions (exiting freeways, long pulls uphill, etc.) Monitor for odd heater/defroster temp changes.

I'm betting it will be OK. search for 'cooling system burping' and similar info.
Yes, all recent events, since this past Sunday.

I've been watching the temp gauge like a hawk, and it doesn't budge from about a needlewidth below "normal". Heat in the cabin seems utterly normal, and the gauge reaches normal in about the same place in the drive every day.

I'll look into burping. Fingers strongly crossed!

Thanks, I appreciate the advice you offer each time I post here.

Coolant was leaking from around the radiator cap.
Coolant should not leak out from around the radiator cap. When the pressure in the engine increases, excess coolant should be released to the external reservoir. If it's coming out around the radiator cap, the hose to the reservoir might be blocked or the radiator cap is defective.

If, the overflow hose to the reservoir is blocked, or the rad cap is defective, as the car heats up, the excess coolant is lost (as was evident -- leaking around the cap) and won't be recovered as the car cools, leaving the contents of the radiator down by what was lost.

I'd start by checking the hose to the external reservoir, making sure it's not blocked, that the bottom of the reservoir isn't full of accumulated "gunk", and that the hose fits snugly on the nipple just below the radiator cap. I'd also inspect the rad cap to be sure the rubber gaskets are intact and that the cap is locking down fully on the radiator neck.
See less See more
he said he changed the cap.

I suppose though, there could also be a crack near the rad cap neck.
If the cap has been changed and overflow tube is clear, must be a crack in the neck of the radiator. Especially if this is where the steam is visible, and under pressure.
Cap has been replaced, and overflow is now going to the overflow tank. I'm still using a cup or so of coolant per drive cycle (about 50km). I suspect it IS a head gasket, but the car's at the shop now for diagnosis.

I've called a few shops around, and keep getting a guestimate price of about $2k, which is kinda hard to justify for this vehicle.

We'll see.
Mechanic has confirmed a head gasket leak.


Seems almost 2k is the going rate for this repair 'round here. Hard choices in the near future.

Well, Job's Done!

Ultimately, it took me and my friend (an Aircraft Maintenance Engineer, ie airplane mechanic, LOL) 15 hours, 2 coffees (not counting morning doses), 3 Cokes and 8 beer. We started at 7pm Friday, worked to 10:30, then Saturday from 9:30 till 10:30 at night, with lunch breaks and the like.

We basically followed the sequence of operations in this post from the Outback forums:

I'd downloaded the factory service manual as well. Ain't the internet great!

This describes changing the head gaskets without removing the engine from the vehicle, we jacked the engine up by about 2", and used a rachet strap to pull it an inch or 2 side to side to access each side. The basic order of ops was:

remove induction plastic
remove windshield bottle
remove battery
remove bellypan
remove radfans
remove belts and tensioners
remove crank pulley (one of the few stumbling blocks, the pulley was stuck, but came loose eventually)
remove timing belt
undo engine mounts x 3
unbolt intake/exhaust mani's
jack engine
pull engine right
remove valve cover
remove head
pull engine left
remove valve cover
remove head
Reassembly reverse.

We learned alot about the car. The left head gasket had been changed before, but not the right. The right one failed. The right head's bolts were VERY tight, but we didn't need a snipe on the breaker bar, at least. Tight but not scary tight.
We took the heads to my friend's hangar, and cleaned 'em up with the varsol bin, plastic scrapers and finger nails. Straight-edging them against strong back light showed them to be as flat as the mark 1 eyeball could see. Subaru had previously advised that they have to plane less than 10% of the heads they take off, so we felt good about that.

We took about an hour cleaning surfaces on the block and nearby, with rags and sprays.

Torquing the heads is an odd thing on this car. THe recommended sequence is to torque 'em all in specified order to a medium torque (25ftlb approx), then tighter, to about 51ftlb, then detorque everything 360 degrees in 2 steps, then torque middle bolts to 22 ft lb, corners to 11 ft lb (in sequence), then 180 degrees on each bolt, 2 steps, in sequence. Took about half-an-hour to torque each head. At finish, they're VERY tight (squeaking/ticking), but not SCARY tight.

We also re-packed the timing belt idlers, as a short term measure. Their bearings felt smooth, but dry. It'll do for a while, till I feel like going back in. A timing belt job on this vehicle borders on easy.

So, by mid-afternoon, the heads were back on. At suppertime, it was down to the accessories and induction. Around 9:30, we were cranking the engine without plugs installed to push oil around. At 10, we had it idled up to temperature, burping the coolant system. A 15 minute test drive, and we loaded up the kids and went home.

Yesterday morning, the car took 2oz coolant in the rad. This morning, none. No coolant smell. No burning oil smell. Engine still looks clean.

It runs quieter.

Total cost: About $500, using all OEM parts from the dealership (whose prices were competative).

See less See more
Well, I thought I'd post some photos... Not much to share, just a couple snaps on my friend's cell.

First, the engine disassembled, lifted, and pulled to the driver's side to access the right head. Both heads are removed in this shot:

Here's the failed head gasket, from the passenger's side. The gasket is obviously blown near my finger, all the black coating is gone. Interestingly, the other cylinder in that gasket had a failure too, right about at 7 o'clock. The gasket seems to be made of thin steel sheet, and one is rolled around another to make the "ring" around each cylinder. The steel cracked along that "fold", maybe 1/2" long. There was minor "washing" of the black coating near this crack.

Looking at the heads and cylinders, all the cylinders had normal carbon buildup in them, except the blown cylinder. It had been steamed clean by the leak. The blown cylinder also had a pink residue on the plug tip.

All 4 cylinders still have hone marks visible in the bores. Was a bit surprised to see that. Heartening.

It was pretty cool to see the engine's design in such depth. The head/valvetrain design is ingenious in its simplicity.

We were both pleasantly surprised on how easy it was to work on. We had to dig in deep, but there was always room for the next operation. I have a few friends with Subaru's. I'll be willing to help when (if?) theirs blows up.

See less See more
I haven't heard anyone use "snipe" in years.

Good job. Education is great.
The gasket seems to be made of thin steel sheet, and one is rolled around another to make the "ring" around each cylinder. The steel cracked along that "fold", maybe 1/2" long.
Interesting. So is the original head gasket actually two full pieces, with one used to form the rolled around ring around the other at the cylinder opening, or is the gasket just one sheet with a rolled over section at the opening?

Any chance of a close-up photo of that particular area with the crack?
1 - 12 of 12 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.