Subaru WRX Forum banner

2005 WRX - Oil cooler leaking water?

1684 Views 2 Replies 2 Participants Last post by  Nwge1234
I changed the oil yesterday and noticed a green stain around the oil cooler. I'm assuming the cooler contains copper and I'm seeing the green copper corrosion. As far as I know that doesn't happen with oil! The two coolant hoses seems well attached, and the green (corrosion?) is higher up than them anyway. A few ideas:

- It's an oil leak and turning green for some reason -> change the oil cooler seal. Easiest fix, but I doubt it's that.
- The oil cooler itself is leaking water from somewhere, maybe through the walls/tubes to the external surface. No idea... replace cooler. $$$
- There's a water leak higher up, but I couldn't see anything like that all. Any ideas?
- It's water splashed up from the road. I can hope, but I don't think it's this one unfortunately. Happy to be proven wrong!

I've had the car just over half a year. I changed the coolant soon after getting it and it wasn't clean, but didn't look tooo bad. There are a few different leaks around the place, and I should've been suspicous of the overly clean engine bay when I looked at it. Too late now! Any ideas about this one would be very welcome. I've read loads of threads about oil leaks from the oil cooler, but haven't seen any about water leaks. Thanks.

Oh yeh, I was worried about the head gasket leaking coolant, but it looks reasonably clean around there. I tried to catch it in the photo so you could see it, so any thoughts about that would be appreciated too.
See less See more
1 - 3 of 3 Posts
They appear to be copper. Considering how much corrosive stuff has splashed up on the heat shield right next to it, I don't think it is a stretch to think the corrosion on the copper came from road splash. Head gaskets on turbo Subarus are very very reliable, unlike the NA 2.5 for example, but I can't really see what you might be pointing out there. You could polish it up and if it turns green again right away there's your answer. Personally I wouldn't be worried about any of that stuff, unless you are noticing coolant loss and having to refill it.
Hi, thanks for your reply. I'm inclined to agree about the road spray causing it. Two things make me hesitate. Firstly is that it's had the plastic undertray on all it's life. Although as you point out, the heat shield is very corroded too so some spray must be getting in there. Secondly, I do have a slight coolant problem. Here's the essay!

I do have a slight coolant problem as well, which again makes me suspect something odd going on. I bled the coolant about 6 months ago because of a gurgling noise in the heater core and a very slightly leaking radiator. At the same time I noticed that the radiator fans would come on very quickly once stopped (in cool weather), and then not go off again. Using the Torque app I could see the temperature reach 96 and the fans kick in as intended, but then the temperature wouldn't drop at all until I revved the engine. Also on long gentle uphills in 5th gear around 2-3k rpm, the temperature would creep up to around 94-95 degrees, and then only drop when I revved the engine for a second or two in neutral while coasting along. Within a few seconds it would drop back to 88 degrees. On the flat it sits at 88 degrees all day long, which is correct. As the air builds up in the system it is less inclined to sit at 88, but instead sits and 90, then 91, then 92 over a period of weeks. There was air in the system causing a problem, which was getting worse. Why?

I flushed the heater core. No debris, easy flow.
New radiator (the last one was weeping anyway so needed to be done).
Replaced coolant.
New radiator caps.
Took ages to bleed, constant small bubbles in turbo tank that only very slowly decreased, but never went away completely. (I've tried every documented way and spents days and days on this, it's not to do with how I'm bleeding it).
All was fine for a month or so, until the gurgling (and radiator fans) would slowly start again. Air was building up again.
When this happened, I could tell the flow through the heater core was restricted/slow - inlet was hot and outlet was luke warm, i.e. no/little flow, with heaters on or off.
I bypassed the heater core and all was good again for the few weeks I tested it. It was easy to bleed because there was nowhere for air to hide.
Reflushed the core (still ok) and reconnected. Hard to bleed, and the same problems gradually built up again over a month or so which is where I am now.
New radiator caps again too.
Checked all hoses, new hose clamps on all, no obvious leaks.
Thought about head gaskets. If it is them, it must be very early stages of failure. No bubbles in overflow/expansion tank at all, only turbo tank. No overpressurising of the system. Gaskets visually looked ok from under the car. Bubble test (for exhaust gases in coolant, tested on turbo res. tank) was non-conclusive. It did change colour, slowly, but only after about 6-7 minutes, not the expected 1-2 minutes. It takes a month or so for a significant quantity of air to build up on the coolant system and start to cause problems. No other signs of head gasket failure (exhaust steam/smell, oil in coolant, coolant in oil) so I'm assuming for now that they're ok...

Air is entering the system (gasket, leaky hose, oil cooler etc). This accumulates in the heater core (I can hear it and feel the flow/return pipes). When sufficient air accumulates it causes a (partial) blockage and the engine temperature starts to increase (never actually properly overheated, just seen the dash gauge go up a bit and Torque show 98/99 degrees (rather than the ususal 88), the rad fans come on at 96 degrees but nothing cools down until I rev it 3-4k rpm for a few seconds. It then cools down straight away, along with a big gurgling noise. People queueing in traffic must think I'm annoying!

The radiator fans stay on for 2 possible reasons that I can think of: 1, The heater core is (partially) blocked by air and no flow is returning to the thermostat to open it, until I rev the engine and get everything moving again. 2, Air is accumulating around the temperature sensor and it's reading higher than it actually is, until I rev the engine and get things moving again and surround the sensor with water rather than air. I think option 1 is most likely, but it doesn't really matter - air is the problem. 3, It feels like a sticking thermostat, but I'm 99% sure it's not - see next paragraph for why.

When it's fully bled normal operation is: 88 degrees on the flat all day long, +/- 1 degrees max when going uphill/downhill. Stopping at lights/traffic and the coolant creeps up as you'd expect. Hits 96 degrees, the fans come on, drops to about 90 degrees in a few seconds. Fans shutdown. No problem at all. This tells me the thermostat, water pump and fans are all ok.

The odd thing is, the coolant level (in turbo tank and expansion tank) never changes (measured when cold). Literally it's always *exactly* the same when it's cold. Level goes up when hot as expected, but always drops to the exact same level overnight. So if air is accumulating, where's the coolant going? Not worked that bit out yet. Maybe it's a one in, one out policy at a small leak site...

Assuming the head gaskets are ok, air must be leaking in somewhere else. Which is why I'm now querying the oil cooler (it might explain a relatively slow leak under certain conditons), and my other thread trying to identify another very slow leak near the water pump/oil sump area... Sorry for the essay, but you did ask!
See less See more
1 - 3 of 3 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.