Jump to content

2005 WRX - What is this leak? Water pump?


Nwge1234
 Share

Recommended Posts

Hi all, long time lurker and first time poster. I was investigating another leak (I'll post about that later...) when I noticed a leak at the edge of the sump/oil pan. I suspect it's the water pump, but it doesn't look all that much like coolant, it looks more like oil. I've cleaned it all up now so I'll check back in a few days and have another look. Any ideas what it is? My guesses are water pump gasket, or oil pan gasket. Just want to make sure before throwing money at it! Cheers.

 

w70pbm.jpg

 

28uitc9.jpg

Link to comment
Share on other sites

Thanks for the ideas. I'll check the sump bolts then, worth a go.

This is on the passenger side, right below the water pump. The black tube you can see in the 1st pic near the bottom the dipstick housing. The rocker cover is weeping slightly, but hasn't spread much. I'm pretty certain that's not the cause, but I'll double check.

Link to comment
Share on other sites


Ignore that last post ,as I couldn't see the pics ,no "4G" on the phone at the time and that looks like oil not coolant.

 

Be careful tightening the sump bolts too much (google the torque settings) as the block is ally and the steel bolts will strip the threads quite easily. Most subarus don't have a sump gasket and use liquid sealant instead , so sometimes nipping them up a bit doesn't help and can pull the sump outta shape . You can buy a group n Cork sump gasket as a upgrade though [emoji6]

 

Without seeing it in the flesh it's a bit difficult to pin point it ,so it could be any of the above . You could try removing the outer cambelt covers and checking if there's any oil leaking from the front crank and cam seals .

 

Link to comment
Share on other sites

Thanks for the ideas everyone. The rocker cover gaskets are weeping, but the oil hasn't extended that far from where it's coming out. I am certain that they're not causing this.

For now, I've cleaned it up around the leak area so everything is shiny again. I'll look back every few days to see if I can see anything. Fingers crossed it is the sump and I can just nip the bolts up a bit. Point taken about the torque for them, thanks.

The idea about removing the cambelt covers is a good one. Only thing is, looking at the 2nd photo above it looks like the leak is coming from around the water pump/block gasket area. The timing belt cover just is visible at the bottom of that picture where the 90deg black hose is, and seems pretty clean. I'll keep an eye on it anyway. First of all I want to know if it's oil or water!

You also mentioned the hardpipe that goes from the water pump to oil cooler. I'm suspicious of that anyway, so I'll have another look. It does seem to pop up on forums quite often as a pinhole leak sort of thing...

Thanks all.

Link to comment
Share on other sites

No problem bud, hope it's a cheap fix for you.

 

The cam belt covers have lines of seals , one to seal the cambelt backing plate to the block/heads and 1 to seal the front covers to the backing plate .

 

I mentioned this as it means oil can leak from a few places if the cam or crank seals are leaking .

It also would be piece of mind to know that the cambelt isn't oil contaminated [emoji6]

Link to comment
Share on other sites

The leak hasn't come back in a week now, so must be very slow whatever it is. Bigger fish to fry anyway, I think the head gasket/s might be on their way out, but things don't quite add up...

 

Air slowly (over a few months) accumulates in the coolant sufficient to cause the rad fans to come on and not turn off unless I rev the engine and churn up the coolant again. I suspect air accumulates against the back of the thermostat and causes it to stay shut. When the fans come on and don't go off, the bottom hose is cold. When they work correctly the bottom hose is hot and the 'stat works fine. If they don't go off, I rev the engine (3-4k rpm for 1-2 seconds) and the bottom hose goes hot and temperature will drop straight away (measuring with Torque app).

Driving 60-70mph uphills the temperaure starts to creep up. A brief blip of the throttle drops the temp down again in a few seconds.

I pressure tested the coolant system last night to 15 psi and then 20 psi. No leaks, held pressure. Dropped less than 0.5psi in a few minutes with a cold engine. No obvious leaks with a hot engine either.

Did a block test after draining some coolant out of the turbo tank. Both cold engine and hot engine. It changed from blue to yellow, but took almost 10 minutes of continuous bubbles, maybe 1-2 bubbles per second. Not conclusive.

Looking in the expansion tank it's clear that there is air in there - bubbles coming up. If there are that many, shouldn't it fail a block test pretty rapidly? I think air might get trapped in the heater core. It slowly builds up (slow leak?). But if it's from the cylinders then all that air should be exhaust gas, and hence make the block test fail. I'm stumped.

Air is getting in somehow. Has to be an internal or external leak. I can't see any external leaks, so I guess it must be internal. And yet it didn't show up bad on the block test. No oil in coolant. No coolant in oil (just did an oil change).

Radiator was replaced 6 months ago, and coolant at the same time.

Flushed heat core last month and it was fine.

The coolant in the expansion tank is always at the same level (checked cold). It goes up when the engine is hot, then back down as it cools. No bubbles I can see. If air is getting in, where is the coolant getting out? Level in turbo tank is always to the brim.

When it's cold I get steam from the exhaust and a few drips. When it's warm outside there's no steam. When the engine is hot there's no steam, either hot or cold outside.

If I bleed the air out (takes some effort but I can get it out) then it's good for a month or so, and then fills with air again.

I love the car but gonna have to get rid if I can't solve this. Any ideas very welcome!

Link to comment
Share on other sites

It's a bit strange the coolant level in the expansion tank doesn't go down. Whether it is loosing coolant internally or externally, when it cools down, it should sick coolant back from the expansion tank. Could it be that you've had a air lock since having the new rad? I'd stick some K-Seal in it.

 

Sent from my HTC Desire 530 using Tapatalk

 

 

Link to comment
Share on other sites

It sucks coolant back in no problem. When it's hot the expansion tank level is high, and drops down when the engine cools down. Tried several different rad caps, and they all worked ok.

It does have an airlock which causes 'sticking' thermostat issues and heater gurgling. Revving the engine shifts it, until it reaccumulates. That's not the problem though! The problem is where is the air coming from in the first place...

I put in some Holts radweld when I replaced the coolant 6 months ago. How much is too much? Should I put another bottle in? That sort of 'fix' doesn't sit well with me, but happy to be told otherwise.

Cheers.

Link to comment
Share on other sites

Headgaskets on scoobs do tend to blow between cylinder and coolant . Initial symptoms are ,when on boost it forces combustion gasses (hydrocarbons) into the coolant.

 

Is a "block test" to check for hydrocarbons ? .

 

Are you sure it's air in the system and not a sticky stat ,faulty pump or a flow issue (clogged coolant galleries or a kinked coolant pipe)

 

 

Link to comment
Share on other sites

If the block test changed colour from blue to yellow, that's a sure sign you have combustion gas getting into the coolant and the only way that'll happen that I know of is HG failure. It's probably early stage but still...There is probably blow through at one cylinder. Pressure test wouldn't necessarily prove anything in this regard - the problem occurs when the engine is hot and expansion happens.

Good luck anyway.

Link to comment
Share on other sites

Hi guys, thanks for getting back to me.

The block test checks for 'combustion gases'. I'm pretty sure it's just a pH reaction with CO2, seeing as you can make it change colour by blowing through it for a few minutes (there are no exhaust hydrocarbons in your lungs!) That's why I'm sceptical of it only changing after 5+ minutes, and even then only very slowly changing, because even air would make it change eventually (in my mind at least). If it changed in 30 seconds I'd be convinced.

I have found a few small chuncks of white crusty stuff floating around (a bit like hard chalk, probably from the radweld stuff). So it's possible there might be some small blockages or bits moving around, but that still doesn't explain how air is getting in. The gurgling noise is getting worse over time (several months) so I would assume more air is getting in.

I suspect you're right and it's early stage head gasket. Unfortunately if it is I'll have to move it on as the gf has had enough of it (and me too by now) and I can't justify £1000+ for someone to do it for me. Shame as I've spent quite a bit on the car already... What's really frustrating me is that it could be something major like the heads, or something tiny like a pinprick leak I haven't found yet.

Are there any other tests I can do to narrow things down a bit? All I can think of is an exhuast gas analyser stuffed inside the turbo tank.

Link to comment
Share on other sites

Fortunately I've never had to do a block test or "sniff test " as I know them .

 

It's possible that it could just be a massive air lock from when you replaced the rad a while back . As scoobs are notorious for air locks and have to be filled gradually making sure the header tank never get sucked dry during the refill .

 

Otherwise it does seem to be early signs of headgasket ,as initially it sometimes only leaks when on boost or heavy load (giving it beans uphill) . Due to the extra compression in these conditions forcing gasses past the partially failing headgasket .this also means that there's not always a lot of hydrocarbons in the coolant to start off with and it takes a while to build up the detectable levels (especially if you've refreshed the coolant recently )

 

You could try dropping the coolant and refilling without the engine running . Squeezing the hoses and making sure the header tank doesn't end up completely empty while topping up .

Once it's as full as possible turn the heating controls to hot ,start it up and continue to fill the header tank as necessary untill the fans cut in .

 

Hopefully it's just a case of air in the system and not a headgasket issue .

 

 

Link to comment
Share on other sites

Not sure about the thermostat being OEM. It opens at 88 degrees (bottom hose goes warm, temp verified by Torque app) so it certainly works. I've not heard of them intermittently failing closed, but I guess it's possible.

I've been advised that leak down tests are pretty useless for early stage head leaks. If the leak only occurs on the power stroke (several thousand psi?) then 100psi on a leakdown test likely won't be enough to trigger it. Do you think it a worthwhile test?

I was wondering about pumping compressed air into the cylinders one at a time and manually turning the engine over, to see if there's any point in the stroke that it leaks (looking for bubbles in turbo tank while I do it). Might be tricky to do.

My next plan was drain the system (again). Remove thermostat and check it (99% sure it's ok). Look inside the hole to see condition of water pump. Check timing belt tension (to see if it's slipping on the pump occasionally, long shot there). Flush coolant system (because of the white chalky stuff). Refill and try again. More radweld. Give up.

 

Last bit of the puzzle... If it's early stage HG leak, and it leaks compression gases to the coolant when under boost, then how does this work? Long gentle uphill gradient in 5th and the temp will creep up. Probably only a few psi boost at this point. Then drop it into 3rd, floor it (but maintain the same speed with the brake at the same time), i.e. lots of boost but the same airflow over the rad, and the temp will plummit back to normal. Eh!? The 'overheating' is related to engine speed, not boost/power. If I thrash it around the hills with high revs, the temperature sits stable at 88 degrees all day long.

Thanks.

Link to comment
Share on other sites

My friends old car would overheat, hot air only coming out when he turned or accelerated, fan would stay on etc.

He said later on the cause of it was broken blade in the waterpump

 

 

Sent from my iPad using Tapatalk

Link to comment
Share on other sites

If you do drop the coolant and remove the stat it might be possible to visually check the blades at the back of the pump ?

 

Tbh I can't remember as it's been a while since I built mine but due to the strange symptoms of the temps going down, in conditions when you'd imagine them to climb . A visual check of the pump ,oil cooler and general pipework from the pump might be worth doing , just to check if the galleries are clear and pumps not broken

Link to comment
Share on other sites

Hi all, thanks for the ongoing help.

As an intermediate step I've bypassed the heater matrix. The idea being that there is nowhere for the air to hide in the system, and it's one less thing that might not be working 100%. Apart from not having any heaters, the temperature is stable and working as expected.

When I bypassed the heater, I then bled the system to get any air out and top up the lost coolant. There is now less air in there than there was before, so that's probably why it's behaving itself. It's now a matter of waiting to see what happens next, which from previous experience is a couple of months for air to build up to noticable levels... Glad we're coming into summer!

If I end up doing more work on it then all the suggestions above are great - check thermostat, inspect pump, check pipes/galleries are clear as far as possible. I'll give it all a flush with the hosepipe too.

Let's see what happens next!

Link to comment
Share on other sites

  • 2 weeks later...

Two weeks on and it's all working fine. Thermostat must be working as temperatures are stable no matter what, as must the water pump. It's cold without the heater though! Level in the expansion tank goes up and down as normal. If anything, it's 1mm higher than it was when I first filled it up after bypassing the heater. That might just be my imagination so I'll keep an eye on it.

Took it for a hard drive yesterday and smelled the expansion tank when I got back (before it cooled down and drew fresh air back into it). Smelled of coolant, no exhaust. No bubbles. Nothing. I'm seriously confused with this now. I'll leave it another two weeks and then put the heater back in, after a full coolant flush.

Link to comment
Share on other sites

 

Glad it appears to have resolved the issue [emoji41]

 

 

Sounds like it might have been a airlock or blockage in the heater matrix. You could try attaching a bit of hosepipe to 1 the heater matrix pipes and run that into a bucket .Then attach the other heater matrix pipe to a hose pipe and see if you can flush it through with mains pressure .

 

 

This might rule out a blockage if it's free flowing or push a blockage out and into the bucket so you can see what was blocking it .

 

 

Best of luck and keep us posted bud

 

 

Sent from my SM-G900F using Tapatalk

 

 

 

 

Link to comment
Share on other sites

I hope you're right, but I've been through this before...

Air was gradually accumulating in the heater, causing intermittent overheating (not boiling over, but needle creeping up, fans not cutting out). Bled the system and it was fine. Then a few months later the same problem again.

So I removed the heater and it was fine again. Flushed the heater (clean) and put it back in again, and the same thing happened again. Took the heater back out, and checked for leaks, which is where I am now. Only this time I'm going to leave the heater out for longer and see what happens. I hope you're right! Cheers.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share




×
×
  • Create New...




Forums


News


Membership