Astra J 1.7 Diesel EcoFlex major coolant loss

[chriscross123]

Evening,

Long post, but hoping someone can help me.

Got a strange issue with my mark 6 Astra J (1.7 EcoFlex diesel, 12 plate 80K miles) - it is using huge amounts of coolant but no sign of a leak.

The coolant leak has got worse over time, until now it gets through about a litre of coolant in 30 miles. The symptoms are that when reaching the end of a journey the coolant level in the tank would look ok, or appear to have dropped slightly. It was only when you restarted the engine after it has cooled down or remove the header tank lid that the coolant level would drop suddenly and majorly. Temperature was rock solid as normal at 90°C on the gauge.

No evidence of an external leak, oil looks ok (no mayonnaise), no evidence of water out of the exhaust, coolant in the header tank looks ok (no oil evident).

Took it to my local Vauxhall dealer about a month ago, they had it for a day and couldn't find anything wrong .

I noticed droplets of water around the header tank vent, so took a punt and replaced header tank and lid. Seemed ok after a 200 mile motorway journey, then did a 30 mile journey a couple of days later and the problem was there. Did another long journey and used a little coolant, but not much.

Since then it has got worse, until a couple of weeks ago after about 30 motorway miles the car came up with an alarm sound and message 'Reduced engine power' and I noticed the temp gauge was up a bit from normal (maybe 95°C instead of 90°C).

Pulled over and pooped the bonnet - coolant level slightly low. Left it for a bit then very carefully loosened the header cap - lots of hissing and boiling and the level dropped like a stone until the tank was close to empty.

Back to the Vauxhall dealer - they had it for a week, said they had tested everything - block test, road test, etc - and couldn't find anything wrong.

Still the only problem I could see was a drip from the header tank vent.

So, I took the car to a different main dealer a week ago - he ran the diagnostics on the car and said it needed a new thermostat. £300+ later, and the problem was still there. Back to the dealer today - after a couple of hours he called to say no fault found so I asked him to keep looking, and after another few hours he called to say they had done another bottle test and this one was positive (previous ones were negative). He said this means either the EGR valve cooler (£1000) or head gasket (£1600) and the only way to find out was to do the job, so which would I like to try first . He did say Vauxhall may contribute as a good will gesture, but they would have to do the work first.

I would have thought that from the amount of coolant the car is losing it would be really obvious if it was going into the exhaust? It still looks to me like a circulation issue - ie there is a hot spot somewhere and the water is boiling off through the header tank vent, so I called an independent Vx specialist to get a price for a water pump, however he said my engine (A17DTE) requires a special tool so is a main dealer only job to replace, and water pump parts alone is £350.

I'd be grateful for any thoughts / suggestions of anything to check / anything else to do before I put the blindfold on and give my credit card to the Vx dealer.

Cheers

Chris

[stevieturbo]

You need to find a mechanic with a brain. That does rule out most main dealers.

Really...if it's losing that much water, it shouldnt be difficult to test and find out where from and daignose it properly.

[chriscross123]

stevieturbo said:

You need to find a mechanic with a brain. That does rule out most main dealers.

Really...if it's losing that much water, it shouldnt be difficult to test and find out where from and daignose it properly.

If anyone can suggest a good mechanic in the Southampton / Portsmouth area I'd be really grateful.

Cheers

Chris

[Pumaracing]

The most likely scenario here is a very small leak into the headgasket. As coolant is consumed it generates a vacuum in the cooling system which then prevents further coolant loss until you take the cap off. For this reason it could never have been the expansion tank or cap which would have to seal properly for this behaviour to manifest and certainly not the thermostat as the engine temperature always stayed constant.

However as Stevie says, main dealers are invariably idiots and can only either replace things at random or respond to fault codes on their computer.

1 litre of water in 30 miles won't show up in the exhaust. It'll just be a bit more invisible steam to add to the gallon of water produced in normal operation for every gallon of fuel consumed.

[stevieturbo]

chriscross123 said:

If anyone can suggest a good mechanic in the Southampton / Portsmouth area I'd be really grateful.

Cheers

Chris

I'd love to be able to answer that one, but too far away for me to know anywhere. But it's the correct question to ask.

This is more aimed at tech guys, but they might be able to point you in the direction of a reputable and competent garage local to you.

http://www.bba-reman.com/forums/

[chriscross123]

Pumaracing said:

The most likely scenario here is a very small leak into the headgasket.

Many thanks for the pointer and the explanation. After the last time I drove it, there was evidence again of coolant dripping (slowly) out of the header tank vent. I also noted that when I switched the engine off I could hear hissing from the vent - I grabbed some of my lad's bubble mix from the house and painted it onto the vent and it seemed to blow bubbles, as if it were over pressure rather than under vacuum?

Bearing in mind the comments regarding main dealers, how easy is it to do a bottle sniff test? The dealer said the last sniff test was positive, and when I queried how this had not been picked up before he said that previous sniff tests had been negative. From your comment about the likelihood of a very small headgasket leak does this mean that it might be 'easy' to miss with a sniff test?

The return line (small bore hose into the top of the header tank) appears to be spurting quite a lot when engine is hot (as if aerated). Never paid any attention to this before so not sure if it is normal or not?

I am assuming a worn water pump or blocked radiator would show up as a high temperature on the gauge? I ran the car up to temperature and monitored the radiator hoses - the hose on the right hand side of the radiator (nearest to the water pump) got hot but the hose on the left hand side (nearest the thermostat) stayed completely cold. Not sure if this indicates anything other than I didn't leave it to run for long enough?

Cheers

Chris

[chriscross123]

stevieturbo said:

This is more aimed at tech guys, but they might be able to point you in the direction of a reputable and competent garage local to you.

http://www.bba-reman.com/forums/

Thanks, I will give them a try.

Cheers

Chris

[chriscross123]

A bit more info - not tried putting videos on Youtube before but hopefully this will work. Both videos are with car idling at normal operating temperature indicated on the gauge after a short run.

Video of return line into header tank:

http://youtu.be/NWrG1rvP8aM

Video of drip from header tank vent:

http://youtu.be/BfRUkPr9vU4

Cheers

Chris

[stevieturbo]

What you will see on the gauge may bare no resemblence to actual temperature. For any sensible reading you'll need to hook up to the OBD port

I would expect a nice smooth flow of liquid via the return to the header, although it can be hard to see through the bottle

And under normal operation, no fluid should be chucked out the vent, but it does look like the fluid level in the bottle is quite high

[andyiley]

Can't be 100% sure, but I think I would be suspecting water pump not operating at full flow (possibly caused by a damaged impeller) is causing the water to overheat. The pressure in the system stops it from actually boiling off in the rad/expansion tank. Then when you open the cap, the pressure is released and the water boils out causing the level to end up low.

When you did your long motorway journey there COULD have been enough air-flow under the bonnet & through the engine bay to stop it from boiling, but then you do a slow 30 mile drive & the temperature is much higher due to less air flow through the engine bay, and the water temp rises enough to boil the water off through the pressure relief cap (expansion tank cap) which is the drops you are seeing.

If that makes sense.

As above the gauge is pretty meaningless.

Unfortunately my work servers do not allow me to see the videos.

Not the same (I stress) but several years ago I did have a 318 (e30 shape) that gave all signs of CHG failure but was actually the water pump impeller had actually mostly rusted away to virtually nothing.

Edited by andyiley on Saturday 1st August 20:28

[Pumaracing]

I'm puzzled here as to how there can be a vent in the header tank. Coolant systems need to be sealed to generate pressure to raise the boiling point of the coolant above 100 degrees C. Unless there's some kind of blow-off valve in that vent it shouldn't be there.

[andyiley]

Expansion tank lids also fulfil the function of over pressure relief for exactly that reason.

A quick search on the bay finds many that even show the pressure rating on them.

http://www.ebay.co.uk/itm/LAND-ROVER-DISCOVERY-3-a...

Edited by andyiley on Saturday 1st August 21:13

[stevieturbo]

Pumaracing said:

I'm puzzled here as to how there can be a vent in the header tank. Coolant systems need to be sealed to generate pressure to raise the boiling point of the coolant above 100 degrees C. Unless there's some kind of blow-off valve in that vent it shouldn't be there.

It's perfectly normal for most rad systems to have a pressure relief, this is what he is referring to. Modern plastic bottles sometimes route this discharge down and underneath the bottle

No different to the spring on the old pressed steel caps. The header tank here is part of the pressurised system

Some setups will also have an expansion bottle and cap that allows anything to be drawn back in as the engine cools.

Others just dump the excess, which could leave it low.

[chriscross123]

Thanks for all the responses,

As mentioned, yes the header tank cap includes a pressure relief valve (rated 20 psi according to the label on it), and the vent is downstream of this.

Thanks to for the suggestion about the water pump. I was suspicious about this, and even more so when I asked an independent Vauxhall specialist about changing it - as well as specifying it was a main dealer only job o replace, he also commented that there had been a design change to the pump (though didn't go into any detail). Only thing that points me away from the water pump is that the heater is still getting good and hot?

Cheers

Chris

[Pumaracing]

Ok so this vent is has a pressure relief valve in it. It still points to the head gasket. Either water is being sucked into the engine through the gasket or the coolant system is being pressurised until the vent blows off. Any other type of leak would leave the header tank unaffected until the coolant level dropped so low it started to boil.

[Pumaracing]

If it was the water pump the engine would overheat. You say it doesn't. That's pretty much the end of that debate.

[andyiley]

Not necessarily.

If it fails completely, yes, I agree, but as I said above if you have one with a steel impeller they can rust when incorrect coolant additives are used & will do so to the extent of rusting away the impeller blades until the pump efficiency decreases over time doing exactly what the OP says.

I had it happen many years ago on an e30.

The engine/coolant got hot, but did not get as bad as boiling. It was first noticeable in standing traffic and when I took the pump off to check it there were about half of the blades missing & the others weren't going to be far behind.

Whether the pump on this car is like that or not, I do not know.

[Pumaracing]

But the coolant doesn't get hot until it's lost so much that it boils. A water pump can't do that.

[crossy67]

Pumaracing said:

Ok so this vent is has a pressure relief valve in it. It still points to the head gasket. Either water is being sucked into the engine through the gasket or the coolant system is being pressurised until the vent blows off. Any other type of leak would leave the header tank unaffected until the coolant level dropped so low it started to boil.

I'm sorry but I can't see how an engine can suck coolant into its self.

Think of the pressures involved. Oil pressure ~2-5bar. Water pressure ~1.5bar static compression pressure ~20bar minimum, max vacuum -1bar.

I would check the rad is getting hot all over (watch those fans don't start when your hand's on the rad). Check temp via OBD. Sniff test. If the sniff test has come back positive you have a small combustion leak into the coolant system. weeping head gasket or small crack in head or block. This to me sounds the most plausible.

You cold have a small HG leak leading to over pressurisation of the coolant system, this blows the burst disc in the tank cap or forces the spring back allowing coolant to escape leaking through the overflow (the only time water should come out of it, it's a sealed circuit).

If the EGR cooler is leaking the water would be either leaking outside and visible or inside and again, visible in the oil (increased level and mayo).

I have no idea what the compression should be at cranking on that engine but I'd expect 20 bar minimum, at tick over more again then under acceleration when the turbo is more than doubling the charge pressure.... no idea but a hell of a lot. Any one know? When you do a motorway cruise you're not making as much boost as when you're cruising so the compression (which is what forces gas into the coolant) will be significantly less.

[mighty kitten]

When the plastic impeller falls of the vw Audi engines it rarely causes pressure to get high enough enough to push past the cap . Earlier versions of that diesel are well known for cracking the heads but I've never seen one that late with the same issue