I own a 2003, C5, 2 litre HDI diesel with 40,000 miles on the clock.
About six weeks ago the red stop warning came on and the lcd pannel
indicated the coolant was low. Coolant was topped up 30 miles later. Since
then it needs topped up after ever 50 - 60 miles taking between one, and
one and a half litres of coolant.
I'm doing 70 miles a day and most of that mileage is done at between 60 and
The system has been pressure tested and is ok. The header tank top and the
cylinder head gasket have been replaced (started simple and then went for
the big one), but neither has made any difference.
At the end of each 35 mile leg of my daily drive, with the engine still hot,
I have been loosening the blue valve on the RHS of the engine. This results
in a release of air under pressure and the header tank level drops further
down. Siumilar happens if I release the valve at the top of the radiator.
When allowed to cool overnight there is no pressure release.
At no time has the engine overheated - temperature guage always reads
normal. There has never been any trace of oil in the water in the header
tank and there has never been any sign of emulsification in the engine oil,
so no apparent leak from the cooling system into the engine oil. The coolant
just keeps being used - I suspect it is being vented through the header tank
cap but I can't be sure - when the bonnet is opened immediatly after a run
there is a faint hissing from the area of the cap.
Anyone any idea what the source of the problem might be?
Why do that? It doesn't sound like normal practice, to me; I'm not
familiar with the C5 though.
Are you sure it isn't water-vapour? Air would be unlikely. If it is
water-vapour then you are letting water out of of the system, which would
otherwise turn back into water as it cooled and the pressure in the system
returns to 'atmospheric'.
Releasing pressure from the hot system would be expected to result in
liquid being drawn in from the header-tank (which is probably still at
'working pressure'), on any liquid-cooled car I've lifted the bonnet on.
Does the water-level drop overnight if you don't open any valves while the
engine is hot?
The pressure-cap is meant to release excess pressure so it will sometimes
'hiss'. I think that's normal, as far as it goes. Having a pressure-cap
that opens at too low a pressure would probably lose coolant quicker than
normal - but don't fit a pressure-cap with a higher rating than the
As far as the loss of coolant you actually detect for yourself is
concerned, I would suspect you, releasing pressure from the system while
it's hot, for a start ;))
In the old days, radiator hoses could become porous and lose liquid slowly
and invisibly (one usually discovered the faulty hose when it ruptured
spectacularly while the car was being driven, if the hoses weren't all
replaced at the first sign of trouble). There could also be tiny leaks at
the places where the hoses are fixed onto the metal parts. A brief
'pressure test' might not show such things. I've even encountered a
radiator with tiny leaks where metal had eroded away or stones had damaged
the tubes, but that was in the '70s on a very old and neglected car with
many worse faults than that.
If the warning light is giving false warnings when the engine is in fact
not over-heating or running low on coolant, then the detector(s)
controlling that warning light would seem to be worth checking.
Of course, I could be barking up entirely the wrong tree.
Whiskers ( firstname.lastname@example.org) gurgled happily, sounding much like
they were saying :
The technical term is "Steam"...
There shouldn't BE any steam in the system if it's running at normal temp
(about 80deg), and correctly pressurised.
No. The bleed screws should only release water. There's definitely air
getting in there from somewhere, and that somewhere is where the water's
My suspicion is that the head's cracked, the block's gone porous, or some
other internal water leak.
Agreed; but releasing the pressure by opening the valve will cause the
water to evaporate. (You won't necessarily see 'clouds of steam' if the
air around the valve is warm and the pressure is enough to disperse the
vapour before it cools).
Any air getting in would have to be at a higher pressure than the liquid
in the cooling system, wouldn't it? If that is happening then the cooling
system will be getting over-pressured and the pressure-valve will be
operating to release the surplus, and that would be where the coolant is
escaping. But how is air getting pumped in?
Replacing the head gasket should have revealed anything as drastic as
that, shouldn't it?
Which should be attended to without any delay. Certainly no more
high-speed journeys in the meantime!
Wouldn't some sort of check be part of the job, if it were done
professionally? A suspect gasket is itself a sign of other possible
problems either as a result of the gasket failing, or causing the failure,
Well it won't show up a porous block, how can it, as for a warped or
cracked head - depends on the garage doing the job and what the owner
has agreed to pay - "My car is loosing water, fix it" and "How much to
change the head gasket on my car, Mate?" are two totally different
A porous cylinder head or block is a strange concept though; solid tough
metal capable of taking the stress of the combustion cycle, but letting
gas pass through to the cooling system? I'm not saying it can't happen,
just that its a strange thing to grasp.
In my experience an iffy head gasket leads to 'white smoke' from the
exhaust rather than 'air in the radiator'.
Hmm, I bet BMW wished that is was to, just how many engines did they
end up replacing for free due to their manufacturing methods /
technology creating porous cylinder blocks within the cars normal
Err, one of the by-products of burning petrol is water, so steam from
an exhaust is normal (one gets it even on air cooled engines), OTOH if
you rally are getting 'white smoke' from the exhaust you should
suspect your brake servo....
<Replying to both 'Wiskers' and the OP>
When the head gasket was replaced was it just swapped or was the head
checked for flatness (warp) and crack tested - if not you could still
have problems in that area. :~(
Indeed, many modern cooling systems need a 'head' of water to bleed
air out of the system when being filled, opening a bleed screw without
that 'head' could well induce air into the system rather than remove
If you are suggesting that there is *steam* in the system then there
should not be steam in the system - OTOH if you are suggesting that by
venting a pressurised system to atmosphere the water is turning to
steam after release then you might well have a point!
That would be normal, even if the head gasket is blowing.
The latter scenario is the most likely, I think. If "steam" is getting
trapped anywhere in the cooling system then there is a serious blockage
somewhere and the coolant is not able to circulate - and I'm surprised the
engine is still in one piece.
I used the term "water-vapour" because in colloquial usage, "steam"
usually implies visible clouds of condensing water rather than water in
its gaseous state as such - which might easily pass for "air".
Thanks for the input.
Releasing the blue valve is the latest attempt to fix this 6 week old
problem - I've only been doing it this week.
As the pressure releases - the water is being drawn in from the header tank
which is logical if there is an air pocket which has air which has expanded
with heat and forced the water in the system out into the header tank
thereby holding its level falsely high.
My suspicion is that it is air coming out, possibly with some steam as I can
release the valve with my bare fingers and still keep the skin on them. I
suspect steam that was hot enough to be invisible would be a bit less kind.
When I release the valve I keep it open until it sputters water then close
it before I get scalded.
The head gasket change and pressure testing were undertaken professionally
but not by a Citroen dealeer. I hope they would have detected a cracked
head. Also remember, no oil in the water and no water in the engine oil - no
guarantee but likely (?)
Today's attempt to deal with it was when I started the engine from cold with
the header tank cap off and bleed the blue valve until it ran water, then
did the same with the radiator bleed valve untill it ran water. I then put
the header tank cap on and did the first leg of my journey. One flaw in the
process was I didn't top up the header tank before putting the cap on. This
seemed hopeful but when cool the level had dropped - trouble is that was at
least partially caused by not topping up the tank at the start.
I repeated the process for the return journey, but topped up the tank before
putting the cap on. It looked promising (the tank level), but I wasn't about
to bleed the blue valve this time in case I got scalded. I'll see how it
looks when cool.
Thanks again for the input and discussion of ideas.
The service manager in the dealers asked me if there was white smoke on
startup. There isn't.
Re. the being able to touch it with bare hands - the valve is a screw on
plastic cap, off what I think is the water pump housing which also appears
to be plastic (?). As the cap is loosened the presure release of air / steam
then water is away from the fingers. Having said that I am vary enough of it
when I thought there was a chance of very hot water coming out of it . The
cap itself can be handled easily even whenthe metal of the engine is
The whole business is extremely baffling for the sheer volume of coolant it
can go through with no apparent indication of where it is going.
My current pet theory - which will be tested tomorrow - is as outlined
earlier that there is an air pocket, possibly in the radiator itself or the
hose between the radiator and water pump. Having bled the system with the
header tank top off and each bleed valve (radiator and water pump) opened in
turn, I'll see how it goes. I took it a run tonight and so far it's looking
I've followed the thread with great interest, it is a tricky one, no?
I hate to insult your intelligence but just to check:
You have had a new header tank cap?
You do know the header tank is not topped up to the top but only to the
level mark which is usually about halfway?
You are certain that there is no external leak of water when the engine is
Assuming all the above to be checked out I reckon you've got a rad leak. The
water comes out but evaporates on the hot rad so you don't see it. Then, as
it begins to cool it draws air in. You then open the bleed screw and it, er,
You could try looking for any crystalline deposits on the rad or joints but
there will be areas you can't see properly.
I don't usually reccommend this but after all you've been through I think
I'd be inclined to chuck some Radweld in and see if that cures the fault.
NO! NO! NO!!!!!!!!!!!
DO NOT USE THAT STUFF, IT WILL BUGGER THE ENTIRE COOLING SYSTEM UP AND
MORE THAN LIKELY CAUSE A BLOWN HEAD GASKET - OR EVEN SCRAP THE
I've actually had to replace engines due to damage caused by Radweld,
it certainly block up any leaks, the trouble is, it also blocks up
Cure the problem, don't mask it and store up even more problems.
Yes, New header tank cap fitted - same effect with new or old cap
Re header tank level: yes only to the recommended level.
Certainly no visible signs of a leak.
I think you description of the sequence is right just not sure of the
source - could perhaps be the radiator as you say - I also suspect the air
that is drawn in is expanding with the heat of the engine and pushing
coolant out of the cap which accounts for the volume being lost - if I bleed
the system each time I fill it - as I started doing last week - I'm getting
an extra 30 mile journey ourt of it before topping up with 1.5 - 2 litres
The car is back at the dealer and they are trying to identify the problem -
nothing showed when the engine was run while the car was on a ramp, so I've
said by all means give it a 30 mile run at m'way speeds.
You may well be onto something - proving / confirming it is the problem.
So far the work has been carried out under warranty - therefore I've
resisted the Radweld option. Reading Jerry's post I'll maybe give it a miss
I'll post the solution if and when it is found - some other C5 owner may be
out there with a coolant loss just waitin' to happen!
Motorsforum.com is a website by car enthusiasts for car enthusiasts. It is not affiliated with any of the car or spare part manufacturers or car dealers discussed here.
All logos and trade names are the property of their respective owners.