Boxer losing coolant

Well, after a 500 mile non stop trip from Newcastle at motorway speeds (70mph ish mostly) I was horrified to find the engine coolant header tank was about 10mm under the minimum Mark. The system has never previously lost coolant (owned by me from new) and it never should. I could find no external leak , so fearing a head gasket failure, booked it in to the garage. They kept the van for a couple of days for an extended pressure drop test, sniffer test etc and could find absolutely nothing wrong. To be honest I think they thought I was imagining it as I had topped up prior to taking it to them. The system continues to use about a mugful per 500 miles, seemingly at higher speeds (could just be that long journeys are usually at higher speeds). 
Has anybody out there had a similar problem and cured it? I have a posh expensive sealant to try, but would far rather know what the problem is and cure it 'properly'.
Thanks all, Tim

i know this will sound  silly,  as I know little / nothing about engines especially yours, 

But could it have anything to do with the heat we are having at the moment.?


Best wishes,
Margaret

Thanks Margaret, but no.
First occasion this happened was back in November. The coolant temperature has always remained normal, bang in the middle on the gauge.
Tim.

I had a similar problem, with about the same amount of coolant loss. I took it to the local Peugeot garage and they did the same tests you describe. I took it to another garage who found a very slight radiator leak where it could not be easily seen. It was weeping, rather than an obvious leak, and just looked a little damp. The radiator had corroded and was replaced. No further problem! I suggest you get a second opinion.

Just a thought, if you 'don't' top it up, will it drop by another 10mm?

Go for a pressure cap first.

PB

David, yes I've had a second (and 3rd) opinion, but I agree with you. There is no water in the oil or vice versa so the water does have to be going somewhere. The engine bay is still very clean and any antifreeze staining should be easy to spot and smell - I'll have a really close look at the radiator. I did have doubts about the hose connected to the oil cooler/heat exchanger. The hose clip is not standard Peugeot and was badly fitted, cutting into the hose. I've refitted it though without improvement. Only been serviced by Peugeot from new, so the 'wrong' hose clip is a mystery - can't imagine why it should ever have been removed unless there was some problem I've not been told about.
PB , yes the level does continue to drop until I panic when it gets below the minimum mark. This does take many miles though. As a matter of interest, when the pressure cap (1.4bar I think) releases, does the water actually spew out of the cap? Other forums talk about the coolant  'finding its own level' but nobody ever says where this coolant actually goes. My cap is actually clean as a whistle and I can be sure nothing has ever passed through, if this is indeed how pressure is released.
Many thanks to all. This is driving me potty and all suggestions are followed up.
Cheers, Tim

Have you ever seen it go under the 10mm lower than the minimum mark on the expansion tank? I ask because this is the level on my van, and has been for the eight years I've owned it. This is despite topping it up with antifreeze mix now and again. So yes, it does find it's own level.
Al.

hi
could it be leaking from the expansion tank overflow when under heavy load. look at temp gauge when going up long hills does it raise then go down on down hill runs.

possible cause radiator crystallization in bottom rad tank seen this many times while engineer at Peugeot./ mainly in older diesel models.
I found main cause was incorrect antifreeze allowing Borax corrosion. correct antifreeze inhibits borax corrosion.
this is just one cause could easily be other more simple causes but worth thinking about
johndean

Al, I've never had the nerve to see if the level will drop any further. I have owned the van from new though (it's a 2012 model) and the level had always been constant midway between max and min so never had to top up until the fairly radical drop in level the first time last November. I'm still baffled where the excess coolant goes when it finds it's own level - definitely not through the cap in my case - do you know if one of the small diameter hoses out of the top of the header tank is an overflow? I would be surprised as both pipes return into the system after routing along the top of the engine and radiator respectively. I can't find a schematic diagram of the cooling system anywhere and don't feel like throwing money at garages right now, as nobody can find what's wrong.
johndean, temp gauge never varies at all, no matter what speed/load/hills/ambient temp. Antifreeze is original Peugeot stuff and I doubt corrosion or crystallization at this age. Do you know which is the header tank overflow? Sorry can't add photos, but two small dia rubber hoses on top of Tank are plumbed back into the system, not vented to atmosphere.They are both hot when engine is warmed, so guess they are actually part of the flow cycle.Is the cap itself the overflow?
Thanks again all. Tim

Hi, don't know much about your engine but had a Peugeot car that was losing coolant like that, eventually found problem at cambelt change, water pump was leaking on longer journey's but drying up as fast, had water pump changed with belt and never lost a drop again, since then any cambelt I've had changed with a water pump behind it, had pump changed as well as only a little extra at time of belt change, just another idea if belt is due a change soon

Matt

I was losing coolant in a VW Topaz. A mystery for about six months until it dropped a considerable amount in a short distance. Transpired that there was a small leak in the coolant pump which only became visible when it failed.   Keith

Just a thought, but are you checking the coolant level when the engine is cold or at working temperature? As I've said, the level on mine is just below minimum, but this is when the engine is cold. If I look under the bonnet when the engine is hot the level is almost bang on the maximum line.
Al.

Al, I have a strip of masking tape stuck vertically on the header tank and check cold,hot, mid journey, end of journey and mark/photograph (told you I was going batty over this!). The cold to hot variance is nothing like as radical as on your van, perhaps 3mm. The drops in level on a long journey are only a few mm at a time.

Keith and Matt, two votes for water pump. I was assured this was checked at the time of the pressure test etc, but it looks a good bet to me. I do think that such a small loss could be a fine spray at higher speeds, temps and pressures , spraying onto hot metal and drying/evaporating to leave no visible evidence. Further investigation required.Only 25000 miles on the clock but would probably do a cambelt and water pump change for peace of mind. The current loss is easily manageable at present, but I do worry about a more catastrophic failure.
I will report back.
Thanks, Tim

Have you looked in the oil filler on the rocker box, I was told when I had a VW T5 a sure sign of water ingress would be a white oily residue around the hole and cap. As I say I was only told this and have no experience, but it was or is a common problem with water pumps leaking into the engine/oil. This was from a VW mechanic.
----------
Richard

Richard, yes you are quite right. The filler cap is nice and clean. No signs of water in the oil or vice versa.
Thanks.

Water in oil is more a sign of head gasket blown. Don't think a leak from water pump could find its way into oil.

postman wrote:Richard, yes you are quite right. The filler cap is nice and clean. No signs of water in the oil or vice versa.
Thanks.

Good news, hope you find the problem.
-----------
Richard

Paulmold wrote:Water in oil is more a sign of head gasket blown. Don't think a leak from water pump could find its way into oil.

True, but I was told by VW that water leak from pump can find it's way into the sump. In the case of VW T5's the pump tends to rust. I didn't have such a problem with mine.

Not sure if it happens in Warwick based engines, not read any reports, just a thought.
----------
Richard