2.5TD oil cooler repeated failure

[White Exec]

Back in 2005, and with 80k miles on the clock, I was forced to replace the dastardly oil-water heat exchanger on our S2 2,5TD. After a 12-month lay-up (which might well have been the trigger), and even with a decent antifreeze mix in place, the first outing for the car produced an alarming and spectacular eruption of black engine oil from the coolant header tank. It turned out that 2-3 litres of oil has crossed the oil-water boundary of the heat exchanger as its internals perforated. Interestingly, although the display immediately showed “water level low” (as the non-conductive oil coated the level sensor probes), the running temperature never climbed above 110degC. Just as well, as I had to limp home up some steep mountain tracks.

Club-XM’s site was really helpful then, and articles by Noz and xmexclusive invaluable. I removed the offending lump from beneath the oil filter, and cleaned it up. A test with compressed air revealed the internal leak, and so I checked on the price of a new replacement: Radiator huille, part no. 1103.F1, and a mere 617euros! Now I had noticed that the item was made by Valeo (and it had a number, which I’ve currently mislaid), and so off to my local Valeo distributor. But no, part available only from Peugeot-Citroen. So back to my Citroen parts depot, and out with the plastic.

Knowing now that these bits do fail, and knowing that (i) the original had lasted 80k miles, (ii) the car had been laid up, and (iii) I would henceforward be utterly meticulous with the antifreeze mix, I reckoned I would probably never have to replace this item again.

How wrong I was.

Over the next 2000mi, remnants of engine oil continued to float up to the top of the coolant header tank, producing a 1-2mm black oil slick, with clean coolant mix immediately below. This also destroyed the large and small rubber washers in the pressure cap. However often I removed the slick, a day later it reappeared.
Eventually, I came to suspect the head gasket (probably the oil-feed O-ring in one corner) as being the only other likely oil-water crossover point, on an otherwise perfectly-running engine. So, out comes engine (I have good friend with a full workshop facility), and head is checked (ok) and valves reseated, head gasket replaced (there might just have been a weep at the O-ring into a waterway, but nothing really conclusive), and every hose, belt and thermostat renewed - plus both radiators and clutch (while the engine’s out...).

Despite copious flushing, I quite expected the last remains of oil in the water to reappear for a while in the header tank, but, alarmingly, it continued as before: a slick of 1-2mm (which always coated the level sensor electrodes and thus indicated "low water level" – I just disconnected the sensor and shorted the connector out).

After a further 1500mi, in desperation, I jumped the replacement oil-water heat exchanger out of circuit (bridged the two water pipes together with a piece of 15mm tubing), and gave the entire cooling system a flush out with Liqui Moly Radiator Cleaner (their product no.2506). Brilliant stuff; Wurth make a similar product, and both are designed to remove oil from water coolant.

With fresh antifreeze in, and the oil cooler out of circuit on the water side, the system remained spotlessly clean! I ran it for a week or two (running temperature seemed unaffected), and then removed the unit and reverse air-pressure tested it. No internal leak evident. So I put it back on the car, and re-inserted it into the water circuit. Immediately, tiny beads of oil started to appear in the header tank.
So, the hugely expensive replacement had lasted less than 4000mi and a couple of years. My helpful local Spanish Citroen parts counter tell me they have seen all this before.

And so last month I fitted a Setrab 19-row oil cooling radiator, in the nose cone, and in front of the electric fans. Cooler, braided pipework and take-off plate (fits where the 1103.F1 did) were all from UK Mocal distributor, www.thinkauto.com, and cost around £250.

It looks . . . sorted.

I do, however, have an important caution for us on antifreeze, which I’ll post separately.

[xmexclusive]

Hi Chris

An interesting story that cost you dear in time and money.
Your final solution sounds quite elegant.
Not yet cut one of the original coolers up to see the insides.
They do seem not just failure prone but also very unpredictable for working life.

Regards

John

[moulin82150]

as the owner of a 2.5 td this happened to me and my local citroen expert told me youdont even need it so cut it out along with the diesel pre heater that goes porous that you dont need

[noz]

Hi Chris,

I've been through the whole cooler testing stage myself and been baffled as to the source of the oil in the water (see the other posts on this issue). However, I managed to find that the source was indeed the oil cooler rather than the head gasket which was also a leading suspect, before taking the engine out.

As a word of warning, the oil in the water not only goes for the rubber seal in the header tank cap, it also goes for the rubber hoses. If you squeeze each hose in turn you will easily be able to tell. When they have been attacked by the oil they go really soft and spongy to the touch. If that's the case, you may want to consider searching for replacement hoses now rather than waiting until they blow because when they do blow it will be at the worst possible moment, by definition.

Cheers

noz

[xmexclusive]

A couple of my 2.5's are now fitted with secondhand 2.1TD oil coolers.
You need to recover and use the threaded centre fixing as well as the cooler.
This is because of the different cooler thickness.
I ran one estate for over a year with the cooler bypassed without problems.
I agree there is little difference to water temperature.
In the end I decided that it was better to refit another cooler.
Without a cooler there is much less control over oil temperature.
The cooler and second thermostat fine control oil temperature.
This operates independantly of the main stat and engine water cooling.

John

[Dieselman]

As a doner cooler the one from the HP18 autobox looks like it would probably be suitable, if not the same, as the 2.5 engine heat exchanger as it's larger than the 2.1 engine exchanger.

[rg]

Folks,

I've just seen the first "tadpole" of oil in my header tank. Oh dear.

After a 380-mile motorway thrash, including 40 minutes crawling on the M62 with the fans cutting in (as normal), the level in the tank has not moved at all, so I'm not fearing the head gasket. At the moment.

So, is it safe to do the mod with the 15mm copper pipe and by-pass the thing? Surely XM oil temperatures are not all that critical? Can anyone comment?

And this is all after running the system with 50/50 blue anti-freeze and deionised water for the last ten years...

Rob

[techmanagain]

I have seen many messages on this site confirming that there has been no trouble after by-passing the wretched , expensive oil cooler. I am sure in my own mind that unless being in countries with extremes of temperature, no trouble will be experienced with that (fairly) common modification.

[rg]

Thank-you, sir!

I think it was Paul Johnson of XM Centre that it was pretty important. Maybe he wanted me to buy one...?

r

[citroenxm]

What would worry me on this matter is if the cooler is putting oil into the cooling system, bypassing the cooler then leaves the cooler pipes open.. Would you not get oil seeping from the water pipes on the cooler!!?? Making a mess?

Why is it the 2.5's suffer from this so commonly! In all the XUD 1.9 engines Ive been across, and theres been a good fair few Ive seen about TWO failed 1.9 coolers! One on a Xantia, the other on my mums 306 TD some years back..

In honesty, if the 1.9/2.1 coolers are so much better, whats wrong with leaving one on permanetly?

Surely a smaller cooler is better then none fitted at all?

The HDi coolers, though the pipes are in a slightly different position to the 1.9 XUD ones have also prooved reliable too..

[rg]

I don't think that there is a huge amount of oil coming out - I think the issue is generally pin-holes, not a total collapse. It just looks dramatic in the coolant!

I think it would be an easy matter to put plastic/rubber blanks in the holes and seal them in with sump-grade black silicon. The ones I'm thinking of are the type you get to blank holes in newly-supplied components, such as brake callipers. I'm currently doing a major sort-out of my workshop, and have found dozens of the things kept "just in case".

[citroenxm]

Thats just like me with 1.9 XUD ones, shed load.. (lol) but kept them Just in Case! Though they dont fail as frequently...

[xmexclusive]

I do not plug the pipes on an out of circuit 2.5 cooler initially.
I wait and see if they fill up and start to drip.
Push a short length of rubber pipe on then to join the ends and stop the oil loss.
Looking for the really bad one to section and see how its failed.

John

[rg]

A straight piece of 15mm copper pipe and clips has by-passed it effectively without putting too much strain on the hoses. The oil seems to have gone.

Any other suggestions on flushing additive, apart from the one mentioned above? Is detergent powder too corrosive? I'm sure I've read about it's use on a Skoda Octavia.

[xmexclusive]

I tend to use a dishwasher tablet from the box under the sink.
All sparkly and no rust in our dishwasher so must be ok.

John