Citroen Xantia 1.9 Turbo Diesel 1996

[scropionking]

Hi,

Another update ran the combustion leak test today, car upto 80 decrees, thermostat opened and watched it for half an hour and everything seems OK. The dye was meant to turn green if gases present in the system, but stayed blue. So everything looks good. But like I said the radiator has been getting worse every day and was leaking quite a bit today. Thermostat should be opening at 82 degrees, but gauge only reading 80c. So my question is would the gases still be present in the system even with a leak, does it seem a fair test. I think my fans are not working because the engine is not getting upto the correct temperature, so it would make sense to me, becuase its a sealed system. So I am going to buy another new radiator and do the whole job again. I don't know if any sensors are faulty at present, but first things first. No bubbles present in the system and there's no pressure
when I release the cap, but this is all void if the system not upto pressure. (ie) radiator leaking. Also no water vapour out of exhaust and no water present in the oil.


Thanks


Kevin

[white exec]

Pretty much good signs on your tests, Kevin. Obviously that rad needs replacing with a non-leaky one! The lack of pressure won't affect normal running temp, but pressure will stop it boiling!!!

You've already replaced the thermostat with a new one (82C), so that's probably ok, and nothing much changed in terms of gauge reading before and after its replacement.

Seem a couple of options for checking what the gauge is showing:
- check the resistance-vs-temp action of the gauge sender (gauge temp sensor). For this someone will have to find the technical data for it; I don't have that here.
- check exactly what temp the engine block is running at (measured at the water outlet, where the sensors are) . . . two options there:
(i) Use a digital multimeter temp probe, if you have one.
(ii) Use one of these, or similar: https://www.amazon.co.uk/Etekcity-Laser ... hermometer.

That last option would tell you exactly what's going on. You could check its accuracy with a container of boiling water.
At sea level, before anyone points out. Seriously, a 100°C and 0°C (ice-water mix) is easily done.

[sparksie]

And, always remember, there's usually a 5% tolerance so 80C versus 82C, especially measured with an analogue (probably moving iron) dashboard instrument, is pretty damn good.
Replace the leaky rad and accept that your different thermostat and disturbed sender electrics are going to be very slightly different from the old ones.
The colder weather may well be a factor, too.
My own Xantia is a non runner, waiting for me to fix the injector pump, but the Cordoba I'm driving only warms up enough to open the stat if I push it hard, or climb a steep hill. Other than that, the heater rad is enough to keep it way below normal, all by itself.
If you're really paranoid about it, wrap up warm, set the heater to cold, tape a cardboard box across the front of the rad and go for a drive.
Thus impaired, the cooling system should get up to something close to summer temperatures and bring on the fans.
Don't forget to remove the cardboard immediately after the test.
In general, you should expect slight differences when you change things and, as long as those differences are still within tolerance, it's all good

[white exec]

I really cannot agree with a lot of that, sparksie.

I haven't driven a car in decades which suffered from either gross over-cooling, or failure to get ACTUAL coolant temperature (not necessarily what the gauge displays) up to something approaching a proper 90°C in normal driving, even in the depths of winter, assuming everything, was working as it should. That is what a good thermostat is for - to exclude the main radiator circuit, and to allow the block and cabin heater to run at proper, hot, temperature.

If the coolant temperature is really as low as reported (hence the advice to check with a thermometer), then something is badly amiss. Sure, diesels driven away from cold and going downhill use little or no fuel, but a mile or so of normal driving should get things properly warmed up.

PSA's thermostats (82C in ths case) normally deliver what is needed, but if after all thorough temperature and gauge checks the system still runs verifiably chilly, then a slightly higher temp 'stat (eg 86 or 88C) may be called for . . . but only after proper temp measurement has been carried out.

There is nothing amiss in these diesels running at 88-90-95 in normal use, as the cooling fans will easily deal with the higher temperatures, and the system is pressurised to prevent boiling, even at 120+.

[GiveMeABreak]

And we established a few posts previously when your fans are supposed to kick in - not until 90 °C & 101 °C respectively.

Here is the data for your vehicle with A/C and the BITRON unit for info:

Vehicle type	-	Colour identification	Fan: Slow speed	Fan: High speed	Engine hot (refrigeration authorised)	Air conditioning cut-off	Pre-warning	Warning	Post-cooling (temperature and operation time) : degrees	Post-cooling (temperature and operation time) : Time
Turbocharged Diesel	With air conditioning	violet	90 °C	101 °C	70 °C	112 °C	115 °C	118 °C	101 °C	6 mn

I can only ever remember my fan coming on high speed a few summers ago on 220 mile trip home from Yorkshire in the blistering heat when I was stuck on a motorway traffic jam. And typically, that's also when my A/C compressor decided it was time to pack in - so with windows open that's how I heard the fan going at warp speed.

[sparksie]

Well, my thermostat didn't open at all yesterday and I drove nearly 100 miles!
The cabin was nice and toasty, the gauge was around quarter way, compared to the normal half way, and the radiator was cold. The top hose was warm to the touch when I was topping up the washer fluid, but that's due to heat transfer, not circulation. The 'stat on the Cordoba lives at the bottom of the engine, so this is normal behaviour. I've had several cars with this basic engine, bearing various badges, over the years and all had excellent heaters and ran noticeably cooler when those heaters were used in anger!
The OP is concerned about a 2degree IMPROVEMENT in cooling efficiency after fitting a new radiator and thermostat. As far as I can see, he's judging this by the dashboard instrument and those are notoriously inaccurate at the best of times.
I really don't think fitting a higher temperature 'stat would be advisable until he's tried it under more typical conditions. The current extremes of temperature are not typical and if he gets it up to the point where it brings on the fans under these conditions, it will struggle to keep cool when summer comes

[white exec]

No it won't. A higher temperature thermostat will make a difference in cold conditions, but under hot conditions will not - as it will be open. Under hot conditions, the fans and radiator will cope as usual, and the thermostat plays no part (and no restriction) above its fully-open temperature.

All the 'stat does, importantly, is to lift up the operating conditions of a cold or gently-used engine into a decent temperature band. Some low-temp 'stats (78, 80, 82°C...) just don't do that adequately, it seems.

[sparksie]

My advice remains not to deviate from standard until such time as it has been proven to be inadequate.
If, like mine, the OP's existing 82 deg stat is not opening, then neither will the higher temp one and, in any case, the fans will still not operate.
It is entirely possible that the heater is capable of emitting all the heat generated by a sensibly driven engine when the outside air temp is sub zero, meaning the engine never gets to the required 82(+/-)deg, especially when that engine is a diesel. Despite their recent bad press, diesels are much more thermally efficient and waste a lot less energy, as heat, than their petrol siblings.
Also, the opening temperature is actually relevant, because a thermostat is not a binary device. It is not merely open, or closed, but has an infinitely variable state, in between those two extremes, and peak cooling efficiency only occurs when FULLY open.
PSA decided 82deg was the optimum compromise to ensure reliability across a wide range of temperatures and deviating from that should only be done with full knowledge of the potential consequenses

[white exec]

My advice remains not to deviate from standard until such time as it has been proven to be inadequate.

Which is exactly what I said above.

I'm well aware that thermostats are progressive devices, and not just binary open or closed. At very high running temperatures - which you suggested could be adversely affected by using a higher figure thermostat - the device will be wide open, and not restricting coolant circulation or radiator/fan action. The thermostat's function is to control low temperature performance, not high ones. I did say all this in my earlier post.

[sparksie]

Not entering into a "you said, I said" discussion.
It's not getting hot enough to open the stat, or turn on the fans.
Fitting a hotter stat won't change that. Period.
I cut my teeth on this stuff...
Now, if we were talking about an ecu and multiple sensors, I'd be open to correction, but this stuff is as old as Santa Clause and I'm the best there is on that.
Cars run cooler in Winter. Now there's a shock!

[white exec]

Sorry, but simply saying 'I'm the best there is' isn't good enough.

What sort of engine - and especially a Xantia 1.9TD - doesn't get hot enough in normal driving (under anything less than sub-Arctic conditions) to begin to open the thermostat?

With a properly chosen thermostat, the running temperature of the coolant round the block (and thus cabin heater, and hopefully that shown on the gauge) will be able to rise to normal even when it is cold outside. That is the whole point of the 'stat: to prevent the engine running temperature being significantly influenced by low outside temperatures.

If the engine never reaches proper temperature in normal driving, then it is being over-cooled, and the 'stat is the very first port of call.

[scropionking]

Hi,

I didn't what to cause an arguement on the forum over this thread subject, the thermostat is opening and doing its job. So maybe I have a faulty gauge or faulty sensor. I am going to do the test that Chris has suggested and see if the temperature is above what the gauge is reading. I was only stating what the tempareture was before and when the fans would kick in, (ie) 85c, before I took out the old radiator. The fans kick in when A/C is turned on, and cut off but from the tables from Marc, it looks like a lower temperature margin. They come on when I disconnect the brown plug from the sensor cold ignition on or off and yesterday I decided to clean up the contacts after a run, engine switched off, disconnected plug again, fans came on low speed. I really don't know whats going on with them or if this should be happening. Maybe I have disturbed the wiring, because the loom is right in the way of the bottom hose. Anyway I've now got another new radiator and will fit sometime next week. Thanks everybody for your help.

Thanks

Kevin

[white exec]

Hi Kevin,

You've done really well with the tests, and the system looks as if it's doing its stuff now. Possible that a contact has been disturbed/cleaned up, but at least you know most of it works ok now.

Gauges have something of a reputation for inaccurate reading, and it could be that the sender isn't ok. As said, I don't have resistance-vs-temperature data for the sensor, so difficult for it to be tested. You could always just replace it (with a genuine item) and see what happens.

Re: pulling the brown plug on the (fan) sensor, what you describe is normal. It is designed to do that - fail safe (i.e. bring the fans on) in case the sensor or the connections to it fail open-circuit, to prevent engine overheating.

The disagreement about thermostat function is in no way down to you, so absolutely no need to apologise or feel responsible! Posts here (and on most technical forums) throw up all manner of different views, and is guaranteed to shake up the grey matter.

Hope all goes well with the new radiator. Let us know how you get on.

[sparksie]

Okay, I was in a lot of pain when I last posted on this topic and was neither as attentive, nor as diplomatic, as usual.
It is true that I'm rarely, if ever, wrong in the fields of thermodynamics, or analogue electronics. I either know the answer, or hold my tongue, but to claim to be the best is a bit OTT.
However, I did miss a crucial clue in an earlier post. The thermostat is supposed to start opening at 82deg, yet the OP said in one post, that at an indicated 80deg the 'stat opened. That is a very strong pointer to either a duff stat (not unheard of, but unlikely), or a difference in the electrical circuit for the gauge. Given the amount of disturbance to the wiring during all the chasing that has occurred, it's highly likely that some connectors have been cleaned up and are less resistive than before, or have loosened and become more so.
In any case, as I said before, dash gauges are not precision instruments and they are NOT to be trusted for such small variations.
Many French cars do away with gauges altogether and rely on a light.
The trouble is, by the time the light comes on, it's often too late.
With a gauge, the driver has the opportunity to observe a trend, as a problem develops.
With regard to the other comments, to simplify things, regard the heater as another radiator, not governed by the thermostat. It also has the benefit of forced airflow, thanks to the fan being used to circulate warm air into the cabin.
Under very cold conditions, that can be capable of dissipating enough heat, all by itself, to keep the engine cool, leaving the thermostat, cooling rad and fans doing nothing.
If, as was suggested, you alter the spec from standard, then you need to try and mimic the most severe extreme you're ever likely to encounter and make sure the modified system can cope. So, in the case of fitting a hotter stat, you need to replicate climbing a step hill, with a loaded trailer and full complement of passengers, with an outside temp in the 30s and the air-con going full blast.
Difficult to do when it's actually sub zero, but my suggested experiment, if carried out first, will at least prove that the cooling system operates as intended and turns on the fans at the appropriate times, which I thought might reassure the OP that he was chasing ghosts. It also costs nothing, can be done in the course of a normal journey and without tools, or professional equipment. For somebody as obviously attentive as the OP, the information gleaned would be of more use than the other tests he's had done and, potentially, save him further expense down the line.
I offered the suggestions with the intention of saving readers from making expensive mistakes, not to put anybody's nose out.
I apologise for having caused upset and for my knee-jerk reaction to it.
This, more reasoned, post will be my last, in this thread. People can make up their own minds as to whether I'm right or not.

[myglaren]

Something like this may or may not be useful, hard to be sure:

This example is on eBay and there are many variations