S3 Cooling (or not!)

[kevroe]

I know there's probably loads on this but the search engine is not working so any guidance appreciated. 1991 S3 by the way. So turn engine on and before long the swirl pipe is emitting steam. The pipe between it and the rad is cold. Drive belt to water pump is driving. No heat through car heater. Any ideas?

Thermostat? Block in radiator? Water pump? What would be best place to start?

[phillpot]

Huge air-lock ?

have you bleed the radiator (little vent screw on top,n/s)

[AutoAndy]

phillpot said:

have you bleed the radiator (little vent screw on top,n/s)

..or bottom opposite corner if its it like mine - In that case I have to un-clamp the top hose to let the air out.

in addition, I had an issue when changing swirl pots as the new one had a leak in the weld, so check where the steam is coming from and it is not a weld pin-hole that needs repairing in the swirl pot

[kevroe]

Yes thanks for this. Steam is definitely pouring out of top of swirl pot. I have the non-spring cap on that with the spring/pressure cap version on the expansion tank. The rad does have a bleed valve on top n/s so will have a fiddle with that.

[phillpot]

AutoAndy said:

..or bottom opposite corner if its it like mine - In that case I have to un-clamp the top hose to let the air out.

get yourself one of these, drill a hole and solder it in


Guess it makes a handy drain plug where it is !

[TurboTony]

kevroe said:

I have the non-spring cap on that with the spring/pressure cap version on the expansion tank. .

Anyone thinking of debating this ? No! Just don't go there.

Wars have started over less........

[phillpot]

Nothing to debate, our friend has it wrong !!!!!

[teamlines]

The swirl pot on my 91 S3 was replaced last year. Over the past few months I'd started to notice murky water puddles under the car after long spins and the swirl pot itself had dirty water stains all over it. Closer inspection uncovered a couple of problems. The cap on the swirl pot wasn't fitting properly and there was a crack in one of the welds at the top. There was also a block between the pot and the expansion tank. I've had it rewelded, fixed the block in the pipe and fitted a new pressure cap to the swirl pot. All seems to be ok now.

Will

Ps I've found that the swirl pot doesn't like to be too full. I imagine all the cars are different though, so probably not much use to declare that.

Pps I've got the pressure cap on the swirl pot and a blank on the expansion tank. Don't want to start any arguments though!

[glenrobbo]

phillpot said:

Nothing to debate, our friend has it wrong !!!!!

Well, Austin runs perfectly well with the opposite arrangement, a la Kevroe, never overheated* or lost coolant, and has always been that way to the best of my knowledge.
That's Fate well & truly tempted then.

Each to their own I guess, whatever works for them.

And that's when the fighting started officer.........

• Well to be honest it was getting a bit warm on lap 4 of Spa, but I was caning it a bit, and it soon came back to normal when I eased off a touch.

Edited by glenrobbo on Wednesday 27th March 20:50

[techbotics]

Assuming car not overheating I would look at cap (either) seal to swirlpot if you have the pressure cap on the swirlpot remember the cap seal will only come into play when system pressurised up to this point the inner cap seals the pot. If blank cap there you will prob notice leak all the time.

Do you have the rubber washer fitted to the swirlpot cap? Is it still in good condition?

Also check cap is pulling down properly.

Good luck

Damian S3

[TJC46]

TurboTony said:

kevroe said:

I have the non-spring cap on that with the spring/pressure cap version on the expansion tank. .

Anyone thinking of debating this ? No! Just don't go there.

Wars have started over less........

Go on then ill give you all food for thought.

Im in the plain cap on the swirl pot, and pressure cap on the expansion tank camp.

Why?

Well on my Chimaera the swirl pot does not have any type of cap.

It is basically a sealed system, with a small outlet in the swirl pot, with a hose connection to the expansion tank.

The expansion tank only holds 25mm to 50mm of coolant when cold.All other volume within the tank is, [as its name suggests] for expansion.

Once the engine is running the temp, and specific to this argument the pressure, of the coolant increases, and it expands into the expansion tank.

When the engine is switched off,eventually everything starts to cool down,the pressure decreases and coolant is drawn out of the expansion tank and back into the swirl pot/engine block.

This is a "designed" system so if all is working as it should there should be no leaks, no overheating problems etc.

The EXPANSION TANK HAS A PRESSURE CAP on top of it, and if the temp and pressure goes above the systems designed tolerance, it will vent it to atmosphere.[usually the garage floor]

Now relate the description of the above system, to the S series cars.

In this example..
The swirl pot has a plain cap on top.

With the engine running,there is increase in temp and pressure. The coolant expands and is forced out of the swirl pot, via the small hose and into the expansion tank.
As with the chim, switch off the engine, it eventually starts to cool and coolant returns back from the expansion tank, to the swirl pot/engine block.

Again if the temp and pressure goes above the design tolerances of the cooling system on the S series cars, it can vent to the atmosphere, via the PRESSURE CAP FITTED ONTO THE EXPANSION TANK.

Thats my theory anyway......awaits to be shot down in flames.


Now fit the pressure cap on the swirl pot, and as the temp and pressure rise it will have to reach the design limit of the cooling system to "lift" the spring in the pressure cap and vent itself into the expansion tank.

This has to be wrong, the coolant as it expands, needs to do so without resistance.
Only when the temp and pressure is too high should it be lifting pressure cap springs, and then only to vent to atmosphere.

NOW MY QUESTION IS.......How is this vented out coolant able to return past the spring loaded pressure cap and go back into the swirl pot as everything cools down?

Edited to add....off to get a beer and await replies.

Edited by TJC46 on Wednesday 27th March 21:45

[phillpot]

TJC46 said:

NOW MY QUESTION IS.......How is this vented out coolant able to return past the spring loaded pressure cap and go back into the swirl pot as everything cools down?

Because the pressure cap also has a return valve that will open if the cooling system drops to a negative pressure,allowing coolant (that hasn't been dumped on the floor but contained in the expansion tank) to be drawn back into the system. this is why the little pipe from the swirl pot neck goes into the bottom of the expansion tank (a tank to collect coolant as it expands, the clue's in the name)so it is submerged and will draw coolant back and not air.

Systems like this, and things like crankcase breathers going back into the induction, were developed because it became unacceptable to have coolant (i.e. anti-freeze which is not nice stuff) and oil vapour dumped into the air or onto the roads we have to drive on.

Chimaera is "next generation" totally different.....Lecture over

[Scoobimax]

Strangely I find this subject riveting due to my "problems" over the past couple of years. I can see both arguments but once I've got my cooling system all back together with the new bits, I'm going to try Phillpot's (and the bible's) way - I'm currently running Tom's set up and have been for the past 12 months, but wonder if my issues have been symptoms rather than causes :-/

I'll keep you all posted. I feel the topic has plenty of mileage in it yet though.

[greymrj]

Hi Kevin, when you say you soon get steam from the swirl pot, is this with the engine idling? My S runs absolutely fine on the road, no sign of overheating except in traffic when the fan comes on and all then works as it should. If however I start the car and leave it ticking over I often get the same issue as you. What I found is that if I blip the throttle a couple of times the flow through the cooling system starts to work properly. It seems that I get an air lock forming as the engine cools which the water pump can deal with once it is going fast enough! Good enough excuse for waking the neighbours on a Sunday morning
In my swirl pot the water always settles a couple of inches down. If I fill it more than that it just gets pushed into the expansion tank.

As far as the great debate on placing of the two caps is concerned, what I do is..............and that works fine

[glenrobbo]

When cool, my swirlpot is always brim full of coolant, & the expansion bottle level is just above the moulded indicator bar thingy. Works for me.

So far everyone has avoided pointing a finger at the thermostat.
Is it not possible that this is the culprit if stuck closed?

Edit: Oh, just re-read o.p . No heat from the heater.
Def sounds like airlock.

Edited by glenrobbo on Wednesday 27th March 23:51

[Griffinr]

greymrj said:

In my swirl pot the water always settles a couple of inches down. If I fill it more than that it just gets pushed into the expansion tank.

Probably due to the swirl pot cap not quite sealing properly as the air should be forced out when it's hot and then coolant sucked back as it cools. A very slight leak in the seal on top of the filler neck will allow air in as it cools. My swirl pot is always full to the top as long as I keep the expansion tank topped up.
Rob.

[Gerald-TVR]

phillpot said:

Because the pressure cap also has a return valve

As long as you get the correct pressure cap. I fitted one which wasnt and every time the system cooled down I had to top up in the swirl pot by my 2-3 inches. re-fitted the earlier (not so pretty) cap and all was well.

EITHER way works with the correct kit!!!