Expansion tank

manners · Post by **manners** » Thu Nov 15, 2007 9:00 pm

Hi does anybody know if the coolant in the expansion tank circulates or does it just sit there untill coolant is lost and the pressure just pushes enough through to replace the lost coolant?

mikeonb4c · Post by **mikeonb4c** » Thu Nov 15, 2007 11:16 pm

I think its the latter, the only circulation (and its not a designed circulation) being by convective circulation (assuming it has an 'in' pipe and an 'out' pipe - I've not checked mine). But I'm not a car mechanic!

dandywarhol · Post by **dandywarhol** » Fri Nov 16, 2007 12:02 am

Circulation I'd say. It's got 2 pipes - one down to the radiator and one coming from the head, therefore it will circulate by convection IMO

Socket Set Sue · Post by **Socket Set Sue** » Fri Nov 16, 2007 12:29 am

Two pipes, large one goes to the top of the radiator, the small one off the bottom side of the tank goes to the front heater feed pipe not the head on my 2.5 turbo 1995 2WD Bongo. I don't think you will get any circulation here but you will get a change in level due to coolant expansion when it gets hot and again when it cools down.

dandywarhol · Post by **dandywarhol** » Fri Nov 16, 2007 1:10 am

Ok, got that one wrong as to the small one but the head also feeds to the matrix surely? I still say it will circulate - I've watched it as I've changed the coolant and purged the system of air.

Socket Set Sue · Post by **Socket Set Sue** » Fri Nov 16, 2007 10:35 am

Hi dandy, must admit I have never looked to see if there is a coolant flow through the header tank but if you say so thats good enough for me.

alphabetter · Post by **alphabetter** » Fri Nov 16, 2007 1:00 pm

We (well me with help from others) drew this scehamtic a while back.

I agree that there is some circulation through the tank, but the whole scheme is rather odd as there are two circulation paths - one through the rad and the other through the tank.

Let me know if you have any comments/corrections.

Click image to see fullsize.

mikeonb4c · Post by **mikeonb4c** » Fri Nov 16, 2007 1:30 pm

Are we in danger of overcomplicating things here.

* Both pipes allow coolant to move to and from the header tank as expansion (with heating) and contraction (with cooling) take place.

* The net flow in or out of the header tank depends on whether the volume of coolant increases more or less than the volume of the system (piping etc).

* The 'spare' coolant in the tank is a reserve designed to ensure that no air gets drawn into the system as expansion and contraction and pulling/pushing coolant to and fro takes place. The coolant level in the header tank needs to be sufficient to allow for the most extreme case of heating and cooling. The different levels of coolant in the tank reflect the change in system displacement from cold to hot.

* The header tank needs a pressure cap so that the hydraulic pressure in the system caused by the water pump is opposed, thus making the header tank 'dead' (a backwater) to forced circulation. It needs a release pressure so it can vent in the event of system boil up.

So, no circulation is required for the header tank to do its job. However, convection caused by hot coolant rising and cold stuff falling MAY set up a lazy (but not pumped) circulation through the pipes. I would think, given that one pipe comes from the top of the rad. where the hottest coolant is (I assume?), that any flow will be into the header tank from the rad. and out of it towards wherever it is the other pipe goes.

Shoot me down chums, but gently huh?

dandywarhol · Post by **dandywarhol** » Fri Nov 16, 2007 2:47 pm

Coolant circulation in any system is by convection from the engine block/head to a cooler part of the system. The water pump is really only there to lift the cool liquid up to the engine from the lower extremities of the system (lower radiator tank for example).

In the days before water pumps were incorporated in cooling systems vehicles had the radiator higher than the engine block and relied purely on a thermo-syphon system which used convection currents.

I still say the "expansion" tank is a part where coolant circulates whether the 'stat is open or not. The bypass valve on the top of the 'stat will allow the coolant to circulate in the secondary (expansion) tank and the myriad of metal/rubber pipes when the 'stat is closed.
When the conditions dictate that extra cooling is required the 'stat (on the return circuit) opens and allows additional cooling.

That's my take on it - this could be another LONG thread

PS - well seeing we're both incapacitated with bad backs Mike - my keyboard has been taking a pounding these last few days

mikeonb4c · Post by **mikeonb4c** » Fri Nov 16, 2007 3:06 pm

Its the Curse of the Diesel O Ring Dandy

Hmmm another piece of the jigsaw. Which side of the thermostat does the header tank pipe connect to. My guess would be the downstream side. That way it would be in constant communication with the bulk of the coolant system and water could flow either way through it as the engine heats and cools, irrespective of whether the thermostat is shut. That would make sense as the thermostat stays shut precisely so circulating coolant doesnt slow up the engine from reaching operating temp. And so a lot of the expansion during engine warm up would take place with the thermostat shut and it would be important that the pipe to the header tank wasnt out of communication due to a shut thermostat. Would that sound about right to others?

dandywarhol · Post by **dandywarhol** » Fri Nov 16, 2007 6:12 pm

I would need to se one in the flesh/metal but according to the diagram it's at the closed side. This makes some kind of sense as when the main thermostat is closed then the bypass valve on the top of it is open allowing the coolant to bypas the radiator circuit.

haydn callow · Post by **haydn callow** » Fri Nov 16, 2007 6:18 pm

It does circulate through the header tank weather the stat is open or not. Just feel the heat in the tank and both pipes so soon after starting. This enables the heaters to work before the stat opens.

mikeonb4c · Post by **mikeonb4c** » Fri Nov 16, 2007 6:45 pm

Ah ha! We're getting somewhere maybe. If there is a bypass valve then the pipe to the header tank would be in communication with the coolant system whether the thermostat is open or closed. So I guess it wouldn't matter then which side of the thermostat the pipe joined the main system at. If it is upstream of the thermostat, I would imagine there is a positive head of pressure on the pipe by the thermostat due to v restricted flow around the main system. Not sure if this is of any significance.

Don't you wish the system designer could walk in the room at this point and tell it like it is

dandywarhol · Post by **dandywarhol** » Fri Nov 16, 2007 6:58 pm

mikeonb4c wrote:Ah ha! We're getting somewhere maybe. If there is a bypass valve then the pipe to the header tank would be in communication with the coolant system whether the thermostat is open or closed. So I guess it wouldn't matter then which side of the thermostat the pipe joined the main system at. If it is upstream of the thermostat, I would imagine there is a positive head of pressure on the pipe by the thermostat due to v restricted flow around the main system. Not sure if this is of any significance.

Don't you wish the system designer could walk in the room at this point and tell it like it is

Only if he spoke English

Expansion tank

Expansion tank

Header tank

Header tank