Cooling system caps - what pressure rating?
Discussion
I run my S3 with the pressure cap on the expansion tank + plain cap on the swirl tank. This works fine and does not cause any problems with coolant loss or overheating.
I have compared this set up with other S3's and they all have been the same. Interesting that your car returmed after servicing was also set up this way.
If you look at the pipework in the car and the various connections, this must be the way it was intended.
Putting the pressure cap on the swirl tank, just serves to isolate the all pipework between it and the expansion tank. I think someone long long ago, in a time now forgotten (Star Wars fan) did this and it cured their "problems". When his problems were probably leaks and splits in the isolated pipes.
I have compared this set up with other S3's and they all have been the same. Interesting that your car returmed after servicing was also set up this way.
If you look at the pipework in the car and the various connections, this must be the way it was intended.
Putting the pressure cap on the swirl tank, just serves to isolate the all pipework between it and the expansion tank. I think someone long long ago, in a time now forgotten (Star Wars fan) did this and it cured their "problems". When his problems were probably leaks and splits in the isolated pipes.
quote:
If you look at the pipework in the car and the various connections, this must be the way it was intended.
Putting the pressure cap on the swirl tank, just serves to isolate the all pipework between it and the expansion tank.
That's all it does, but isolating that overflow tank is a *good* thing. Since the overflow tank doesn't have water circulating through it, it is unheated and the air in there will stay pretty cold. This means it won't build up pressure until the system starts to gas. While the system is unpressurised it has less ability to apply cooling to hot spots. Normally this style of tank is used as a swirl tank (with water coming in a high speed at the top and leaving from the bottom at low speed) and this heats the air volume, pressurising the system earlier. The tank on our TVRs is *only* an overflow catch tank.
Bottom line is with a pressure cap on either tank (but not both!) it will work, but it works best with the pressure cap on the swirl pot. To be honest I suspect there are quite a few mechanics out there who don't really know which way round it should be or why. And I don't think everyone at the factory fully understands it either, if the 'V8 bypass' affair is anything to go by.
So put the caps on whichever way round makes you happy, but there's no doubt in my mind that pressure cap *should* go on the swirl tank.
Cheers,
Peter Humphries (and a green V8S)
If I put the pressure cap on the overflow and the standard cap on the swirl tank and run the car up to temp then it pushes all the coolant out of the overflow pipe in the expansion tank and onto the floor as there is nothing to maintain the pressure in the system and as it all expands it looks for a way to escape.
Peter is definitely right. If however the expansion tank on the S3 has not got an overflow then the pressure would be maintained, thus allowing this configuration of caps....
Peter is definitely right. If however the expansion tank on the S3 has not got an overflow then the pressure would be maintained, thus allowing this configuration of caps....
quote:
If I put the pressure cap on the overflow and the standard cap on the swirl tank and run the car up to temp then it pushes all the coolant out of the overflow pipe in the expansion tank and onto the floor as there is nothing to maintain the pressure in the system and as it all expands it looks for a way to escape.
Peter is definitely right. If however the expansion tank on the S3 has not got an overflow then the pressure would be maintained, thus allowing this configuration of caps....
Read "overflow PIPE", a small omission but a critical one !
All pressure caps have this return usually except for the ones that don't. 
I looked into this really big time for the S bibles and found that pressure cap on the swirl tank is the way the vast majority of cars will work correctly. There are a few out there that either were built differently or were replumbed that will work either way. If you car is happy one way then fine leave it. If it is losing water try swapping them over.
Like many things TVR, nothing is black or white.
Steve
www.tvrbooks.co.uk
I looked into this really big time for the S bibles and found that pressure cap on the swirl tank is the way the vast majority of cars will work correctly. There are a few out there that either were built differently or were replumbed that will work either way. If you car is happy one way then fine leave it. If it is losing water try swapping them over.
Like many things TVR, nothing is black or white.
Steve
www.tvrbooks.co.uk
quote:
Interested in replies, but does anyone have the sprung cap with the return valve? I have looked at mine and it looks bog standard. My car runs fine with the pressure cap on the expansion tank so I'll leave well alone for now
cheers John
Is the return valve not the brass round thingimebob under the bottom rubber seal? Pull it and see - I think!
quote:
Buy a radiator cap from Halfords, would I hell! I revealed the whole sworded story about what happened to my V8S last Autumn on this site. Beware, Halford 15lb pressure rated caps are inaccurate. When I reported the matter to them they didnt give a monkeys ....!Go to a dealer.
PS I did find the right 15 lb cap as Halfords for £3.50...I think the Halfords part number is 601.. I saw they also did a 13lb cap the right size as well
well, went to 3, yes 3 dealers yesterday.....and none had either of the caps! gotta order em......having been a former Porsche owner, and still love the marque with a passion, its funny how both differ in their sales and parts dept. Like most TVR parts dept look like my grandfathers shed! (only i can find anything there!)...did make me laugh, even the wife had a chuckle. But did finally get to see the Tamora production car at last.....am i the only one to like the rear end? oh well roll on next week when i can finally get the caps!
Just to continue the discussion – a few further thoughts. Laws of Physics come up with two issues in normal circumstances
Liquids cannot be compressed – gases can
Both liquids and gases expand/contract on temperature changes
If the factory (and bible) fitting is followed with pressure cap on swirl tank, then very early in warm-up, water expansion works against the pressure cap, forcing it open to overflow into expansion tank. (Check on cold engine with full swirl tank and both caps fitted – starts to increase level in expansion tank quite soon after starting the engine) This continues until max temp is reached and then, unless there is a rubber seal on upper flange of pressure cap (Halfords 601 does have one – my original did not) starts to weep from top of pressure cap. Yes, I did follow Bible point of filing swirl tank top to ensure level neck.
On cooling, any system will find it easier to suck in air through the return valve if the system is not air-tight, (again the rubber seal on upper flange is necessary) causing the need to check levels in both tanks as well indicated in the Bible. Both warming and cooling have no effect on the air in the expansion tank which can freely exit through the overflow – along with water if too full.
If the reverse fitting (pressure cap on expansion tank) is followed, then the warming water in the system and swirl tank has freedom to expand apart from the air in the pressurized expansion tank. The air is compressed by the incoming water and also tries to expand as the water heats it up (I do not want to come back on this earth as expansion tank air!). At 15lbs pressure, air is forced out through the pressure seal if necessary and, again, if overfull, followed by water.
On cooling, the pressurized air assists in ensuring that the expansion tank water is forced back into the swirl tank. This is, of course, what happens in a ‘normal’ car as Steve points out and, therefore, only the expansion tank needs to be checked for water level.
So, given that the 2.9i engine fitted in ford cars does not have a swirl tank (nor does it seem to have an overflow from the expansion tank) is the issue to do with the swirl tank, the function of which I know sod all about?
Liquids cannot be compressed – gases can
Both liquids and gases expand/contract on temperature changes
If the factory (and bible) fitting is followed with pressure cap on swirl tank, then very early in warm-up, water expansion works against the pressure cap, forcing it open to overflow into expansion tank. (Check on cold engine with full swirl tank and both caps fitted – starts to increase level in expansion tank quite soon after starting the engine) This continues until max temp is reached and then, unless there is a rubber seal on upper flange of pressure cap (Halfords 601 does have one – my original did not) starts to weep from top of pressure cap. Yes, I did follow Bible point of filing swirl tank top to ensure level neck.
On cooling, any system will find it easier to suck in air through the return valve if the system is not air-tight, (again the rubber seal on upper flange is necessary) causing the need to check levels in both tanks as well indicated in the Bible. Both warming and cooling have no effect on the air in the expansion tank which can freely exit through the overflow – along with water if too full.
If the reverse fitting (pressure cap on expansion tank) is followed, then the warming water in the system and swirl tank has freedom to expand apart from the air in the pressurized expansion tank. The air is compressed by the incoming water and also tries to expand as the water heats it up (I do not want to come back on this earth as expansion tank air!). At 15lbs pressure, air is forced out through the pressure seal if necessary and, again, if overfull, followed by water.
On cooling, the pressurized air assists in ensuring that the expansion tank water is forced back into the swirl tank. This is, of course, what happens in a ‘normal’ car as Steve points out and, therefore, only the expansion tank needs to be checked for water level.
So, given that the 2.9i engine fitted in ford cars does not have a swirl tank (nor does it seem to have an overflow from the expansion tank) is the issue to do with the swirl tank, the function of which I know sod all about?
quote:
So, given that the 2.9i engine fitted in ford cars does not have a swirl tank (nor does it seem to have an overflow from the expansion tank) is the issue to do with the swirl tank, the function of which I know sod all about?
You have most of it already, but there are a couple of points that I think will make it clearer how this works.
The first point is that the cooling system continually generates gas bubbles as a result of porus castings, leaking gaskets, chemical reactions and so on. These bubbles are normally carried around in the water until they are separated out. The cooling system has to be designed to separate and blow out this gas without blowing out water.
The second is that just as water and gas expand when heated, the metal of the engine also expands, and water pipes expand under pressure. The result is that the cooling system *takes in* water as it heats up and pressurises, rather than expanding and blowing water out as you might think. The cooling system has to be designed to self-pressurise as soon as possible to prevent nucleate boiling at hot spots.
Conventional production engines are designed with a swirl tank that neatly solves both these problems.
It is typically a plastic tank with pipes in and out, and a pressure cap on top. Depending how the engine bay is laid out it may be a strange shape with swan necks and so on so it may not be obvious what it is, but if you spot a tank with a pressure cap on, and two water pipes, chances are this is the swirl tank. They often look very similar to the plastic overflow tank used on TVRs but don't be fooled, the plastic tanks in our TVRs don't have water circulating through them and are only overflow catch tanks.
The first job of the swirl tank is to hold an air pocket. As the engine warms up, the hot water heats this air pocket, the air expands and displaces water into the rest of the system and pressurises it. The second job of the swirl tank is to continuously de-gas the cooling water by separating out small gas bubbles from the water passing through it. The water enters as a high velocity tangential jet at the top of the tank, flows down the tank and leaves at low velocity at the bottom. The tangential entry produces an axial swirl within the tank which tends to centrefuge the bubbles out to a column in the middle of the tank. The relatively low exit velocity prevents these bubbles from being dragged down and out along with the water.
The result of the separation process is a small pocket of air at the top of the swirl tank which is continually fed by the small bubbles extracted from the water. Since the pressure cap is at the top of this tank, as the system pressure reaches the pressure limit of the cap and it starts to weep, it is the separated air which is blown out. Since the separation process is continuous and gradual, air is normally blown out at a very low rate.
If your cooling system is routinely blowing out water, either it is not well designed, or it is not in good condition. For example, you may have a leaking pressure cap or the pressure caps the wrong way round. The systems are fairly resilient and will cope with many faults, but only up to a point.
Incidentally, current technology pressure caps and thermostats do not work particularly well. The thermostats induce thermal cycles of under/over cooling, and the pressure caps don't like being unscrewed. Remember there are two seals, the inner seal blows off under pressure and the outer seal is fixed. The more often they are removed the more likely the outer seal is to weep under pressure. The design life is just 1 (one) cycle. If you take the pressure cap off every week 'just to check' you will soon need a new pressure cap.
There are now far better designs of thermostat and pressure cap available, invented by an engineer at a British company
, unfortunately some of the IPR is now held by a Bavarian company 
. But if you keep an eye on the surviving UK industry you might see the 'next generation' coming out soon. 
Cheers,
Peter Humphries (and a green V8S)
quote:
air is normally blown out at a very low rate.
My Rover 220 GSi recently blew it's head gasket. The gasket failed between a cylinder and a waterway, the expansion tank expoded with the extra pressure now leaking from the compresion of the piston.
I can laugh about it now, but it's been a real pain to fix all the mess caused by this.
I have done some research on the swirl tank pressure cap ratings to try to find the right cap here in Sweden.
After converting between units I have found that a pressure of 15 lb/square inch (psi) would be approximately 1.0 bar.
Please correct me if I'm wrong.
Pressure caps are rated in bar here in Sweden (and perhaps in the rest of Europe?).
I have bought a new cap now and hope that it will stop my S3C from leaking at the swirl tank. The new cap looked like space technology stuff compared to the old one, so it should definitely be better.
Johan
>> Edited by Johan G on Tuesday 28th January 08:03
After converting between units I have found that a pressure of 15 lb/square inch (psi) would be approximately 1.0 bar.
Please correct me if I'm wrong.
Pressure caps are rated in bar here in Sweden (and perhaps in the rest of Europe?).
I have bought a new cap now and hope that it will stop my S3C from leaking at the swirl tank. The new cap looked like space technology stuff compared to the old one, so it should definitely be better.
Johan
>> Edited by Johan G on Tuesday 28th January 08:03
Bit late for my say, but hey. The pressure cap goes on teh swirl tank, no question in my mind. The manual says so, and all other cars I have owned have been this way. There are lots of good reasons, the most obvious of which is why have a catch tank otherwise? All cooling systems need an air space and will make one, this should be high up, thus the catch tank is above the swirl tank. The water doesn't circulate to the catch tank, only allows the movement of water relative to volume changes. Hot water takes more space than cold, (even given the metalwork expanding), thus if you overfill your tank, it will be less full after you run the car as the expanded water has been ejected via the pressure cap.
In saying this, if it works don't worry about it.
Nic
In saying this, if it works don't worry about it.
Nic
Gassing Station | S Series | Top of Page | What's New | My Stuff