Coolant pressure at M30 engines

[deQuincey]

knowing that there is a cap in the expansion tank rated at 1 Bar we can assume that coolant pressure must be under 1Bar

but have you found any specifications on this topic ?

what is the correct cooland pressure ?

does it change with rpms ?

i will appreciate your experience on this topic

regards

 

[Mike Goble]

There is no single coolant pressure. It varies around the system, with the highest point being the outlet of the pump, and the lowest being the inlet of the pump. If the pressure exceeds the relief valve in the cap, it vents. One drawback of older-style down flow radiator systems was that the pressure relief was on the high-pressure side of the radiator, resulting in coolant being vented at high pump speeds.

 

[Stevehose]

There is no single coolant pressure. It varies around the system, with the highest point being the outlet of the pump, and the lowest being the inlet of the pump. If the pressure exceeds the relief valve in the cap, it vents. One drawback of older-style down flow radiator systems was that the pressure relief was on the high-pressure side of the radiator, resulting in coolant being vented at high pump speeds.

What about when the engine is not on, and the pressure is built up in the system so caution is needed when taking the cap off, isn't this the pressure DeQ is referring to or is it the same thing?

 

[Mike Goble]

What about when the engine is not on, and the pressure is built up in the system so caution is needed when taking the cap off, isn't this the pressure DeQ is referring to or is it the same thing?

You should always be cautious when removing the radiator cap on a warm engine. With no flow, the pressure is the same throughout the system. Even though the radiator may be just warm, there can be hot spots in the coolant that will boil if you release the pressure quickly.
When the pressure cap is on the coolant temperature can exceed the atmospheric boiling point. Once the pressure cap removed, the coolant will boil if the temperature is above the atmospheric point. It can boil all at once, resulting in a dangerous geyser erupting from the system.

 

[deQuincey]

What about when the engine is not on, and the pressure is built up in the system so caution is needed when taking the cap off, isn't this the pressure DeQ is referring to or is it the same thing?

thank you
i refer to the pressure in the system during engine performance

If we assume that pressure varies through the system, and if we consider that the expansion tank is sort of control point

but my question is still open

which is the pressure that a healty engine shows ? maybe to focus in one point, say in the expansion tank

between 0 until 1 bar for sure, but should it be 0,5 0,8 ?
does it vary with rpms ?

thank you

 

[Mike Goble]

thank you
i refer to the pressure in the system during engine performance

If we assume that pressure varies through the system, and if we consider that the expansion tank is sort of control point

but my question is still open

which is the pressure that a healty engine shows ? maybe to focus in one point, say in the expansion tank

between 0 until 1 bar for sure, but should it be 0,5 0,8 ?
does it vary with rpms ?

thank you

You can fix the expansion tank pressure at 0 bar by loosening the cap. It doesn't really affect anything until the temperature exceeds the boiling point of the coolant. I had a leak in my expansion tank right at the solder line and it would blow the coolant out until the level of the coolant was lower than the leak, and then it would operate just fine. I would check the lever every week or so and it stayed consistently at that level.
I will install a pressure gauge on my new expansion tank today and report back on the readings under various conditions.

 

[deQuincey]

Elaborating my ideas: TO CHECK HERE FOR MISTAKES....

The cooling system is a closed system that has an expansion tank where a quantity of air accumulates.
This tank has a cap calibrated to 1 bar of pressure
the meassured pressure is AIR pressure.
The objective is to have a system in which the internal pressure is higher than the atmospheric pressure and thus ensure that the boiling temperature of the coolant increases above 100ºC and in this way the liquid does not boil when passing through the walls of the cylinders (which in fact are the hottest point in the engine for the coolant) and can dissipate heat from them.

If the boiling temperature increases with the pressure, it seems reasonable to think that the system will try to get as close as possible to the maximum pressure of 1 Bar, since at that pressure the boiling temperature will be higher.

It is understood that the pump provides a pressure to the coolant, but that this is negligible compared to the pressure due to temperature. I myself have verified that by increasing the rpms the general pressure only rises 0.1 or 0.2 bar.

all of this is right ?
any mistakes on my side ?

 

[Stevehose]

I would replace the word "refrigerant" with "cooling" and "coolant" to avoid confusion with the air conditioning "refrigerant" namely R134, R12 etc.

 

[Arde]

I do not think the expansion tank is to regulate the pressure to a constant value. If it were, one would consistently lose coolant that escapes when open or lose air when air escapes. If so one would need to add coolant over time, or the cap would need to open to let air in when negative pressure in the expansion tank is excessive as it cools off.

I view the cap as a mechanism to avoid catastrophically high pressure that can blow up hoses, and thus cascade into faster overheating through coolant loss. The cap valve does not prevent overheating, for that we have temp gauge, light, and shutting down the engine.

If the above is correct, the engine is meant to operate at pressures well below the 2 Bar so that it never has to open in the absence of a fault.

My E10 had no cap, which means that pressure was indeed limited to 1 Bar, but I had to keep adding coolant every couple of weeks...

 

[eriknetherlands]

Interesting question @deQuincey ; I do not have the answer.

I do like to brawl about things , and dump my thoughts anywhere however

(Note that the 1.0 bar purge pressure is actually 2.0 bar atmosferic pressure; it is 1.0 bar OVER atmosferic pressure. )

I would think the system would operate during normal driving condition at a near zero state of pressure for the following reason: I would consider the system to be closed; closed as there is no addition of air or coolant during driving. There are thus only two factors that can create pressure: 1 thermal expansion of the media: coolant and the air in the tank, and 2. gas formation.
- The coolant expansion is a linear thing, and even from freezing point (0) to 99 degree Celsius the volume chance effect is small: (99*the factor of 0.00021 = 0,02 thus 5 liters will become 5,1 Liters (gallons) volume increase (Wikipedia: for every 1 degree Celsius increase in temperature, water will expand by 0.00021 times its original volume)
If the pressure in the air volume in the tank is ambient (1bar) at the begin, and the airvolume was 1 liter, then the 0.1 liter of fluid expension will increase the pressure by + 10% going to 1.1 bar.
- To the second point; Gas formation. IF it occurs, it increases the volume by a factor of ~1600, so that would indeed seriously build pressure. But I doubt that gas formation occurs during normal operation, as the system tries to limit the max temp to 95 or 97 Celsius, doesn't it? (correct me if i'm wrong, isn't that what the temperature sender does; open the cooling circuit to push the temp down? ) If that is correct, then gas will not form throughout the system. This leaves gas formation close to the cylinder wall: for this I think the speed at which the coolant passes through the cylinder head is too high for the coolant to heat up beyond the boiling point.

But for this last thought i have little argumentation to defend it, it's just a thought.

 

[mulberryworks]

Plain water under 2 Bar pressure (atmospheric +1) will boil at 120°C. Adding coolant will raise that temperature some depending on the mix and the product. A 90° thermostat should open at the design temperature, but the capacity of the cooling system and load on the engine will determine what the ultimate temperature will be. Certainly, it will be hottest next to the combustion chambers and boiling there would be bad as the cooling ability will fall off quickly and damage could occur.
When overheating happens the coolant cap should release excess pressure so that damage to hoses and radiators doesn't happen. The engine could be in danger of damage then and should be shut down. I warped the head in my Volvo when I lost coolant because of a failed head gasket and the engine got quite hot.
I agree with the posts above that state engine coolant would normally be under more or less atmospheric pressure, ignoring the pressure of the water pump pushing it around and through the engine and radiator.

 

[eriknetherlands]

A further consideration: We always talk about that the pump builds pressure, but remind that the pump also actively pulls in the fluid on the the other side.
I see no reason to not call it a "hot-water-removal-suction-device".

it's another argument to think that the pump actually does not build up much pressure during normal operation; it's there to just overcoming the friction created for the needed relocation of the hot water. Directly after the pump is the highest pressure. Just in front of the pump we should see the lowest pressure of the system.
Would the pressure in the expansion tank be roughly halfway between those 2 levels?

but, I am violating my own adage here: measure first, then comment.

 

[Arde]

but, I am violating my own adage here: measure first, then comment.

Here are my measurements:

1) E10 had a water leak on an engine block plug. The temperature needle would start going above normal, if I stopped and waited idling it would not drop very fast, if I kept driving it would go up faster, if I kept driving but downshifted the temperature would drop.I did about 90km of highway in a lower gear just to get it safely home and repaired.

That means that when the water volume was not enough to cool the engine, moving the same volume faster was enough to cool it. This is unrelated to water pressure and fully related to water speed, and shows that speed is proportional to revs. That must mean that pressure after the pump is proportional to revs because pressure delta is what provides motion speed.

If the water path was obstructed somewhere then most of the pressure delta would accumulate around the obstruction. I think Erik is right that the pressure delta would be an increase on one side and a decrease on the other as the pump sucks...

2) Again driving 1977 E24 for 5 highway hours low coolant level caused higher temperature and the auxiliary fan kicked loudly in resulting in lower temperature immediately. In this case it means that the air speed at the radiator matters a lot even for a lower water volume. That car had auxiliary fan and oil cooler. It never overheated. The main conclusion is that a 77 BMW dies from rust before overheating can kill it.

3) Volvo 240 with 1/4 Million miles. I think the expansion tank cap was not working, in retrospect. The expansion tank would visibly expand and contract probably due to head gasket leaking engine pressure into coolant. The water pump leaked. I replaced the water pump to fix the leak and the car blew head gasket...The bad water pump was acting as pressure valve...

 

[Beaudave]

knowing that there is a cap in the expansion tank rated at 1 Bar we can assume that coolant pressure must be under 1Bar

but have you found any specifications on this topic ?

what is the correct cooland pressure ?

does it change with rpms ?

i will appreciate your experience on this topic

regards

Hi there! My experience is that the cap is not perfectly leak-proof. The rubber washer that is supposed to do the sealing seems to leak a tiny bit. If not liquid, at least pressurized steam. I drove about 1000 km last weekend, and lost about 2 or 3 cm in the reservoir. Just filled it up again... but noticed my liquid is full of rust. So I need to bleed that circuit. Curiously, I noticed that when filling again the level, the heating inside the car was working better... I don't know it there is an electric resistance that does the heating, of if air goes through a specific radiator containing the coolant. Another thing I need to check.

 

[deQuincey]

Hi there! My experience is that the cap is not perfectly leak-proof. The rubber washer that is supposed to do the sealing seems to leak a tiny bit. If not liquid, at least pressurized steam. I drove about 1000 km last weekend, and lost about 2 or 3 cm in the reservoir. Just filled it up again... but noticed my liquid is full of rust. So I need to bleed that circuit. Curiously, I noticed that when filling again the level, the heating inside the car was working better... I don't know it there is an electric resistance that does the heating, of if air goes through a specific radiator containing the coolant. Another thing I need to check.

IMHO,
lots of things here
your cap or any other point in the system has a problem !
no way losing 2cm coolant in the reservoir
rubber seal might be too hard after years of service, or defective, or broken
when you compare cold engine situation before and after, level MUST be the same, else there is a leak
I have a long experience in checking this before starting engine, absolutely no level drop, not 1mm in 6000km


liquid with rust particles is usual, rust deposits coming from block, must remove block bolt, and if possible use a fresh water hose on top to move the debris away,

sometimes you will need to unclog the blockage inside the bolt

no electric resistance for heating, there is a small radiator inside the heater box

 

[eriknetherlands]

@Beaudave,
I think along the lines of @deQuincey as well.
Loosing 2-3 cm of cooling fluid must mean that it leaks somewhere; as a fluid in the engine compartment (hoses, couplings, block caps etc), or (worst case ) it may disappear internally into the combustion zone (bad gasket, cracked head): do you get a bit sugary smell in the exhaust? I hope not.

Yes, it could leak from the cap, but only until the level drops below the level of the rubber that you think is leaking. Once the fluid level drops below the rubber gasket, the only thing that can leak is just air +possibly some vapors that escape. But you will not loose 2 cm of fluid level as vapor, as that would show immediately; 2cm of fluid volume is many, many cubic feet of dense vapor. (1 block of water makes up ~2600 same sized blocks of vapor - image how bad that looks like!)

Anyway, I think the fluid level is never up soo high that it could spill through a cracked rubber cap gasket. So I would bet on another route for your cooling fluid loss.
I hope it's just a bad hose or hose clamp. Other options soon start to be big bucks and mucho work.

 

[Honolulu]

It's just physics.

1. Coolant system pressure is always the same throughout the cooling system. It is a continuous system filled with an incompressible fluid thus equally pressurized throughout.
2. Cooling system pressure will vary with temperature up to the release setting of the cap. There are tables online which show the relationship of coolant concentration and boiling point.
3. An overheated system may continue to boil if the pressure after release still exceeds the boiling point of the coolant at the cap release pressure.

For the nit-pickers, the pressure at the bottom of the system will be very slightly higher because of the "height" of the coolant above it; this amount is negligible for all intents and purposes. More for the nit-pickers - water is not actually incompressible but for the purposes of coolant, the slight compressibility of water is neglected.

The pressure rating on the cap (if functioning correctly and not loose) will limit the maximum pressure by releasing coolant when the rating is exceeded.

A (normally very small) amount of coolant can leak past the seal on the cap, when you realize that the coolant in the system (particularly at the cap where there is a small vapor space) can be turbulent, wetting the gasket and very slowly escaping.

I've gotten away with driving with my radiator cap on loosely back when I had a small leak in the radiator and needed to conserve fluid.

I went through the coolant cap setting bizness last week with my E46 convertible which had been inexplicably loosing coolant over the two years of ownership. I came to the realization that the expansion tank cap is supposed to have two o-rings, but they were missing. Further investigation showed that the stock E46 cap had a release pressure of 2.0 bar, while most previous BMW models used 1.4 bar. I obtained a 1.4 bar cap with o-rings from a friend and pitched the 2.0 bar cap in the scrap heap. Back story, the M54 engine has an all aluminum block and head with a plastic expansion tank, and we know well the bad rep of BMW with regard to use of plastic, particularly in cooling system. Combined with a old, brittle plastic expansion tank and a coolant temp gage whose electronics are fudged so that the gage reads "normal" until it's too late, there are anecdotally a lot of former E46 owners whose engines overheated, blew the expansion tank and almost instantly lost all coolant, all while the gage reads "normal".

The coding for the E46 gage can be revised to more accurately display the actual coolant temperature but my local wizards refuse to do that for me, and I don't have INPA, ISTA-P or similar to revise it myself.

 

[ScottAndrews]

Elaborating my ideas: TO CHECK HERE FOR MISTAKES....

The cooling system is a closed system that has an expansion tank where a quantity of air accumulates.
This tank has a cap calibrated to 1 bar of pressure
the meassured pressure is AIR pressure.
The objective is to have a system in which the internal pressure is higher than the atmospheric pressure and thus ensure that the boiling temperature of the coolant increases above 100ºC and in this way the liquid does not boil when passing through the walls of the cylinders (which in fact are the hottest point in the engine for the coolant) and can dissipate heat from them.

If the boiling temperature increases with the pressure, it seems reasonable to think that the system will try to get as close as possible to the maximum pressure of 1 Bar, since at that pressure the boiling temperature will be higher.

It is understood that the pump provides a pressure to the coolant, but that this is negligible compared to the pressure due to temperature. I myself have verified that by increasing the rpms the general pressure only rises 0.1 or 0.2 bar.

all of this is right ?
any mistakes on my side ?

Your thought process is basically correct. However here are a few fine points.

The boiling point does increase as the pressure in the system increases. Since there is a relief valve at 1 Bar, then we know that is the upper limit of pressure. The cooling system is always working to reduce the temperature of the coolant, while the engine is continuously raising it. At some point the system reaches an equilibrium point, where the radiator (and any other components in the system that are radiating heat) radiates as much heat as the engine is generating. Presumably this equilibrium point is at a temperature below the 1 Bar boiling temp.

If the heat input from the engine rises above the heat dissipation capacity of the cooling system, then the temperature will rise, and, because it is a closed system, the pressure will rise...at least up to the point where the pressure exceeds the relief valve pressure. Once that point it reached, the temp does not really rise (as long as there is an infinite supply of coolant), because the heat is dissipated as steam.

We can assume that since each cylinder consumes the same volume of air and fuel for each piston cycle, it also generates the same amount of heat. However, since the engine speed varies, the RATE of heat generation will rise as the engine speed rises. At the same time, the water pump speed also increases, and the air flow through the radiator also increases. So these factors act in opposition to the increased heat generated by the engine as the engine speed increases.

In a perfectly designed system, one would assume that these factors are all balanced, so the heat transfer from the radiator (in terms of BTU per second, or Joule/sec) would go up at the same rate as the heat generation (in BTU/sec or Joule/sec) would also go up. In this perfect system the coolant pump would move the coolant at a rate that carried away the heat from the engine at the same rate it was generated. This perfect system would be designed so that the steady state temperature of the coolant was constant, and remained right at a temeperature where the vapor pressure was below 1 Bar.

So any changes in pressure in the system (below 1 Bar) are going to be based primarily on the temperature variation in the coolant

 

[Arde]

How about the chances of cavitation of the M30 coolant?

 

[Beaudave]

@Beaudave,
I think along the lines of @deQuincey as well.
Loosing 2-3 cm of cooling fluid must mean that it leaks somewhere; as a fluid in the engine compartment (hoses, couplings, block caps etc), or (worst case ) it may disappear internally into the combustion zone (bad gasket, cracked head): do you get a bit sugary smell in the exhaust? I hope not.

Yes, it could leak from the cap, but only until the level drops below the level of the rubber that you think is leaking. Once the fluid level drops below the rubber gasket, the only thing that can leak is just air +possibly some vapors that escape. But you will not loose 2 cm of fluid level as vapor, as that would show immediately; 2cm of fluid volume is many, many cubic feet of dense vapor. (1 block of water makes up ~2600 same sized blocks of vapor - image how bad that looks like!)

Anyway, I think the fluid level is never up soo high that it could spill through a cracked rubber cap gasket. So I would bet on another route for your cooling fluid loss.
I hope it's just a bad hose or hose clamp. Other options soon start to be big bucks and mucho work.

Hi Erik. I don't totally agree with you. Without falling into rocket scientist calculations, I am assuming (I may be wrong of course), that I am loosing 2-3 cm of vaporized fluid thrugh the cap, simply because I don't lose any on small distances, even on a daily base. But last weekend I drove 450 km straight; twice... And it wouldn't be easy to notice, at 120-130km/hr... that steam is coming out of the hood... But not big deal... I just now need to check that I won't be loosing any further.
Because indeed, if it's leaking inside the cylinders... yes, that's another story. I hope not though. I don't have a sugary smell on the exhaust fumes, although in the cold and humid mornings, it does fumes a lot, until engine reaches a certain temperature... But I guess we all experience this same situation. And it does not smell Chanel N. 5... (as my sweet wife says)
Last argument pro that potential cap leakage: my mechanic told me many times to buy a new cap for that reservoir. When I asked him why, he simply answered, with his nasa rocket science : "Because when you restore old cars, it's always better to do so".