2014 intercooling

[ringo]

Some people don't get it. Water to air has no advantage on an F1 car.

These are the heat flow paths:

Air to Air
Compressor air --> metal----> side pod air. Done.

Water to Air
Compressor air---> metal----> water-----> metal -----> air. Done.


If that's not enough, then I can do a sample calculation if pushed. I'm feeling lazy, but if the discussion goes further, it can be demonstrated with heat transfer equations.
The point is that air is at both ends of both methods. So your heat transfer is limited by the airs ability to conduct the heat from the metal and to give heat to the metal. You could even have magical gypsy tears in the radiators, the air will only take heat away from it at it's own limitation and nothing more.


The advantage the air to water has is that in the case of a road car where you have a lot volumes that are not in contact with air directly you can place your intercooler there for low drag reasons and packaging. The other advantage is that if you have a large water reservoir, it will absorb the heat from the air for a while before it would actually need to be cooled by outside air. the larger the reservoir the longer it can absorb heat without needing to be cooled itself.
Now when that time is up, you got a problem.

Looking on the F1 car now, it really has no advantage of water to air. I doubt they would want to keep a large reservoir of water in the car. And I doubt they would use engine water to cool compressor air. And for 300km, the reservoir size would eventually be a moot point as the water would eventually get as hot as the air.
The other concern is the need for additional radiators and a water pump, and a water tank. It doesn't add up if ultimately there is no advantage thermally or geometrically.

[Ra8]

So you're saying no one uses water-air cooler?

[ringo]

I don't know for sure, i'm just going off what seems to make more sense.
I just don't see the reason for that system.
If they do, it must be done in some unconventional way or for a particular reason.

As it appears, Ferrari doesn't have the smallest sidepods, neither does Mercedes.

Scarbs put a picture of what seemed to be a air to water intercooler. And it's possible that the team must have placed a huge water radiator in the side pod for that particular intercooler.
They made space for that water intercooler above the engine, then they still had to go back to using a radiator in the sidepods. So I'm not seeing the packaging advantage if the car will have more components.

[basti313]

Some people don't get it. Water to air has no advantage on an F1 car.

You are absolutely right. =D>

If that's not enough, then I can do a sample calculation if pushed. I'm feeling lazy, but if the discussion goes further, it can be demonstrated with heat transfer equations.

If you look at the equations the water solution is only as good as the air solution if you have "perfect" water flow. In practice the water flow is limited, so you can not even reach the cooling performance of the air to air solution with the same cooling area.
Air-water intercoolers like in the new M3 have a much lager area due to this fact...as you I can not see any advantages for F1.

The advantage the air to water has is that in the case of a road car where you have a lot volumes that are not in contact with air directly you can place your intercooler there for low drag reasons and packaging. The other advantage is that if you have a large water reservoir, it will absorb the heat from the air for a while before it would actually need to be cooled by outside air. the larger the reservoir the longer it can absorb heat without needing to be cooled itself.
Now when that time is up, you got a problem.

The main benefits for me are the shorter air tubes. A road car benefits a lot in turbo reaction time from them, but in an F1 with a turbo that is kept on speed with the MGU H there is no need for short tubes.

[aleks_ader]

The main benefits for me are the shorter air tubes. A road car benefits a lot in turbo reaction time from them, but in an F1 with a turbo that is kept on speed with the MGU H there is no need for short tubes.

I dont agree with that. Now in times of fuel formula every saved joule of energy counts. Beacuse the turbo lag is smaller or shorter we dont need so much turbine angular acceleration (lees electric energy needed for same engine performance or if we turn that around could maker even better and higher torque curve). Mercedes could use that saved energy in other parts of circuit.

I think that could make the engine operation window much wider. That benefit overall car drivebility. Engineers will have so much more options to optimize the car on exact driver, track demands.

So Merc engineers conclude that the mass penalty of additional heat exchanger (+ longer shaft etc.) will be manageable and it could be negligible compared gains in other areas.

That i think is main reason behind that decision.

[ringo]

It can't be negligible. How heavy is 2 heat exchangers full of water?
The ask yourself, how about the water pump and the water tank?
Those are additional weight and energy consumption and also more potential for reliability failures. Not to mention the car would have more crammed packaging.
There's no tangible advantage. A shorter intake run is infact the negligible point, the kers eliminates any response issues. In fact the shorter run of pipe doesn't affect response time as much as we'd like to think.

What would be interesting is if there was both air and water cooling. With a radiator somewhere in the region below the driver above the splitter.

[aleks_ader]

It can't be negligible. How heavy is 2 heat exchangers full of water?
The ask yourself, how about the water pump and the water tank?
Those are additional weight and energy consumption and also more potential for reliability failures. Not to mention the car would have more crammed packaging.
There's no tangible advantage. A shorter intake run is infact the negligible point, the kers eliminates any response issues. In fact the shorter run of pipe doesn't affect response time as much as we'd like to think.

What would be interesting is if there was both air and water cooling. With a radiator somewhere in the region below the driver above the splitter.

So you say that Ferrari,probably Merc and Lotus got that wrong? It most be compromise of perfomance in other areas for sure. I dont think that team goes in season with know deficit.

But i give you right that additional complexity of systems make nightmare to optimize. And make even more chances to broke something...

[ringo]

Those teams you mentioned having water to air is all speculation. There's no existing proof. So i'm not saying they got it wrong.
I'm just saying what makes sense thermodynamically.
Renault power plant seems to be air to air, from what redbull have shown. Mercedes is a mystery as we haven't seen the car with all it's heat exchangers, and neither have we seen what's in the left side pod.

I do have a suspicion however, that if they are using water to air, it may not be water at all. It may be gasoline to air.
Where the heat exchanger is that bulk in the side of the tank, and the gasoline is being preheated by the air, and vice versa for the air. 100kg of fuel is a good heat absorber for compressor air... but that's an extreme idea.
The power unit itself and why it has it's advantages are still a mystery.

[flyboy2160]

Water cooling can be an advantage if the component to be cooled is hopelessly buried, thus making the required amount of cooling airflow impossible or requiring an increase in the overall packaging. As already stated, it's of no thermodynamic advantage and comes with a weight and complexity penalty. But sometimes you're stuck in a corner and can only get relatively small diameter coolant pipes routed. Huh? Does that sound like an F1 sidepod or what.........

[flyboy2160]

...
I do have a suspicion however, that if they are using water to air, it may not be water at all. It may be gasoline to air.
Where the heat exchanger is that bulk in the side of the tank, and the gasoline is being preheated by the air, and vice versa for the air. 100kg of fuel is a good heat absorber for compressor air... but that's an extreme idea...

Not extreme at all. It's done in high altitude aircraft - sometimes also to warm the fuel! But I can't see it in an F1 car. Air to air would be way simpler and seems like the clear winner if you can route the air.

[ringo]

yep,
though water to air has these plusses:-

smaller hose diameter. instead of big charge air piping.

lower pressure drop, and friction losses through the shorter charge air piping.

flexibility in packaging, though not necessarily smaller.

isolation of vibrations from the radiator to the charge piping.

flexibility in shaping the radiator.

However, for raw performance and efficiency and packaging, especially when space is very critical. Air to air seems to have edged it IMO.

The use of a log manifold also indicates that something is probably beside the engine as well.

[aleks_ader]

Those teams you mentioned having water to air is all speculation. There's no existing proof.

For Ferrari is pretty 100% that they use it (pictures already exists). For rest of grid i sad in my comments "probably" so i newer stated that is a fact.

Anyway that there are some + and minuses and that teams research pretty well so we could trust them.

But on end of the day we need some simulation. How water-air intercooler effect on turbo lag and on mass penalty (must be towards 20 kg+ pump included) of the vehicle. From there we could make raw conclusion. Without that we just speculate.

[miguelalvesreis]

yep,
though water to air has these plusses:-

smaller hose diameter. instead of big charge air piping.

lower pressure drop, and friction losses through the shorter charge air piping.

flexibility in packaging, though not necessarily smaller.

isolation of vibrations from the radiator to the charge piping.

flexibility in shaping the radiator.

However, for raw performance and efficiency and packaging, especially when space is very critical. Air to air seems to have edged it IMO.

The use of a log manifold also indicates that something is probably beside the engine as well.

Considering the metal on those equations as with equal thermal properties on both, you can elliminate it from both. In that case you only have to consider the heat exchange properties of the changing variable. You are considering water but, as already stated, is hardly water. Is a cooling fluid with extended capacity of heat exchange. It's faster exchanging heat from air and to air then doing it directly from air to air. In this last case the efficiency of the system is always dependent of the external temperature for any given surface area, in other words, depends on the delta of temperatures. With the coolant, you may dump the weight of that variable, making the system efficiency less sensitive from it.
Furthermore, you can better position the 2nd radiator, on a higher flow area of the sidepod, with not only thinner pipes but, longer and twister pipes, without loosing pressure (it's a closed system).
It allows to ``isolate`` the air from the compressor from external variables.
Might be important since we know that the optimal thermal window for this systems is very narrow.

At least, to me, seems to be a viable solution even with the non negligeable weight increase.

Just my 2 cents

Regards

[ringo]

[quote="miguelalvesreis]

Considering the metal on those equations as with equal thermal properties on both, you can elliminate it from both.

No you shouldn't. One has double thickness, remember one uses an extra radiator, that's two metal layers.

In that case you only have to consider the heat exchange properties of the changing variable. You are considering water but, as already stated, is hardly water. Is a cooling fluid with extended capacity of heat exchange.

There is no liquid that exists at 100*C at the moment other than water that has better thermal properties.

It's faster exchanging heat from air and to air then doing it directly from air to air.

It's not.
To illustrate using a funny example, if you are playing catch with a person with 2 arms, a normal person, and the aim is to catch the tennis balls from them and put them in a basket. would your rate of catches and putting the balls in the basket increase if you put a 6 armed person between the both you?
Maybe the 6 armed person can catch 6 balls without passing any at all to you. As soon as his hands are full he'll start passing them to you. Maybe a third two arm person can pass them to him before he passes them to you.

But at the end of the day you are the only one who can put the tennis balls in the basket, and no matter what, you are limited by your number of hands.

Now if you want a faster rate, you probably should play the game with 3 six armed persons or at least 2 to at least have a quicker rate of ball release, even though you will still be the "bottle neck" in the system because of your 2 arms.

So you see, with looking at the 2 armed people as air and the six armed person as water. The limitation of thermal exchange is the two substances at the ends, Air and Air.
The water has the advantage in being able to soak up more heat quickly from an increased mass of air, (as shown if the 6 armed person catches from 2 normal people) however the water has no control over how fast the air can take heat from it in the side pod.

In this last case the efficiency of the system is always dependent of the external temperature for any given surface area, in other words, depends on the delta of temperatures. With the coolant, you may dump the weight of that variable, making the system efficiency less sensitive from it.

I'm not understanding.

Furthermore, you can better position the 2nd radiator, on a higher flow area of the sidepod, with not only thinner pipes but, longer and twister pipes, without loosing pressure (it's a closed system).

Yes reduced pressure drop of the charge air is an advantage. You are basically saying that the water can deal with the pressure drops since it separate and not relevant to the boost. I agree. However it still has to exchange with the limited air in the sidepod. And all those twists and turns have an effect on your water pump you will be using. More power losses.

It allows to ``isolate`` the air from the compressor from external variables.
Might be important since we know that the optimal thermal window for this systems is very narrow.

Now that you have said that..
What can be done is if the water pump for the water to air system is controlled so that the charge air temperature is manipulated by controlling the flow of cooling water that it exchanges with. However that all depends on how much times in the race will the sidepod air flow vary. I don't think it changes much for extended periods of time over a lap.
Track to track I see a reason for it. The engine will be less sensitive to the environment with the water pump being controlled.

At least, to me, seems to be a viable solution even with the non negligeable weight increase.

Well it depends a lot on the how big those effects are.

Here is an image of BMW M5 water to air system, notice the network of hoses and pump to control temperature:



There are two auxiliary water pumps (not shown) for the water-air intercoolers. I believe they are electronically controlled.
Maybe Mercedes could be doing this, who knows. If it's legal it's another issue. But as I said it's really weight, reliability, complexity and packaging and energy efficiency where this system loses out to an air-air.

Drivability, flexibility, and control are advantages, but again it all goes back to the magnitude of these effects.

[basti313]

Considering the metal on those equations as with equal thermal properties on both, you can elliminate it from both. In that case you only have to consider the heat exchange properties of the changing variable.

No, you can not. You could do that if both metals would have the same thickness, but with air-water you have double thickness. But the main thing is thermal flux from one medium to the other: In the case of air-air you have a heat flux from air to metal and metal to air. With air-water you have air to metal, metal to coolant, coolant to metal and metal to air and you have to calculate the heat flux for every boundary layer.

You are considering water but, as already stated, is hardly water. Is a cooling fluid with extended capacity of heat exchange.

What is "extended capacity of heat exchange" in technical terms? For me it comes down to heat capacity as we have a temperature hopefully far below 100°C. So for me it comes down to water with some additives...not much, you do not want to destroy hydrogen bonds.

It's faster exchanging heat from air and to air then doing it directly from air to air.

No. Lock at my first section in this post. As long as the temperature of the coolant is not either the temperature of the air or the compressed air and as long as heat flux from metal to coolant is not infinite, you will always end up with lower overall heat flow.

It allows to ``isolate`` the air from the compressor from external variables.
Might be important since we know that the optimal thermal window for this systems is very narrow.

This may be a benefit if you really have to route the air guides round hot surfaces.