Liquid-to-Air instead of Air-to-Air

[ASAP]

Liquid to air is awesome and provides a fantastic amount of cooling for short bursts. It's even better if you ice it. This is why it's preferred for drag racing. The problem is that the water temps can go up very very quickly and once they do, you are done. The air to air is universally preffered for track usage, yes it can heatsoak but as someone already mentioned it will dissipate quickly. As far as lag, size an IC properly with low bend, low travel pipes and you can bring it down tremendously. The non M cars having air to air is one of the reasons I spoke about differentiation.

[accce]

With all the testing on the Ring I think they would have gotten the air to liquid ratio correct and these cars should not heat soak unless air flow is reduced somehow.

[ASAP]

LTA is supposed to be awesome for short sprint drag runs and not so hot for the track or street vs ata.

[Boss330]

Originally Posted by ASAP

LTA is supposed to be awesome for short sprint drag runs and not so hot for the track or street vs ata.

Manage heat soak and you got a better cooling system with liquid to air That's why the S63Tü manages to keep IAT that low IMHO

And, Ferrari are using a liquid to air intercooler in their 2014 F1 car BTW...



Talks about the Ferrari system from 7:30 onwards:

[JoeFromPA]

Sounds like a pretty amazing system then. It all comes down to how much heat it can radiate out vs. how fast it's taking it in, and that's a design function. It's not like Liquid to air (LTA) is inherently inferior in some way, it's all about how quickly it was designed to drop stored heat. It's just more challenged in that regard.

[CanAutM3]

Originally Posted by JoeFromPA

Sounds like a pretty amazing system then. It all comes down to how much heat it can radiate out vs. how fast it's taking it in, and that's a design function. It's not like Liquid to air (LTA) is inherently inferior in some way, it's all about how quickly it was designed to drop stored heat. It's just more challenged in that regard.

An air to water intercooler is more efficient at extracting heat from the intake air, period.

Naturally, it has to be properly designed and sized to avoid heat soak; the same applies to an air to air system. There are drawbacks from an air to water system, such as added weight and complexity, but performance is not one of them.

[ASAP]

Originally Posted by CanAutM3

An air to water intercooler is more efficient at extracting heat from the intake air, period.

Naturally, it has to be properly designed and sized to avoid heat soak; the same applies to an air to air system. There are drawbacks from an air to water system, such as added weight and complexity, but performance is not one of them.

I have now spoken to a propulsion engineer and a few tuners. None of these individuals see the sense of a LTA application for a street car. I guarantee right now that this was done for packaging, differentiation and potentially lag reasons.

[Sedan_Clan]

Originally Posted by ASAP

I have now spoken to a propulsion engineer and a few tuners. None of these individuals see the sense of a LTA application for a street car. I guarantee right now that this was done for packaging, differentiation and potentially lag reasons.

The latter being a significant reason for an liquid-to-air system.

[Boss330]

Originally Posted by ASAP

I have now spoken to a propulsion engineer and a few tuners. None of these individuals see the sense of a LTA application for a street car. I guarantee right now that this was done for packaging, differentiation and potentially lag reasons.

I guess none of these where involved in the design and engineering process of the Bugatti Veyron or McLaren 12C and P1 then?

Nor the 2014 Ferrari F1 car...

Lag is not a "potentially" issue here. You have substantially less volume of air pipes and less distance the compressed air has to travel from compressor to intake with the S55 design. A huge air to air intercooler has a much larger volume that the turbo has to fill than a water to air cooler. Turbo lag is definitely one of the reasons.

Cooling capability of a water to air intercooler is far better than a air to air cooler. There is no argument there.

Heat soak is the water to air intercoolers main issue. If the water that cools the intercooler becomes hot, then it also takes more energy to cool it Down than a air to air intercooler that has heat soaked.

So, heat soak is the thing engineers need to worry about. If they have failed there, then it's a possible bad choice. If they have that under control, water to air is the superior coolant Method

And, Ferrari relies on water to air intercoolers in their current F1 car, so it's doable for racing as well.

[karbonphiber]

Originally Posted by ASAP

I have now spoken to a propulsion engineer and a few tuners. None of these individuals see the sense of a LTA application for a street car. I guarantee right now that this was done for packaging, differentiation and potentially lag reasons.

You're speaking to the wrong type of engineer and tuners. A 5th grade student will know that the specific heat capacity of water is greater than air

Why do you think Underground Racing Twin Turbo Lambos producing 1750HP have been engineered with an air-to-water cooling system...

With air-to-air there are less moving parts to circulate the heat rejection medium which makes it more reliable, but less efficient in pre-cooling intake air.

[Netwon]

The beauty of the M5 system (and others like it is) is that the radiator fans and intercooler/turbo (they're water cooled) cooling pumps (x2) can remain on when you turn the car off removing heat... Something that you cannot easily do in an Air to Air environment. Thus preventing heat soak when parking in hot environments after spirited driving.

[aus]

Originally Posted by ASAP

I have now spoken to a propulsion engineer and a few tuners. None of these individuals see the sense of a LTA application for a street car. I guarantee right now that this was done for packaging, differentiation and potentially lag reasons.

That's the main reason since BMW was trying everything to prevent lag.

Originally Posted by karbonphiber

You're speaking to the wrong type of engineer and tuners. A 5th grade student will know that the specific heat capacity of water is greater than air

Why do you think Underground Racing Twin Turbo Lambos producing 1750HP have been engineered with an air-to-water cooling system...

With air-to-air there are less moving parts to circulate the heat rejection medium which makes it more reliable, but less efficient in pre-cooling intake air.

I'm sure packaging had a LOT to do with using a water to air intercoller iin that car. They make SICK systems.

.

[aus]

Originally Posted by Netwon

The beauty of the M5 system (and others like it is) is that the radiator fans and intercooler/turbo (they're water cooled) cooling pumps (x2) can remain on when you turn the car off removing heat... Something that you cannot easily do in an Air to Air environment. Thus preventing heat soak when parking in hot environments after spirited driving.

I think you're comparing two different things. The intercooler cools the compressed intake air, where you're talking about cooling the turbos themselves.??

.

[turbo8765]

Originally Posted by CanAutM3

You need a lot of "air heat" to heat soak a water system...

Like 431hp for a 20-30min session?

Water has a good ability to absorb heat. The problem is that's once it has absorbed too much heat it's equally hard to dissipate.

If you spend much time at the track you know that an air/air system is easier to manage.

BMW chose air/water for throttle response… bad choice.

[turbo8765]

Originally Posted by Netwon

The beauty of the M5 system (and others like it is) is that the radiator fans and intercooler/turbo (they're water cooled) cooling pumps (x2) can remain on when you turn the car off removing heat... Something that you cannot easily do in an Air to Air environment. Thus preventing heat soak when parking in hot environments after spirited driving.

Not at all unique to an air/water system.

It's all the same to keep the electric fans running to pull air through an air/air heat exchanger.

The fact that the M5 needs this system is telling.

[CanAutM3]

Originally Posted by turbo8765

Like 431hp for a 20-30min session?

Water has a good ability to absorb heat. The problem is that's once it has absorbed too much heat it's equally hard to dissipate.

If you spend much time at the track you know that an air/air system is easier to manage.

BMW chose air/water for throttle response… bad choice.

The bold portion is line with what I posted earlier in this thread. It all has to do with how well the system is sized and balanced. If the system has sufficient heat extraction capacity, heat soak will not be an issue, be it for 5 minute, 20 minute or 2 hour sessions. I am fairly confident BMW did their homework correctly in that respect. Overall, liquid-to-air is a superior system, the drawback is that it is more complex. From a performance point of view, air-to-air is the compromized system, but for some application it is simply "good enough", so there is no need for the additional complexity.

And yes, throttle response is a big reason for going liquid-to-air.

Heat exchange is dependent on 4 important parameters, specific heat, mass, temperature difference and time. To produce more power, a greater mass of air is needed for the engine. To produce more power with a given displacement, more boost is needed, hence more temperature in the intake charge. So relative to an N55 engine, the S55 needs ~40% more air mass flow that is also hotter due the higher boost; a massive air-to-air intercooler would have been needed on the S55 to slow down the greater and hotter air flow to cool it down sufficiently. The huge volume of air stored in that big intercooler would have killed throttle response.

An air-to-air intercooler performance is totally dependant on the ambient air temperature and the speed of the vehicle (mass flow), so it is not possible to have a good control over the intake charge temperature. With a liquid-to-air system, it is possible to control the intake charge temperature by varying the water flow (with the speed of the water pump or through a thermostat) in the system depending on operating conditions.

[OpenFlash]

Originally Posted by Boss330

Bugatti Veyron uses air to liquid intercoolers...

So does McLaren in their cars...

It will be interesting to see how efficient BMW has managed to deal with heat soak and track driving, but if they can cope with that, air to liquid is the best way to cool the charge air

Liquid to air intercoolers are pretty common in mid engine cars since air ducting (for air-to-air ICs) can be difficult. The Porsche manages well because it has fat rear fenders which create a high pressure zone at the dual air inlets and vents in the rear bumper are placed at a low pressure zone to help with cooling air flow through the rear mounted air/air ICs.

The choice to go with liquid to air in the new M cars is probably influenced by the need to keep engine's modular since BMW is getting pretty engine swap happy between different models. Same reason DMEs are not bolted on to the engine and not tucked away somewhere else in the engine bay. It saves a ton of wiring and allows engines to drop in easily from one model to another without much complication.

Going with Liquid to air intercoolers in the M cars also do a fair share to improve boost response. BMW went to great lengths to make the new turbo M engines feel as close to NA as possible. Not requiring a big front mounted IC in the noise of the car also pays dividends when it comes to weight distribution and polar moment. They are trying to keep as much weight between the wheels instead of hanging on either end of the car.

Liquid-to-air ICs do have better cooling characteristics than air-to-air since water does a much better job of cooling the air than air. Assuming BMW sized the water reservoir generously and is running a beefy enough water pump, it should remain very heat-soak resistant. If not, that is all easy stuff to rectify in the aftermarket

Cheers,
shiv

[CanAutM3]

Originally Posted by shiv@vishnu

The choice to go with liquid to air in the new M cars is probably influenced by the need to keep engine's modular since BMW is getting pretty engine swap happy between different models. Same reason DMEs are not bolted on to the engine and not tucked away somewhere else in the engine bay. It saves a ton of wiring and allows engines to drop in easily from one model to another without much complication.

Not sure I agree with this statement since most BMW AG engines use air-to-air. Unless you believe AG models will also be going for liquid-to-air in the future.

Originally Posted by shiv@vishnu

Not requiring a big front mounted IC in the noise of the car also pays dividends when it comes to weight distribution and polar moment. They are trying to keep as much weight between the wheels instead of hanging on either end of the car.

What about having a radiator filled with heavy water hanging forward of the front axle? I am not convinced weight reduction was a driver for going liquid-to-air, quite the opposite actually.

Originally Posted by shiv@vishnu

Liquid-to-air ICs do have better cooling characteristics than air-to-air since water does a much better job of cooling the air than air. Assuming BMW sized the water reservoir generously and is running a beefy enough water pump, it should remain very heat-soak resistant. If not, that is all easy stuff to rectify in the aftermarket

Mostly agree with you here. But wouldn't you think that a properly sized radiator is the most important element for providing resistance to heat soak?

[OpenFlash]

Originally Posted by CanAutM3

Not sure I agree with this statement since most BMW AG engines use air-to-air. Unless you believe AG models will also be going for liquid-to-air in the future.



What about having a radiator filled with heavy water hanging forward of the front axle? I am not convinced weight reduction was a driver for going liquid-to-air, quite the opposite actually.



Mostly agree with you here. But wouldn't you think that a properly sized radiator is the most important element for providing resistance to heat soak?

Heat exchangers don't need to be very big. They are much smaller and lighter than comparable air/air ICs. Even when filled with water. Their small size also means that they can be fitted in smaller frontal space areas without interfering with radiator cooling which is also a drawback to running large air/air FMICs in high output application.

At the end of the day, a good liquid/air IC design is a balancing act between the heat exchanger, pump and intercooler core. BMW tends to engineer these things very well. So who knows? I'm eager to see what design decisions they made

[CanAutM3]

Originally Posted by shiv@vishnu

Heat exchangers don't need to be very big. They are much smaller and lighter than comparable air/air ICs. Even when filled with water. Their small size also means that they can be fitted in smaller frontal space areas without interfering with radiator cooling which is also a drawback to running large air/air FMICs in high output application.

You might be right, I have never actually compared the weights. The radiator of the intercooler system does seem fairly large on the F8X though (see picture below, my guess is that it is the big radiator that sits in front of the main engine cooling radiator, the horizontal one being the oil cooler and the smaller unit to the right being the DCT oil cooler).

Originally Posted by shiv@vishnu

At the end of the day, a good liquid/air IC design is a balancing act between the heat exchanger, pump and intercooler core. BMW tends to engineer these things very well. So who knows? I'm eager to see what design decisions they made

Agreed . I can't wait to try it out a the track .

[turbo8765]

Originally Posted by CanAutM3

If the system has sufficient heat extraction capacity, heat soak will not be an issue, be it for 5 minute, 20 minute or 2 hour sessions. I am fairly confident BMW did their homework correctly in that respect.

An air-to-air intercooler performance is totally dependant on the ambient air temperature and the speed of the vehicle (mass flow), so it is not possible to have a good control over the intake charge temperature. With a liquid-to-air system, it is possible to control the intake charge temperature by varying the water flow (with the speed of the water pump or through a thermostat) in the system depending on operating conditions.

My prediction is the system works well with an intermediate driver on stock tires. I'll be surprised (pleasantly) if the system is adequate during the summer in the SE (RA, RRR, CMP, Barber, Sebring, Daytona etc) with an advanced driver on sticky tires. With any power mods in the summer on sticky tires I predict reduced performance whether by reduced boost, retarded ignition timing etc.

"control over intake charge temp" is not really necessary. The goal is to get as close as possible to ambient in all cases. You might save a little juice by decreasing electric pump speed in an air/water system when demand is reduced but that's more minimizing the downside than it is upside.

[turbo8765]

Originally Posted by CanAutM3

You might be right, I have never actually compared the weights. The radiator of the intercooler system does seem fairly large on the F8X though (see picture below, my guess is that it is the big radiator that sits in front of the main engine cooling radiator, the horizontal one being the oil cooler and the smaller unit to the right being the DCT oil cooler).



Agreed . I can't wait to try it out a the track .

My first thought looking at look at the myriad heat exchangers... how many hours labor to change a lower rad hose?