Air to Water Intercooler

[Gts*Jason]

I am wondering if anyone is currently running an A2W setup on their turbo celica. I am planning on plumbing one in on my celica in the next couple of weeks when I make the switch over.

[Shizuma]

I am wondering if anyone is currently running an A2W setup on their turbo celica. I am planning on plumbing one in on my celica in the next couple of weeks when I make the switch over.

I made a system for an Exige awhile back that the guy used on the street and mild track days. Just curious: why do you want to go air water? Air to air is more efficient, cheaper, and weighs less than an air to water system (on average.... yes you can dump ice cubes into the reservoir but you can also temporarily "freeze" an A2A intercooler with liquid N and achieve similar results).

Anyway that's neither here nor there, for a core if you can make your own it'll save you some $$ otherwise look at spearco or PWR cores. For a pump, I believe I was using a Bosch unit on the last setup I made, just be sure to keep the lines short and of a large enough diameter that you don't starve the pump.

[ano]

I ll use a/w setup with my rotrex build.
Plan is PWR barrel intercooler 4"x8", bosch water pump and a heat exchanger.

Reason for going this route is:
1. It was perfect for my old plan, the small eaton m45 s/c --> shorter route, more boost, less lug (i hope i am using the right wording here).
2. wanted to keep the the fogs
3. didn't count the cost correctly :gap:

PWR barrels cost a fortune, but if you keep an eye on ebay and on some other car boards (that use them a lot, i.e. mr2 spyder), you ll get them new half price.

[Gts*Jason]

I made a system for an Exige awhile back that the guy used on the street and mild track days. Just curious: why do you want to go air water? Air to air is more efficient, cheaper, and weighs less than an air to water system (on average.... yes you can dump ice cubes into the reservoir but you can also temporarily "freeze" an A2A intercooler with liquid N and achieve similar results).

Anyway that's neither here nor there, for a core if you can make your own it'll save you some $$ otherwise look at spearco or PWR cores. For a pump, I believe I was using a Bosch unit on the last setup I made, just be sure to keep the lines short and of a large enough diameter that you don't starve the pump.

So you havent on a celica.

I ll use a/w setup with my rotrex build.
Plan is PWR barrel intercooler 4"x8", bosch water pump and a heat exchanger.

Reason for going this route is:
1. It was perfect for my old plan, the small eaton m45 s/c --> shorter route, more boost, less lug (i hope i am using the right wording here).
2. wanted to keep the the fogs
3. didn't count the cost correctly :gap:

PWR barrels cost a fortune, but if you keep an eye on ebay and on some other car boards (that use them a lot, i.e. mr2 spyder), you ll get them new half price.

And you havent either but plan to. You are going to try and mount a barrel type? Where in the bay? That just seems a bit large to be mounted easily.

[Shizuma]

So you havent on a celica.



And you havent either but plan to. You are going to try and mount a barrel type? Where in the bay? That just seems a bit large to be mounted easily.

Nope because it's a bit large to be mounted easily
:gap:

[Shizuma]

^ You should probably try asking Jlitman, he was running an A2W with a custom core on his Greddy(?) Celica before he sold it.

[ano]

And you havent either but plan to. You are going to try and mount a barrel type? Where in the bay? That just seems a bit large to be mounted easily.
It will fit after a small cheat. I am removing the stock fans and i m putting two 10" thin ones in place. I have the parts in hand and it seems pretty doable (i believe 1zz has a little more space than the 2zz though)

[evilthorne]

Your going turbo now Jason or something else?

[Gts*Jason]

Your going turbo now Jason or something else?

I have all the stuff to switch over to a turbo setup. The A2W setup seems to be the easiest to setup and plumb without have to change a bunch of other stuff. I have already done an ECU and Battery relocate so there is nothing but space above the transmission. Going to give it a shot and see what it takes to get one installed on a celica. After installing a small A2W on a cobalt the other day the setup has really caught my eye.

[Boosted2.0]

For your car I would totally go A2W. Just relocate teh battery to the trunk and put it there. You get a very direct, very short pipe path for the intake air that way, which is ideal for improved throttle response in road racing.

I haven't ever done one on a Celica, but I would sure like to some day. The tC kit we are planning will most likely be using A2W.

[2KGTrd]

not to get ot here but how did you extend the ecu harness from the engine?

i started doing the relocate yesterday and noticed one side is really easy but the other from the engine poses a problem.

[monkeywrench]

Ait to air is more efficient, less expensive and lighter. Air to water is good when routing does not allow for air-air or for drag racing applications where you want to run ice water. Our LSR car has air to water simply because that car has no radiator opening in the front bumper for aerodynamic reasons. Weight is actually a plus with that car because of the high speeds.

The short piping is nice (I think the LSR car has under 12" of IC piping) but the cost and complexity of the system usually makes air-air a better option. We have a large square type air to water mounted above the transmission.

Both ways work.

[Gts*Jason]

not to get ot here but how did you extend the ecu harness from the engine?

i started doing the relocate yesterday and noticed one side is really easy but the other from the engine poses a problem.

I had to do a rewire a couple years ago bc of a loose fuel line, so I extended the harness a little.

Ait to air is more efficient, less expensive and lighter. Air to water is good when routing does not allow for air-air or for drag racing applications where you want to run ice water. Our LSR car has air to water simply because that car has no radiator opening in the front bumper for aerodynamic reasons. Weight is actually a plus with that car because of the high speeds.

The short piping is nice (I think the LSR car has under 12" of IC piping) but the cost and complexity of the system usually makes air-air a better option. We have a large square type air to water mounted above the transmission.

Both ways work.

Could you elaborate on how A2A is more efficient

[Shizuma]

I had to do a rewire a couple years ago bc of a loose fuel line, so I extended the harness a little.



Could you elaborate on how A2A is more efficient


It comes down to what you're using as a heatsink.

For an A2W intercooler, you're actively passing a fluid media (usually water) through a heatsink to take away thermal energy. Water has a pretty high thermal conductivity and therefore can take in heat energy better than air. However, you're stuck with a finite amount of fluid in the system, and once the fluid has taken in all the heat energy it can hold, that's it. In the meantime you're passively trying to take heat energy back out of the system by running the heated water through a heat exchanger, unfortunately this process is usually not as efficient as transferring heat into the water and therefore once the system is "heatsoaked", it takes it a while to come back down to ambient on it's own (efficiencies of the cooling devices and work done to the fluid by the pump come into play). So, if you're running short bursts (drag racing, standing mile, short track, autox) you won't have generated more heat than the water in the system can "hold" and it will generally provide better cooling than an A2A system for that duration of time. Plus you can always ice it down in the paddock or transfer cooler water into the system (think bypass machine).

For an A2A intercooler, you're basically trading the higher thermal conductivity water provides for an infinite heatsink. The system won't be able to take away heat as fast as an A2W setup but it will always take heat away.

Then other things come into play like MWR mentioned: cost, pipe routing, space limitations, pressure drops through the cores, weight, what to do if you're pump explodes, etc.

[Gravel]

It comes down to what you're using as a heatsink.

For an A2W intercooler, you're actively passing a fluid media (usually water) through a heatsink to take away thermal energy. Water has a pretty high thermal conductivity and therefore can take in heat energy better than air. However, you're stuck with a finite amount of fluid in the system, and once the fluid has taken in all the heat energy it can hold, that's it. In the meantime you're passively trying to take heat energy back out of the system by running the heated water through a heat exchanger, unfortunately this process is usually not as efficient as transferring heat into the water and therefore once the system is "heatsoaked", it takes it a while to come back down to ambient on it's own (efficiencies of the cooling devices and work done to the fluid by the pump come into play). So, if you're running short bursts (drag racing, standing mile, short track, autox) you won't have generated more heat than the water in the system can "hold" and it will generally provide better cooling than an A2A system for that duration of time. Plus you can always ice it down in the paddock or transfer cooler water into the system (think bypass machine).

For an A2A intercooler, you're basically trading the higher thermal conductivity water provides for an infinite heatsink. The system won't be able to take away heat as fast as an A2W setup but it will always take heat away.

Then other things come into play like MWR mentioned: cost, pipe routing, space limitations, pressure drops through the cores, weight, what to do if you're pump explodes, etc.

I'm not sure that's quite right...

Water has a much higher specific heat capacity than air, so it takes about 4,000 more heat to heat up 1L of water by 1C than 1L of air.

The problem in a A2W system is that you're really doing hot-air --> warm-water --> ambient-air via 2 heat exchangers. The warm-water to ambient-air heat exchange happens in your charge-coolers' radiators, and is no less 'infinite' as you like to put it than the A2A's hot-air --> ambient-air exchange.

However, the efficiency of the A2W system is hampered by the fact that heat-exchangers work best with big temperature differences:

In air-air you might have 100C charge --> 30C ambient - Delta-T = 70C
In air-water-air you might have 100C charge --> 50C water --> 30C ambient - Delta T1 = 50C, Delta-T2 = 20C

So in a air-water-air system, you will need really large water rads to efficiently dump heat out of the system - but converesly the temp of your cooling water will be much more stable than say the underbonnet air cooling a Scooby WRX's air-to-air...

We all know air-water-air works - that's how the heat gets out of your cylinders, into the coolant and out of the raidator - and that's a shed load more heat than a charge cooler ever sees ;)

[Shizuma]

I'm not sure that's quite right...

Water has a much higher specific heat capacity than air, so it takes about 4,000 more heat to heat up 1L of water by 1C than 1L of air.

The problem in a A2W system is that you're really doing hot-air --> warm-water --> ambient-air via 2 heat exchangers. The warm-water to ambient-air heat exchange happens in your charge-coolers' radiators, and is no less 'infinite' as you like to put it than the A2A's hot-air --> ambient-air exchange.

However, the efficiency of the A2W system is hampered by the fact that heat-exchangers work best with big temperature differences:

In air-air you might have 100C charge --> 30C ambient - Delta-T = 70C
In air-water-air you might have 100C charge --> 50C water --> 30C ambient - Delta T1 = 50C, Delta-T2 = 20C

So in a air-water-air system, you will need really large water rads to efficiently dump heat out of the system - but converesly the temp of your cooling water will be much more stable than say the underbonnet air cooling a Scooby WRX's air-to-air...

We all know air-water-air works - that's how the heat gets out of your cylinders, into the coolant and out of the raidator - and that's a shed load more heat than a charge cooler ever sees ;)

Makes sense! :gap:

The only two things I don't agree with are:

Water has a much higher specific heat capacity than air, so it takes about 4,000 more heat to heat up 1L of water by 1C than 1L of air.

I could be wrong, but the fluid's ability to transfer heat would seem to be a more important property than its ability to store heat in a cooling application. I'm more interested in how well it takes in and gives off heat as opposed to its ability to store it.

The warm-water to ambient-air heat exchange happens in your charge-coolers' radiators, and is no less 'infinite' as you like to put it than the A2A's hot-air --> ambient-air exchange.

The infinite part I was referring to was the fluid you are removing the heat energy from the charge with (water) is a set amount, and though it's ability to transfer heat (conductivity) and store heat (capacitance) is greater than air's, you are stuck with a relatively small amount of it ( less than a car's cooling system and in terms of the amount of fluid present, much less than the ambient air in an A2A setup).


*EDIT*

You know, as I re-read this it seems you did a better job describing the system's efficiency; mine is more a characteristic of the entire system rather than a function of it. http://1.bp.blogspot.com/_ytU6dUQrIfU/S88S3Abj_lI/AAAAAAAACBI/Kn0QreCOXg0/s1600/curses.jpg

[Gravel]

Makes sense! :gap:

The only two things I don't agree with are:



I could be wrong, but the fluid's ability to transfer heat would seem to be a more important property than its ability to store heat in a cooling application. I'm more interested in how well it takes in and gives off heat as opposed to its ability to store it.

OK, take two iron rods, heat them up to red-hot, then put one in a 1L tin filled with still air at 20C for 10s and the other in a 1L tin filled with still water at 20C for the same 10s. Then chose the one you'd rather grip in your bare hands...

As you say you are not trying to store heat, but take it away. Your heat is removed in an air-air system if and only if you have a flow of air across your intercooler. You are heating up the ambient air to cool the charge air and then moving away from the air you just heated into cooler air as the car moves or fans draw cold air in.

In air-water-air, you are heating water to cool your charge air, and you are pumping that warm water away to the charge-cooler rads where hopefully it will dump its heat out in the same way as you lose heat to the atmosphere in air-air.


The infinite part I was referring to was the fluid you are removing the heat energy from the charge with (water) is a set amount, and though it's ability to transfer heat (conductivity) and store heat (capacitance) is greater than air's, you are stuck with a relatively small amount of it ( less than a car's cooling system and in terms of the amount of fluid present, much less than the ambient air in an A2A setup).

You could have 20L of charge cooler water if you didn't care about the weight issue, but that's missing the point. You are using the water as a heat transfer fluid - to pump the heat from the charge-cooler which is stuck under the bonnet somewhere, to the charge-cooler rads which are in nice cool air somewhere. The final destination for the heat is still the atmosphere - the efficiency of air-water system will still depend on the efficiency of your heat-exchangers - and you have 2 to worry about in air-water-air.

What I would like to know is the duty cycle of various styles of engine use - clearly drag racers produce a shed load of power for a very short percentage of their use, but I'd love to know what percentage of the time Ronin is on more than 30-50% throttle in his canyon runs...

[Shizuma]

OK, take two iron rods, heat them up to red-hot, then put one in a 1L tin filled with still air at 20C for 10s and the other in a 1L tin filled with still water at 20C for the same 10s. Then chose the one you'd rather grip in your bare hands...

This is still just a demonstration of each fluid's conductivity, not its capacitance.

As you say you are not trying to store heat, but take it away. Your heat is removed in an air-air system if and only if you have a flow of air across your intercooler. You are heating up the ambient air to cool the charge air and then moving away from the air you just heated into cooler air as the car moves or fans draw cold air in.

In air-water-air, you are heating water to cool your charge air, and you are pumping that warm water away to the charge-cooler rads where hopefully it will dump its heat out in the same way as you lose heat to the atmosphere in air-air.

I don't disagree but as you say, you will only transfer heat out of the water through the heat exchangers if there's airflow across them. Just because you're pumping the water away from the charge air doesn't mean it's going to take heat with it because it can only do so if the water temp is less than the charge air temp which will only happen if the water is dissipating heat through the charge coolers which will only happen if you're moving. In an A2A system, the car moving is the pump, while in a A2W2A system you have a separate pump, but the system is still limited in cooling capacity by having air move across the heat exchanger.


You could have 20L of charge cooler water if you didn't care about the weight issue, but that's missing the point. You are using the water as a heat transfer fluid - to pump the heat from the charge-cooler which is stuck under the bonnet somewhere, to the charge-cooler rads which are in nice cool air somewhere. The final destination for the heat is still the atmosphere - the efficiency of air-water system will still depend on the efficiency of your heat-exchangers - and you have 2 to worry about in air-water-air.


Yup, but at the end of the day however much charge cooler water you have is only going to be able to take in X units of heat energy. Once that happens, it can't take in anymore: it has a finite amount of heat energy it can transfer. With an A2A there is an infinite amount of charge cooler air so there is no limit on how much heat energy it can transfer.

[monkeywrench]

With air-water you're pulling heat out of your charge air with water. Works great. But now what do you do with that heat? You need to use air to cool the water back down. The water is a middle man. With air-air you skip the middle man. It's the same process with both setups, air-water just has an extra step in the middle.

The water does act as a buffer. It takes a while for the water to heat and during that time an air-water may pull more heat out of your charge air than an air-air could. This is why drag cars do it this way. Being able to start with 32F water is a nice bonus. Now, in a road race application where heat soak is an issue that advantage goes away. Many of the Lotus guys have found that to be the case.

Like I said, both are valid approaches. Use whichever suits your application best.

[Gravel]

This is still just a demonstration of each fluid's conductivity, not its capacitance.

Yeah, it was meant to be - I thought you were doubting water's ability to take in heat, not its specific heat capacity.... :shrugs:


I don't disagree but as you say, you will only transfer heat out of the water through the heat exchangers if there's airflow across them. Just because you're pumping the water away from the charge air doesn't mean it's going to take heat with it because it can only do so if the water temp is less than the charge air temp which will only happen if the water is dissipating heat through the charge coolers which will only happen if you're moving. In an A2A system, the car moving is the pump, while in a A2W2A system you have a separate pump, but the system is still limited in cooling capacity by having air move across the heat exchanger.


Yep, agree with all of that...


Yup, but at the end of the day however much charge cooler water you have is only going to be able to take in X units of heat energy. Once that happens, it can't take in anymore: it has a finite amount of heat energy it can transfer. With an A2A there is an infinite amount of charge cooler air so there is no limit on how much heat energy it can transfer.

I agree but I think you're missing the point - the idea is NOT to warm the water up to the point at which it can't cool the charge air. If you're dumping as much heat out of the water in your rads as you're putting in in the charge cooler, you can transfer heat through the system forever - the idea is still to get the heat from your boosted air to outside the moving car somehow - water is just the intermediate step...

I must have come across wrong - I'm not trying to sell air-water-air as better in any way - it is a lot more complicated to get right but sometimes you just can't run 3" air pipes to the front of a car, or can't live with the lag of doing so...

[Gravel]

With air-water you're pulling heat out of your charge air with water. Works great. But now what do you do with that heat? You need to use air to cool the water back down. The water is a middle man. With air-air you skip the middle man. It's the same process with both setups, air-water just has an extra step in the middle.

The water does act as a buffer. It takes a while for the water to heat and during that time an air-water may pull more heat out of your charge air than an air-air could. This is why drag cars do it this way. Being able to start with 32F water is a nice bonus. Now, in a road race application where heat soak is an issue that advantage goes away. Many of the Lotus guys have found that to be the case.

Like I said, both are valid approaches. Use whichever suits your application best.

Yup - agree with all of that - but why are the Lotus guys heat-soaking? Are their water-circuit rads just undersized for the power they're making? :shrugs:

[Shizuma]

Yeah, it was meant to be - I thought you were doubting water's ability to take in heat, not its specific heat capacity.... :shrugs:


Nope, just that it's capacitance isn't the important property here, conductance is. :gap:


I agree but I think you're missing the point - the idea is NOT to warm the water up to the point at which it can't cool the charge air. If you're dumping as much heat out of the water in your rads as you're putting in in the charge cooler, you can transfer heat through the system forever - the idea is still to get the heat from your boosted air to outside the moving car somehow - water is just the intermediate step...

No I get what you're saying but you'll never be able to move heat OUT of the water as well as you move heat into the water, in fact it will only ever be as efficient as an A2A intercooler at this part.

The first part I'm trying to explain goes like this: Say you're moving bricks (heat) from a pile in your backyard (your charge air) to a pile in your neighbor's front yard (ambient air). Now say you have 10 friends with wheelbarrows (the amount of water cooling your A2W system) and 10,000 friends with buckets (the amount of air cooling my A2A system). Since the process will never be 100% efficient, none of them ever completely empty their wheelbarrow/bucket. So at first, your wheelbarrow guys will be moving more bricks than the bucket guys, but eventually they just won't be able to put many if any bricks in their wheelbarrows because with each trip they make, some of the bricks stay in their wheelbarrow. The same thing happens with the bucket guys, but since there's SO many more of them available, there's always going to be some that can still carry away bricks.

What I'm saying is that by using two steps with imperfect efficiencies, you're going to reach a steady-state at which the water cannot carry away heat as effectively as an A2A setup and you're going to reach this state faster than an A2A setup.

I must have come across wrong - I'm not trying to sell air-water-air as better in any way - it is a lot more complicated to get right but sometimes you just can't run 3" air pipes to the front of a car, or can't live with the lag of doing so...

No, no it's all good we're just arguing thermodyanmics :gap:.

[Gravel]

Nope, just that it's capacitance isn't the important property here, conductance is. :gap:

And that conductance isn't constant - it depends on the heat difference between the fluid and the rod, and as the air will heat up way faster for the same heat input than the water (due to their relative heat capacities), the air's effective thermal resistance (1/conductance) will shoot up and prevent any more conduction really quicky. The air filled tin and iron rod reach a hotter equilibrium than the water one, which is why we'd both choose the rod in water...


No I get what you're saying but you'll never be able to move heat OUT of the water as well as you move heat into the water, in fact it will only ever be as efficient as an A2A intercooler at this part.

Yes, and no :gap: The thermodynamics is symmetrical - but the temperature differences between hot charge air and warm cooling water, and warm cooling water and ambient air will make it harder to cool the water to ambient. I say harder, because if you want to build your entire car out of radiators, you will be able to get very close to ambient...


The first part I'm trying to explain goes like this: Say you're moving bricks (heat) from a pile in your backyard (your charge air) to a pile in your neighbor's front yard (ambient air). Now say you have 10 friends with wheelbarrows (the amount of water cooling your A2W system) and 10,000 friends with buckets (the amount of air cooling my A2A system). Since the process will never be 100% efficient, none of them ever completely empty their wheelbarrow/bucket. So at first, your wheelbarrow guys will be moving more bricks than the bucket guys, but eventually they just won't be able to put many if any bricks in their wheelbarrows because with each trip they make, some of the bricks stay in their wheelbarrow. The same thing happens with the bucket guys, but since there's SO many more of them available, there's always going to be some that can still carry away bricks.


The problem with the analogy is that the fuller the wheelbarrows get, the quicker they get unloaded!


What I'm saying is that by using two steps with imperfect efficiencies, you're going to reach a steady-state at which the water cannot carry away heat as effectively as an A2A setup and you're going to reach this state faster than an A2A setup.


But the temperature of your steady state is key here - if you rads are crap and your charge-coolant reaches equilibrium at 80C, you are in trouble. If they are good and you get to within 10C of ambient - e.g. 30-40C, you are doing well - but quite possibly not as well as an A2A system.

It's all down to packaging IMHO - can you get more efficient cooling from the same volume of water rads, or air-rads (i.e. a2a intercoolers), and can plumbing let you do things with water rads that a2a can't :shrugs:

Anyone for a rear wing made of charge-cooler rads? :gap:


No, no it's all good we're just arguing thermodyanmics :gap:.

That's what you say :)

http://imgs.xkcd.com/comics/duty_calls.png

[Shizuma]

And that conductance isn't constant - it depends on the heat difference between the fluid and the rod, and as the air will heat up way faster for the same heat input than the water (due to their relative heat capacities), the air's effective thermal resistance (1/conductance) will shoot up and prevent any more conduction really quicky. The air filled tin and iron rod reach a hotter equilibrium than the water one, which is why we'd both choose the rod in water...


The problem with this analogy is that you're limiting the amount of air and water in the system to the tin. In an A2A setup, this limit doesn't exist.

The problem with the analogy is that the fuller the wheelbarrows get, the quicker they get unloaded!

But you're still limited by how efficiently you can empty them. Even though they'll dump their wheelbarrows faster they're still left with some increasing amount of bricks X that stay in the wheelbarrow. And it's not as if the rate that they're moving to the front yard to dump them increases either because you're pumping speed is going to remain constant.


It's all down to packaging IMHO - can you get more efficient cooling from the same volume of water rads, or air-rads (i.e. a2a intercoolers), and can plumbing let you do things with water rads that a2a can't :shrugs:

If it didn't work well in some application no-one would use it?:gap:






That's what you say :)

http://imgs.xkcd.com/comics/duty_calls.png

:rofl:

[Shizuma]

:gap:

http://www.youtube.com/watch?v=zvP7Jv8rUK4

[Gravel]

But you're still limited by how efficiently you can empty them. Even though they'll dump their wheelbarrows faster they're still left with some increasing amount of bricks X that stay in the wheelbarrow.

Ah, no - this is key, the number of unemptied bricks rises from 0 to X and then stays at X - this is the equilibrium point. The key is to make X as small as possible, it can never be 0 but it can be close...

If it didn't work well in some application no-one would use it?:gap:

:werd:

[Shizuma]

Ah, no - this is key, the number of unemptied bricks rises from 0 to X and then stays at X - this is the equilibrium point. The key is to make X as small as possible, it can never be 0 but it can be close...


BAH! You are correct, I worded that badly (I've now been up for close to 30 hours :marky:).

[PRRedCelicaGT]

I have an A2W system on my celi...no problems or issues and has been working perfect for 6 years...