Intercooling on a race car. Air to air, or air to water?

[Chris Wilson]

I have asked five people, and had five different responses, so I am now asking a wider audience. I have also asked on another race car forum, and had conflicting ideas mooted. I THINK I am about to pass on air to water, but would welcome further input. If I go air to air could people suggest placement of such an exchanger. Thanks.

I have a sports GT race car (Zeus) that currently has a Toyota 1600 4A-GE Formula Atlantic engine in it, making about 210 BHP. I am turbo charging it with the hope of seeing about 350 to 380 BHP at the flywheel. The turbo is a Garrett GT3071R, with integral wastegate. The management will be Motec M800. The current pair of twin choke Weber 48DCOE carb are removed and a plenum with a single 65 mm throttle body made. I am dropping the CR to about 8.5 to 1 and fitting milder cams, of about 270 to 280 degrees duration.

My indecision rests on how to intercool it. Anyone with nothing better to do is welcome to look at some photos of the car I link below. I won't muddy the water with my own, or currently suggested ideas, except to say I think I need to keep the current water radiator set up, I don't think I can get away with less. I don't mind scoops on the rear body work, but don't want any front of the car mounted heat exchanger.

I will consider air to water, but ONLY if I can get information from people with first hand experience of this of this running a similar power output from a small engine, in circuit RACE trim. My fear is an air to water will work fine on the road, or even track days, but may heat soak in full long distance circuit race usage.

Photos of the ongoing project at http://www.gatesgarth.com/intercool...tercooling.html


Thanks for looking.

[tristancliffe]

Also consider that the water in a water:air intercooler isn't very light, which will hurt braking and cornering. Whether it hurts pure acceleration depends on whether it cools more than the water and so allows sufficient extra power to overcome that weight. I would doubt it though.

If the water is cooled efficiently then there is no reason why it should heat soak, but to get rid of the same amount of heat you'll need a similar amount of cooling air going over its radiator, so you will also suffer similar drag issues.

I can't see any benefits of water:air intercooling myself in any application really.

[Chris Wilson]

The link's not working for some people. Let me try again:

http://www.gatesgarth.com/intercooli...ercooling.html

[GORDON STREETER]

The guys next to my workshop had a water intercooler made for a Nissan Pulsar that was fitted on top of the engine with the water rad front mounted .
It was OK on the open road, but in a close racing situation I don't think it will make an awful lot of difference for the money that it cost.
OK if you can get out in front though !

[CNHSS1]

air to air for race applications, chargecooling (water/air) is great as a 'packaging solution' for road cars or stop/start, low speed vehicles, but given the extra weight of the additional rad(s) waterpump, not forgetting the water itself, its difficult to justify in a racer.
the other worry, although very unlikley, is if the charge cooler fractures internally, the water leaking into the inlet will hydraulic the motor with disasterous results!

[CNHSS1]

heat soak at idle/paddock speeds shouldnt be a problem if the charge coolers 'slave' rad is fitted with fans to bring the temps down. the problem could be your Zeus alternator and battery are hardly likely to be huge, so with the C/cooler fans, water pump and fan for the water rad (if fitted) it puts a fair strain on the charging system. most ecus (especially it seems aftermarket rather than OEM) are very voltage sensitive at low engine speeds and can cause grief.
dont underestimate the extra heat that turbocharging will put into the engine water system either (through producing more power), you may need to look at the engine rad size or core thickness. im assuming the Gt3071 you have has a water cooled core? if so it may be better to not use the water cooling for the core but leave it open to the air. the water jacket is really there to save the bearings from lazy drivers who turn the motor off hot and carbonise the oil, but given that the engine will be warmed up and shut down properly, itll help to keep extra heat out of the engine rad system by not hooking the core up in the first place.

[ian_w]

An air/water system has an advantage that the intercooler is very compact and can therefore be fitted close to the engine - sometimes they are built into the plenum itself. This means that the total volume between the turbo and the throttle will be less than with an air/air intercooler, this should have a noticable effect on the transient response of the engine. The lower volume means the boost pressure will build faster giving a better throttle response.

If you have a mid-engined car then it gives you the option of having a front mounted charge cooler radiator - which may help cooling and possibly weight distribution. With an air/air you will have to duct air either from the roof or side pod areas.

[1200Datto27]

Originally Posted by CNHSS1

dont underestimate the extra heat that turbocharging will put into the engine water system either (through producing more power), you may need to look at the engine rad size or core thickness. im assuming the Gt3071 you have has a water cooled core? if so it may be better to not use the water cooling for the core but leave it open to the air. the water jacket is really there to save the bearings from lazy drivers who turn the motor off hot and carbonise the oil, but given that the engine will be warmed up and shut down properly, itll help to keep extra heat out of the engine rad system by not hooking the core up in the first place.

The downside of not running the water through the turbo's cooling jacket is that then you would have to increase the volume of oil being flowed through the bearing core to prevent it from being over heated, this then results in the oil pump needing to be upgraded to maintain adequate flow and pressure, as well as extra cooling for the oil system.

[Chris Wilson]

Garrett are totally adamant that water cooled cores are NOT run dry. So I will have to run the water to and fro the core, with the associated rise in water temperatures. assuming i keep the 2 stock side mounted water rads this won't be an issue, the car was built for racing in a very hot environment. I may be able to jettison one water rad and fit an air to air in there, but it's risky, I feel. I may get advice as to whether an I/C 1/2 the height or length of one of the rads would be enough, and try a 1/2 size water rad paired with it in one side duct. Thanks for the answers so far. A front mounted I/C isn't an option though. Cheers.

[CNHSS1]

you might be right re the GT series as they are ball bearing cores (not used them as the cores arent rebuildable). traditional oil bearing cores are normally run without the water cooling connected in race applications, although im sure Garrett would have a blue fit if they saw what people do with their products
a GT3071 is a big old beast on a 1600cc motor though, although should fairly easily hit the target bhp figures. being a pure race-bred chassis, the resulting narrow power band should be ok, i suspect the current N/A motor is somewhat cammy anyway.
a split height/length IC/water rad could do the trick, is there room to increase the core depth as well? mounting an IC infront of the water rad is a possibilty too, although does tend to raise the water rad temps, but not as horrendously as youd think
im sure the oil pump system isnt stock anyway (dry sump i assume), so should be made to cope with tweaks

[Notso Swift]

No question you should go air to air, if your charge cooling is compemised by packaging and air flow than you should consider suplementary cooling via chemical intercooling (ie. water injection)
Mind you, I think a larger model in the GT28 series would be a better bet for your solution, the 30 is a pretty big frame, great when you go well over 400, but that isn't going to happen here

[phoenix]

I don't see why this engine shouldn't produce over 400 bhp.

2.4 bar gauge with good intercooling will get around 465bhp at 8200 rpm by my estimate. The turbo is too big lower down the rev range - I don't think you would get much over 2.2 bar at 5000 rpm, giving maximum torque of about 340 ft/lb. Still a useful 320ish bhp at that engine speed.

But if you only want 380 bhp, 1.7 bar should do it - so the turbo has much more capability than you require and a smaller housing would have spool up quicker and be less laggy.