This article was first published in September 2001.
The archenemy of any forced induction system - whether it's turbocharging or supercharging - is charge-air heat. Charge-air heat is a by-product of compressing (aka boosting) the engine's intake air to above atmospheric pressure; the introduction of this heat is physically unavoidable, though its severity does vary depending on boost level and compressor efficiency.
What Problems Can Charge-Air Heat Cause?
Excessive charge-air heat can often lead to detonation. Detonation is an unstable and rapidly spiking combustion process that can result in damage to rings, pistons, spark plugs, valves, cylinder heads and head gaskets. Needless to say, the onslaught of detonation is something that must be avoided at all times!
Short of damaging your engine, charge-air heat will at minimum cause a loss of torque - which is not good from a performance point of view. It's important to remember that hot intake air is less dense - this reduces combustion pressure and, therefore, reduces torque output. Furthermore, many electronic engine managed cars are programmed to retard ignition timing and/or enrich the fuel mixtures as intake air temperature increases. Both of these countermeasures are aimed at reducing the risk of detonation - even though they result in further worsened engine response and torque.
So How Can Charge-Air Heat Be Reduced?
Air-to-air and water-to-air intercoolers are the prime strategies used to cool charge-air before it reaches the engine. Air-to-air intercoolers have a heat exchanger core that transfers charge-air heat to the on-coming air stream. Water-to-air intercoolers have a heat exchanger core that transfers charge-air heat to a recirculated water system. These water systems usually comprise an electric pump and control circuit, a reservoir/filler and a water radiator.
In the aftermarket - where chasing power is the Number One priority - it has become common to install the biggest possible modern-design (ie bar-and-plate) air-to-air intercooler to give maximum heat-exchange and induction airflow. These types of intercoolers are certainly effective for increasing potential power and reducing the chance of detonation - but such an installation is often very expensive.
Because of this, many people are resorting to an intercooler water spray to improve the cooling potential of an average air-to-air core. These sprays work on the principle of water evaporating on the core surface and - in doing so - absorbing a portion of charge-air heat. Note that the effectiveness of a water spray is largely determined by the droplet size; a very fine water mist gives the potential to quickly absorb the most heat. (See "Intelligent Intercooler Water Spray - Part 1" for more on water sprays.)
Of course, the benefit of a water spray continues to apply to the aforementioned premium-sized modern-design intercoolers - though, due to their already high effectiveness, there is usually less potential for improvement. Oh, and - despite being largely forgotten - the water radiator of a water-to-air intercooler system can also benefit from a water spray, although because of the thermal mass of this sort of system, you will need to use a lot of spray water.
But There is a Potentially More Effective Option to a Water Spray...
Many dedicated electronics stores sell so-called freeze sprays - an aerosol-style spray can that emits a powerful refrigerant chemical. Freeze spray is designed to rapidly cool electronic components, allow detection of intermittent thermal faults, dry joints and overheating problems. This Servisol-branded freeze spray can be bought for AUS$14.50 through Jaycar Electronics. The 250-gram can uses an environmentally-friendly non-CFC propellant and - like each brand of freeze spray we've come across - is non-flammable. This is very important when you plan to use it in an under-bonnet environment!
Keen to see how effective freeze spray is as an air-to-air intercooler supplement, we organised a chassis dyno session at Adelaide's RPM Performance Centre. Here's how things went...
Our freeze spray testing was conducted on a late-model Toyota Supra that had been converted to a 1JZ-GTE 2.5-litre twin-turbo six with an automatic transmission (a combination found in Japanese-spec Soarers). The engine itself was equipped with a single K&N pod air filter, a 3-inch mandrel exhaust, MicroTech programmable ECU, 15-psi boost and a front-mount tube-and-fin Garrett air-to-air intercooler. This intercooler had been positioned in the main radiator cooling passage, giving easy access for us to hit it with the freeze spray - we could cover most of the intercooler with the spray.
Our testing procedure set out to identify two things - the freeze spray's effect on post-intercooler charge-air temperature, and its accompanying effect on torque (and so power). Charge-air temperature was measured using the MicroTech management system's air temp sensor (which was located in the pipe leading from the intercooler to the throttle body). Power was measured on RPM's in-house Dyno Dynamics chassis dyno.
Let's get to it.
With an ambient temperature of around 13 degrees Celsius, we gave the Supra a succession of power-runs until we'd gained a stop-and-go routine that achieved identical intake temperatures and a stable power curve. In standard form, the twin-turbo Supra (driving through an auto transmission, remember) could repeatedly generate 158kW at the rear tyres. This is shown in the accompanying graph - and all that follow - as the red plot.
Interestingly, post-intercooler air temperatures started off at 37 degrees Celsius (immediately prior to the run) and fell to 35 degrees once under load. This reduction of full-throttle charge-air temperature was indicative of an already effective intercooling arrangement where there was tremendous heat-sink capacity. To further explain things, we'd suggest that the high-velocity post-intercooler induction air (while on boost) didn't get enough time to become affected by under-bonnet radiant heat sources. (The same sort of behaviour can be seen on the road, where cars with good intercoolers will experience a drop in intake air temp when on boost for the first time after a long period of no boost.)
The figures 158kW (at the wheels) and 35 degrees Celsius were established as our 'un-sprayed' baseline.
Next, we set about replicating an environment where somebody might use freeze spray to cool their intercooler immediately prior to making a run at the drags. Certainly, heat soak can be a major problem for drag racing competitors who have to spend extended time queued up in staging lanes; we often see people showering their intercoolers with a hand-held water pump before making their pass.
So, with the Supra prepared for another power run, we doused the intercooler core with freeze spray and let it sit with the engine idling for around 20 seconds - as would happen in real-life at the drags. After being applied, the freeze spray was seen to momentarily condense on the core surface before dissipating. After our 20-second pause was up, the Supra's throttle was again fully opened in order to attain another power graph. And here it is...
Interestingly, the prior dousing of freeze spray had no affect on low-to-mid-range torque - however, there was a small improvement in the very top-end. The pink plot shows that peak power had increased by an indicated 2kW to 160kW (representing a 1-2 percent gain). Initially, this was considered to be simply dyno run variation - but another power-pull with the pre-sprayed core followed by another 'un-sprayed' baseline power run revealed exactly the same power curves.
Certainly, this small 2kW margin did exist.
Furthermore, it was justified when we looked at the post-intercooler charge-air temperatures while under load; these had fallen to 32 degrees Celsius (down from 35). Theory tells us that (as a rule) each 4 degree Celsius reduction of intake air temperature results in a 1 percent increase in peak power - which works out pretty line-ball with what we'd seen.
So why was the freeze spray's improvement seen in the top-end and not throughout the entire rev range? Well, we suspect that this particular Supra was running very rich under load through the lower and middle part of the rev range - as indicated by thin black smoke from its twin tailpipes. This excessively rich mixture (seen only in that part of the rev range) would be responsible for destroying any potential torque gain - a phenomena that we'd recently seen while dyno testing a modified MY01 Subaru Impreza WRX.
Convinced we had seen a definite improvement in the top-end (though extremely small), we now tried another test with the freezing spray; this time we kept spraying the refrigerant onto the core both before and during the run. Note that this process consumed a full can of freeze spray quite rapidly - only two power runs could be performed before having to start with another can. Here is our new dyno graph showing the continuous flow of freeze spray (plotted in blue) versus the 'un-sprayed' baseline (plotted in red)...
As you can see, the accompanying power curve showed near-as-dammit the same peak power increase as previously - 160kW at the wheels (up from 158kW 'un-sprayed'). Full load post-intercooler charge air temperature, meanwhile, was measured at 31 degrees Celsius - down from 32 previously and 35 degrees 'un-sprayed'. As before, these gains were ultra-small - but we repeated them again to be guaranteed they were real.
It appeared that the Jaycar-distributed freeze spray was delivering a j-u-s-t measurable power gain - but was it giving any advantage over a conventional water spray? To draw a comparison, we conducted some more power runs with a plastic hand-held pump spraying water onto the core; as in the previous run, the water spray was applied continuously before and during the dyno run. Note that - once adjusted - the pump nozzle gave a very similar spray pattern and flow volume as we'd seen with the freeze spray.
And here's how traditional water spray fared...
Compared to our 'un-sprayed' baseline, the continuous spray of water perhaps yielded a peak power gain - but it didn't perform as well as freeze spray. As indicated by the green plot, the water spray saw the Supra push out an average 159kW at the back wheels - a 1kW (less the 1 percent) improvement on 'un-sprayed'. This is 1kW down on each test that involved the freeze spray. Post-intercooler charge-air temperatures with the water spray in action were between 32-33 degrees Celsius (compared to 31 degrees with the same continuous flow of freeze spray).
An intercooler spray on a force-inducted vehicle can reduce charge-air temperatures, yielding improved torque with reduced chance of detonation. Freeze spray in particular - from what we have seen here - can give slightly increased gains over a conventional water spray. Also - given that our test Supra already exhibited exceptionally good intercooling and the day was cool anyway - we'd suggest that the gains showed here would be magnified on other vehicles using a more heat-stressed intercooler in hotter weather.
Certainly, if you can measure the post-intercooler charge-air temperature in your force inducted car and you've got a spare AUS$13.77-$14.50, why not give freeze spray a go?
RPM Performance Centre
+61 8 8277 2266
Mounting the Can?
In this story we've only set out to identify whether or not freeze spray had any performance advantage. However, it appears that - with a little ingenuity - you could mount a can of freeze spray in a cool place in your vehicle. Being cylindrical, perhaps the can could be held in place near your intercooler with large diameter hose clamps (making it easy to replace when it runs empty).
Triggering the spray and then transporting the refrigerant from the head of the can to your intercooler core are the tricky bits. One avenue to explore is having a small extension tube from the head of the can to the (relocated) nozzle. This might, at least, get the spray where you want it to go. Next comes the job of pushing the tube down into the can when spray is required. Hmmm. This could, perhaps, be done with an electric central locking solenoid to force the extension tube down - but how it would connect to the tube is your problem!
As we said, a little ingenuity...