Turbo cooling circuit and track-day upgrades

What one owner found about the standard circuit

• The turbo gets little coolant with the thermostat closed. With an 82°C thermostat and a warm engine after track driving, he saw no water in the return line from the turbo unless he revved the engine a lot. His conclusion is that the turbo only sees flow if it self-circulates.
• The electric auxiliary pump does not cool the turbo. Many (including another owner in the thread, and a 2007 post) assumed it runs water through the turbo after shutdown. After tracing all the hoses, he found it only circulates water from the cylinder head to the heater matrix, removing the worst heat soak from the head. He concluded that shutting down with a hot turbo isn't good, although the water will self-circulate a little from the heat. Other owners hadn't checked this themselves.
• Slow to cool. After a track session, even with a larger aluminium radiator and twin Spal fans, his temperature took cool-down laps to fall from 105°C to 95°C and wouldn't drop at idle in the pits unless held at about 2,000 rpm. Another owner reports the same: still 105°C on tickover in the pits unless revved. Revving to 1,500 to 2,000 rpm helps a bit.
• His explanation: the standard pump isn't meant to push coolant round the whole car, with about 8 m of tubing and many bends, so the pressure drop is large. That is his reasoning, not a measurement.
• The charge cooler radiator sits in front of the main radiator, which two owners say blocks the main radiator, reducing airflow and raising the temperature of the air reaching it. Another owner reported second-hand that a car without the pre-radiator ran 91°C on hot track days.
• The oil-water cooler passes the oil heat into the coolant. One owner was told by a cooling engineer at his work that it transfers about 20 to 25 kW against about 80 kW for the front radiator on a standard car. Treat these as unchecked figures.

Upgrades owners tried

Electric water pump

• An owner fitted a Davies Craig EWP80 (not the larger EWP150) in the front under tray, by the anti-roll bar on the cold side, as the maker recommends. He reused the hose and drilled a new hole. He kept the belt-driven pump too, since he wouldn't trust the electric pump enough to remove it. It is a volute (turbine) type that doesn't restrict flow much, and he runs it only at low rpm, high temperature and after shutdown.
• Another owner pointed out that an electric pump to help at low revs, in traffic or in the pits is common on tuned Elises, and lets you circulate water after the engine is off. Another Turbo owner ran a bigger electric pump in the same place, which helped, but he also cut the boot and added vents at the same time, so it wasn't conclusive.
• One-way valve. The first test failed: the pump raised the pressure in the big bottom hose of the expansion tank enough to reverse the flow in the small hose that goes to the thermostat, pushing all the air from the tank back into circulation. A fuel one-way valve in that hose cured it. Air must be bled out first; he did that by moving the car around on a slope with the engine running and no cap on.
• Control: he drove pump and fan from two modules with PWM tables by temperature and rpm (fan at 20 to 40% duty is quiet). The modules have a built-in soft start, so there is no big current draw when they switch on.
• Results (his): at idle at over 90°C on a warm summer day, turning the fans to full had little effect, but raising pump speed brought the temperature down to a steady 85°C straight away. After shutdown the temperature fell instead of rising towards 105°C, and he could see circulation in the turbo return hose at about 70% pump duty. On the road he can idle indefinitely without reaching 90°C. At a trackday in August (about 17°C) the highest temperature was 102°C, falling as soon as he eased off. He called the results not yet conclusive.
• Concerns: he doubts the Davies Craig pump's durability, and doesn't like the maker's controller cycling the pump on and off, which he thinks would cause thermal cycling in the head. His own plan was to move to an OEM Volvo electric pump from Pierburg (about 150 l/min; one used part cost about £200 in Sweden, part number 31368715). That was a plan, and no result was posted.

Airflow and radiators

• One owner added a neoprene ramp in the crash box to push cold air through the radiators, and a phenolic spacer between the charge cooler radiator and the coolant radiator to close the gap so all the air goes through both. Another sealed the radiator and pre-radiator with rubber foam so all air passes both coolers.
• One owner enlarged the hole around the flex pipe. Another suggested cutting a slit in the main rear floor pan and bending it down to direct air onto the sump, and thought the small plastic flaps do little. A third said venting the rear does little for engine temperature because the engines are water-cooled. Owners differ.
• Adding a second water radiator was suggested, but the owner with the pump thought one that would fit wouldn't help much, citing a test where he took off the heater air hose to the cabin and ran the heater on full, venting the hot air under the bonnet, which helped only a little.

Oil cooling

• One owner added an extra oil cooler radiator in the passenger side ear, with a thermostat on the oil filter sandwich plate so it only flows when hot, plus a louvred vent behind it. Another thought the thermostat unnecessary because the oil pump assembly already has one.
• Another considered replacing the oil-water cooler with an air-oil cooler. One owner advised against removing the oil/water matrix: add cooling to the oil circuit and don't take things away, since over-hot oil is more dangerous than over-hot water. A suggestion made on advice from a cooling engineer was to fit any air-oil cooler in parallel with the original at first, since fitting it in series creates a pressure drop that affects oil flow.
• One practical point from another owner: water at about 1 bar is easier to plumb than oil pressure.

Throttle body pipes

• Two owners say the coolant pipes to the throttle body serve no purpose in a hot engine bay (you won't get throttle freezing), and one simply joined his together. Another rerouted his to cool an external wastegate. A suggestion to route that pipe into the turbo for hotter water was floated but no result was reported.

Before you spend money

• Bleed the system properly first. See Changing the coolant and bleeding the system.
• If the engine won't cool at idle, one owner suggested worn pump vanes, though the thread didn't test that. See High coolant temperatures.
• For how the circuit is meant to work, see How the cooling circuit works.