OK.
Background
Found out some times ago that Toyota Prius automobiles are fitted with a "quick warmup system" in the shape of a thermally-insulated coolant tank, which holds about 3 liters of coolant warm for some time, to be pumped back in the cooling system upon engine start-up.
1.8Ts can't be fitted with something similar, unless very complex and costly mods are performed (auxiliary coolant pump, separate "thermos" tank, valves, piping, wiring, dashboard controls etc).
But as we can (not easily) find out, on our 1.8Ts coolant recovery/overflow tanks communicate directly with the engine block, not the radiator. Lower pipe from the tank goes to an Y-shaped connector, from which one pipe goes to the turbo central bearing and the other one goes to the lower part of the block, above the coolant pump and before the thermostat. Tank holds slightly over 1 liter of coolant when filled to the middle joint and maybe 1.5 litres when filled up to the "G12" mark (not the wisest thing in the summer).
The design is made as such because the 1.8T timing belt also runs the coolant pump. So, when changing the timing belt and coolant pump, the coolant in the pump housing and recovery tank has to be drained away and when the tank is filled again, the coolant flows directly to fill the pump housing. (Otherwise the pump would run dry and destroy itself when starting the engine).
Were the tanks thermally insulated, the warm coolant would be drawn into the cooling system while the thermostat was still closed, and the warmup would be quicker, with less engine wear, quicker heating of the cabin heater, and a slight improvement in fuel economy. It takes energy to heat the engine, energy which otherwise goes wasted when cooling down.
First tests
With stock cooling system, the time to reach regular 90°C are:
Stone cold engine, wintery conditions - 10 to 15 minutes of driving
Cold engine after a few hours since last drive - 8-10 minutes of driving
I've performed a small experiment - temporarily insulate the coolant tank with a very good pipe insulator (Eurobatex - similar to neoprene mattresses). Insulator was about 16mm thick when stretched over the plastic tank.
Results:
18:05 - engine stopped. Outside temperature: 4°C. Everything hot.
21:47 - outside temperature: -1°C. Engine block: cool. Radiator and hoses: stone cold. Insulated tank: almost hot.
00:06 - outside temperature: -1°C. Engine block: ice cold. Radiator and hoses: stone cold. Insulated tank: warm.
As we see, the insulated tank could keep the coolant warm for at least 6 hours, and even more.
The dirty work
Removed the tank from the two screws and insulated it thoroughly with same material, every inch, all pieces glued together lip to lip, except a slit cut to see the coolant level. The slit will be also covered by a movable flap. Doubled the insulator thickness on the lower half of the tank, where there is more space, and also where most coolant stays. Glued also thin foamy insulator on the inner side of the fender, just below and behind the tank, insulated full innner fender and suspension turret. Return pipe from the A/C radiator in front of the car, which ran below the tank, had also been insulated previously with a foamy pipe insulator. Carefully left out the plastic catches for the wires and a small "overflow tube" on the side of the tank. Removed the metal plate that holds dipstick and N249 valve and insulated also the hose from the bottom of the tank to coolant ingress in the lower block.
Results:
1st test: engine warmup time (cold engine after a few hours) reduced to
slightly over 3 minutes - until 90°C are reached. Outside temperature 6°C. In about
2 minutes, the temp gauge had risen to the line between 50°C and 90°C (~70 degrees?).
2nd test: stone cold engine, car started after 5 hours of exposure to 0-3°C outside temperature, coolant in the insulated tank rather warm. In about
4 - 4:30 minutes, the temp gauge has begun to rise, regular 90°C attained in
about 7 minutes.
3rd test: stone cold engine, car started from cold in a spring day after exposure to 10-13°C outside temperature. In about
4 minutes, engine had reached regular 90°C operating temperature.
4th test: room-temperature engine, car started after 3 hours in the workshop, outside temperature 7°C. Coolant had barely cooled, maybe 2 needle widths below 90°C. In a few tens of seconds, it stood mightily at 90 degrees.
5th test: stone cold engine, car started on a cold and rainy day after staying for 24 hours, outside temperature 8°C. Warmup time: 5 minutes to 70°C, 6 minutes overall to 90°C.
Coolant temp sensor is on the opposite (airbox-) side of the block so it's not influenced.
When touching the tank after a run, insulator was barely warm at the surface from the hot air in the engine bay (and it remained cool in open air). The hot part through which heat is lost is the blue plastic cap, which may be insulated itself in the future.
Disadvantages: it works for 6 hours at its best. If the car has stood for more than 5-6 hours in winter chill, warmup time is shortened insignificantly, maybe by 1 minute, compared to stock, or not shortened at all.
Picture:
Comparison
The Prius system is obviously light-years more efficient, due to it's specific construction:
- large storage tank, up to 3 liters;
- true "thermos" (Dewar flask) construction, able to hold the heat in for more than a day, not hours;
- separate from coolant recovery/overflow tank;
- valve to control coolant flow in both directions (all coolant is drawn by the pump both ways);
- separate electric pump to feed either storage tank or engine;
- duct goes into the cylinder head, near combustion chamber, not block.
According to user-posted experiences on Toyota forums, it heats up from overnight cold in about
1.5 to 2 minutes.
~Nautilus
