V8 cooling

The cooling is carried out by means of a Davies-Craig electric water pump and electronic controller. There is no conventional thermostat. The electronic unit is connected to a sensor situated in the outlet from the top of the engine.

The radiator is cooled by an electric fan from a Rover two litre turbo engine. There is no mechanical fan powered by the engine. The fan is switched on and off by a capillary style thermostat. The sensor is located in the bottom hose, the cold outlet from the radiator. This means that the engine receives water at a more or less constant temperature which is how I think things should be.

I like capillary temperature sensors as I believe they are pretty reliable compared to electronics and they don’t need electricity to work. (I like capillary water temperature gauges for this reason.)

The original pump has been replaced by a blanking plate fitted with a hose outlet from a Rover T-series engine.

Pump blanking plate

Short belt (no pump pulley)

Davies-Craig electric pump

The sensor for the electric fan fits in the engine bottom hose, following the pump. In the pictures below, copper tube has small tube silver-soldered in and small sections of tube soft-soldered each end to prevent the hose pulling off. The small tube is big enough to pass the capillary sensor through. The tube is then sealed with epoxy and protected with adhesive lined shrink-fit tubing.

The thermostatic switch and 40 amp relay fit in box with a 30 amp strip fuse. It needs a heavy duty supply preferably direct from the battery.

Switch/relay unit

Later, I developed the circuit (see below) to include  a feature whereby the actions of the pump controller and the fan thermo switch could be over-ridden by a dash switch which would turn on both the pump and the fan continuously. This would be useful if the controller failed or the engine developed a fault or you just wanted to cool the engine down rapidly to avoid, say, vapour lock in the fuel system. It is very surprising how rapidly the temperature reduces with the pump and fan on full!

The circuit uses a relay to switch the pump from the controller to a plain 12 volt supply. A diode prevents normal operation of the thermo switch from triggering the over-ride relay. The 470 ohm resistor seems to fool the controller into thinking the fan is still connected otherwise it would register a fault condition I think.

Rover 820 Turbo Vitesse Sport electric fan. If current consumption is anything to go by (it might not be!) this is one heck of a powerful fan.

Switch on dash turns on both radiator fan and water pump for very rapid cooling (as long as there’s still coolant left in the system! - I’m a bit of a glass half empty guy, so says my wife or should that be engine half empty?)

See picture left. The Davies-Craig fan controller lives in the electrical/electronics tray I have made to go under the dash on the drivers side (car is LHD). There are status LEDs and a setting button which can be seen or go at from under the dash as the bottom of the controller sticks out through a hole in the bottom of the tray.

(I didn’t intend for all this stuff to get so complicated, it just happened, officer.

The truth probably is it’s more pleasant to do electrical stuff on a bench than to crawl under the car and get greasy! It’s a sort of displacement activity.)

Since the Davies-Craig pump controller pulses the pump to maintain the desired temperature, there is not a continuous flow of water through the heater. For this reason an extra tiny pump is required if you want to be cosy in the cabin in winter! The pump is controlled by a switch under the dash.

Small heater pump

Later it occurred to me that the Davies-Craig pump was rather vulnerable to being disconnected and immediately losing all the coolant if the car ran over anything more than about 125 mm high! I had to provide some protection …

In implementing this circuit, my first attempt attempt failed after about a 1000 miles! I had a strip fuse clamped between two bolts acting as terminals for the heavy cable that the fan needs. The nuts must have loosened and the terminals overheated, eventually causing the fuse to fracture (rather than blow). Fortunately, the Davies-Craig pump gets its supply from another route (unless the override switch is operated) and the car was not in heavy traffic in the middle of summer so no harm was done except that I needed a new terminal box.

Heat damage to 5 mm tufnol terminal block.

I made a new base for a new terminal block out of tufnol and used M5 brass screws for the high amperage terminals. Brass conducts electricity better than the stainless steel I had used previously so there wont be as much potential for the joints to heat up unless the nuts loosen. That’s where the extra nyloc nuts come in.

I’ve also made the whole thing bigger and fitted it into a bigger box so that the wiring won’t be so squashed which should reduce the risk of heat build up and short circuits.

This time I’ve put the strip fuse in a purpose made holder on the outside of the terminal box for better access.

It should be an improvement, hopefully.

These connectors plug into the thermo-switch which is fitted into the lid of the box (see the picture near the top of this page).

See here for an alternative to the electro-mechanical stat.

    on the mechanical fan   thermostat described on this page below can be seen here.

NB: An improvement (I hope!)