Re-post of one I made some time ago.

Originally Posted by RichardV6
In the course of fitting a SiFab radiator I was able to trace the cooling system as a whole. These were my findings helped by the cutaway image below of the Ford Mustang 3.7 V6 the Roadster engine is sourced from.

[Linked Image]

The top coolant rail (hidden black tube) from cylinder heads runs along the V of the engine forward to the casting with blue label, feeding coolant to radiator via top hose connection. This casting also contains the narrower bore permanent bypass (part with cast in text) to a mixing chamber below (shown as cutaway) to provide quickest warm up. The bottom hose of rad feeds to the thermostat housing with the temperature sensing side of thermostat facing inwards to the mixing chamber and directly onto bypass flow. This chamber also takes coolant from a "T" piece on the Roadster (not shown) connecting to the smaller bore hose union in image. This allows a heater return when there is a flow, and a head of coolant from lower point of header tank above in Roadster installation to keep system topped up. The heater is fed from the rear end of top coolant rail.

The chamber feeds via casting to water pump inlet, hidden behind black pulley below, in turn pushing coolant to lower part of engine. When engine warms, the bypass flow starts to open stat allowing a cooled flow from radiator to mix in with the bypass and heater return, the larger bore rad hoses giving a preferred path for coolant providing stat is open or partially so. In this way the stat is able to control coolant temperature through a wide range of engine power demands.

The header tank also accomodates two very narrow bore bleed/siphon hoses from top of rad and top coolant rail with connection to latter seen adjacent to blue label.

The system works so well temperature gauge readings during warm up and normal running are totally unaffected by the more efficient SiFab radiator and remain just below mid scale under most driving conditions, the only difference being the rad fan rarely comes on now.

I believe the recent four cylinder Ford's cooling systems work in a similar fashion.

I would add that I have never had any issue with long term leaks other than some very minor ones from new on heater hoses. I don't regard loose clamps however slight as indicative of coolant design problem on any particular model. Ditto for the OE plastic side tank rads used across the full classic range in recent times. I will admit to taking a week or two to fully seal hose connections on replacement ally rad when the OE one failed on my car, as with so many others. Finally cured by doubling up clamps on the silicon bottom hose extension required. However I haven't needed to top up since, well over a year and 7000 miles ago, despite a recent rolling road session where the coolant peaked at 106°C (normally 96 - 98°C) due to lack of adequate rad airflow.

I am not aware of any issues unique to the OE 3.7 Roadsters cooling system but happy to be corrected by other owners of this model. The larger aluminium replacement radiator was a seamless replacement in terms of warm up times and temperature indications. Cooling fan aside which rarely needs to run now. This cannot be said of some 4 cylinder models I understand where the extra cooling from ally rad can actually cause overheating on fast motorway runs due to thermostat delay in opening. The cure being lower temp thermostat.

Richard

2018 Roadster 3.7
1966 Land Rover S2a 88
2024 Royal Enfield Guerrilla 450
1945 Guzzi Airone