Engine coolant loop change thought experiment

This is a thought experiment. The goal is to do a coolant exchange without putting air into the system, and without needing an AirLift or equivalent. It is a follow on from my thread about filling the system from the bottom

Refill ICE coolant loop through drain valve under thermos? | PriusChat

So picture this. Attach a Lisle funnel to the radiator neck and fill it with coolant. Attach a tube to the drain port under the thermos. Raise the other end much higher than the drain port and pour coolant into it until it is full to the port. Flick it with a finger to dislodge any stubborn bubbles. Pinch off the far end and place it below the fluid level in a large container with some coolant in it. Open the drain valve. Keep filling the funnel until it gets to 2-x gallons then close the drain valve and cap the radiator. With a fluid transfer pump remove all the coolant from the overflow and then refill to midline with x gallons of new coolant. (I don't know what x is, some fraction of a quart.)

If the funnel was never empty there shouldn't be any air in the system. It shouldn't be necessary to do any of the usual tricks to get the (not there) air out. That would be a big time saver.

Down side - one would have to watch the funnel like a hawk, because if it is ever empty, then air can enter.

I think that with the radiator cap on, pumping out the overflow tank and then refilling it cannot introduce air into the system. If the cap is on and the tube is full of water it shouldn't be able to empty into the overflow tank because that would create a vacuum and prevent any more from coming out. If somehow the level gets below the bottom of the tank the new coolant will fill that back up again.

On the downside, the coolant will be a mixture of old and new coolant. More than usual, where there would only have been the old coolant in the radiator core. That said, the usual coolant change results in roughly a 3:6 ratio of old to new, so it isn't the cleanest exchange either. I think the majority of this mixing would occur in the thermos, which is a large open volume. The mixing would only occur once the new coolant reaches the tank (before that it is just old coolant replacing old coolant). Once the new coolant gets to the tank if the flow is stopped temporarily, the system allowed to come to rest, then the valve opened just a little, there should be much less turbulence. The old coolant would flow (slowly) out the bottom with the new coolant flowing (just as slowly) in the top. Unless a tracer is added though, one would have to just estimate when to stop and restart the flow.

Another issue is that the coolant in the thermos will normally be hot, while the new coolant will be cold. That would promote mixing from convection since the new cold entering would fall to the bottom while the old hot would rise. I'm not sure how much of an issue this would be if the car had sat long enough for the majority of the coolant to have dropped to ambient temperature. The old (now) cold coolant outside the tank should have displaced most of the hot coolant from the thermos before the new also cold coolant arrives, minimizing the temperature difference. The volume of the heat storage tank is around 3 liters, and in a normal change, 6 quarts is added, so roughly 3 quarts of cold would have been pushed through the tank to mix/displace the same volume of hot fluid which was initially present. Or the car could be turned on briefly earlier to trigger the pump to evacuate the tank, then let the car sit for a while to cool off.