Join Date: Jun 2006
Location: Eastern ON Canada
Car: 07 OBW 2.5i D-4AT
Feedback Score: 5 reviews
Just brainstorming here . . . does air have to be drawn into the cooling system to result in bubbles in the overflow. I believe not. Even coolant of the correct concentration and under 12 psi pressure has a boiling point. And, if it boils, there's bubbles, and the bubbles are carried by the flow to the top of the radiator and out the rad cap to the overflow.
Boiling of coolant can happen around the upper cylinder walls and head if the flow is inadequate. This is normally prevented by having the pump designed to circulate the coolant at a high enough rate, even at idle. It's also why the heater circuit is constant flow even when the thermostat is closed -- otherwise, in these engines, there no circular path for coolant to flow inside while the thermostat is closed.
Water pump? Could be -- perhaps a loose impeller or corroded impeller not pumping well enough. (These have been mentioned earlier.) I also seem to recall a case where the timing belt was slightly loose for some reason -- not enough to cause timing problems (because of the "toothed" belt) but enough to allow the smooth back of the belt, which drives the pump, to slip.
Incidentally, I'm thinking that the problem of boiling is made worse when the heater core is in the loop because it isn't exactly a "full flow" device. But when it's bypassed, what limited coolant circulation there is has a better chance to limit the boiling. Air in the coolant isn't a good heat conductor, so the more boiling, the more bubbles and the hotter the coolant can get. In this case, even if the thermostat is open, the limited circulation won't be sufficient.
In regard to the chemical tests, are you using it at the overflow reservoir or at the radiator neck (where the cap would be)? If the former, I wonder if that could be affecting the readings. The air would be passing through the relatively cool coolant in the reservoir. Perhaps temperature affects the way the chemicals react.
Again, just brainstorming . . .