I am confident that my COOLANT consumption is normal but I would like to reduce it. It seems to me that coolant overflow systems are less than ideal. The air space above the coolant in the bottle must contain some components of evaporated coolant. That air is constantly pumping in and out. Engine heat must be increasing that evaporation. In addition, especially whenever the engine is running, air is blowing over the air-bleed hole and whisking coolant vapor away. I wonder too about the size of the bleed hole. My father's old 1978 Mazda 626 had a pinhole whereas my latest two cars are about 25 sq.mm. ... why so large? I also imagine coolants contain antioxidant to reduce rusting and oxygen from the air would consume some of it.

Ideally I would think a sealed system like a bladder that could expand and contract would be the ultimate solution but apparently not? Is there something against heat shielding the bottle? Is the gradual loss just water, meaning keeping the bottle topped up with distilled water rather than coolant is ok ? Keeping it topped up to the max line would keep the air space/vapor loss to a minimum too? Coolant changes are a pain what with flushing, expelling air pockets, disposal of the old, mess etc. and perhaps the intervals could be extended with system design improvements?

Nowadays, to minimise evaporation, I insulate the coolant overflow bottle from radiant heat with a piece of aluminium foil between it and where I suspect engine-heat sources to be. I noticed that some makers shield the battery the same way, so I copied the idea and I also shield plastic parts. My engine compartment looks quite "Christmassy".

  • 4
    If you are loosing coolant, you have a leak.
    – HandyHowie
    Oct 22, 2018 at 9:03
  • What @HandyHowie said. The only times I have ever topped up coolant in 40+ years driving are when something has sprung a leak. My current car hasn't been topped up for the last 8 years, and the level is still slightly above "full," exactly the same as when I bought it!
    – alephzero
    Oct 22, 2018 at 14:24
  • As I said in my first sentence, I have no reason to think I have a leak. All my cars have needed topping up from time to time, and as the overflow-bottle system is vented it must lose some. Perhaps, if your coolant contains propylene glycol which is hygroscopic it is absorbing water from the air at the same rate as it is losing. That thought gave me a clue. We are in a low humidity place and you may be in a place high enough to balance loss to gain. Oct 23, 2018 at 1:08
  • I have a 2010 vehicle with 235k miles, I have never added coolant to it. I did completely flush and change out the coolant about 6 years ago.
    – Glen Yates
    Feb 9, 2022 at 22:49

1 Answer 1


Most cooling systems are NOT vented - as the system is designed to run at a given pressure above atmospheric (15psi is common) to increase the boiling point of the coolant when the system is at running temperature.

An expansion bottle could be vented, but that is there to catch any fluid blown out of the system and stop it polluting the environment.

As for coolant consumption, I have added 100ml in a period of 3 years to my car... Basically that was after the pump was replaced and it held a small amount of trapped air.

  • I believe from what I read when overflow bottle systems were first introduced (around 1970?) the idea is to ensure there is never air in the engine i.e. under the radiator pressure cap. When the coolant expands, the cap is designed to let enough coolant out to avoid undue strain on hoses. When it cools and contracts, that overflow is pushed back by atmospheric pressure. At the same time, as you said, the system is pressurised enough to increase the boiling point of the coolant when the system is at running temperature. As I said above I believe the humidity/hygroscopic theory explains. Oct 23, 2018 at 1:22
  • Our average annual humidity (Adelaide, South Australia) is 47% and we are probably a bit lower as we live 20 km from the coast. It would be interesting if you could tell us yours. p.s. air pockets allow hot-spots in the engine Oct 23, 2018 at 1:33
  • googling further I read that if the ambient humidity (also referred to as RH = relative humidity) is less than 70% then propylene glycol/water mixtures will give up water which suggests my theory is correct. Nov 3, 2018 at 23:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .