Evaporative coolant loss from unpressurised overflow tank?

Status
Not open for further replies.
Joined
May 16, 2019
Messages
15
Location
Singapore
So my overflow tank is one of those unpressurised designs, with a simple snap-on plastic cap that appears fairly watertight. It is exactly this one.

I've been monitoring the cold coolant levels weekly, and noticed that over two months or so, the level would drop about 1cm or so, corresponding to a loss of between one to two ounces. This is happening even with a fairly new radiator, rad cap and hoses, which are bone dry and by now covered with a layer of dust (no evidence of leaks). The radiator neck and external surfaces of the overflow tank are likewise. Zero coolant smells when the heat is on. Head gasket appears good, oil has no evidence of coolant in it at the end of each OCI, and the engine as far as I know has not overheated in its lifetime.

So I get a little frustrated with getting nowhere on this and start sniffing around my engine bay after long trips. Upon lifting the hood there is no coolant smell. I stick my nose next to the radiator cap - no coolant smell. But when I stick my nose next to the cap of the overflow tank, I catch a very brief whiff of coolant sometimes. It lasts not more than a few seconds and it becomes imperceptible again. I check the cap, it looks perfect, as does the mouth of the bottle. The cap snaps back on with a reassuring click.

I'm wondering if minor coolant loss such as what I observe is normal for unpressurised overflow bottles. It is obvious that the cap does let out minute amounts of vapor (by design??). FWIW, I'm running a premixed green coolant at 30/70 v/v, so it's got quite a bit of water right out of the bottle. What I'm also wondering is if it makes sense, and/or if it could be in any way detrimental to perhaps stick an O-ring into the groove of the plastic cap to try to create a tighter seal.
 
Last edited:
Donald might be right. Maybe just continue monitoring. My 6 year old car lost about 12 ounces of coolant during an over-winter OCI. I topped it off and for the next few months I had to add a tiny bit more and then it stabilized and has been fine since.
21.gif
I don't think the overflow tank cap is the problem. Sealing it might prevent the tank from working properly.

I suppose a pressure test kit might be the first attempt at diagnosing a problem. Sometimes there are extreme tiny leaks that only occur while driving that never become obvious. An example is the weep hole of the water pump.

Good luck on this and I am sure others more experienced will chime in - It's BITOG and there are great members here with lots of varied experience.
 
You are correct. When coolant gets hot-ish it's only the pressure cap that keeps it liquid. Once it passes that seal it flash-explodes to vapor inside that little 5/16 rubber hose going off to the overflow.

Now, that gas hits the overflow and bubbles up through it, which gives it a chance to cool down again. I muse that if one kept the overflow somewhat overfull, that gives the vapors more time to cool down and recondense into fluid. Possibly reducing boil-off consumption.

I mean, why do they give you all that space above the "hot" line?
laugh.gif
Worst case, you'll burp a little out, and find your new hot level.
 
My 2.7 Nissan have the same design using non pressurised overflow tank and I also have to add some coolant over time. This is my first Japanese vehicle after a 90s Civic and I haven't seen this design or symptom on any of my Euro cars. When there is a coolant loss on a Euro vehicle you can be assured there is a leak.
 
My Rx7's reservoir is the same style, I have to add some every few months. I don't think it's unusual given that there is space for vapour to escape, and it gets hot which makes it more likely to evaporate.

I wouldn't worry about it, just keep filling with distilled water as necessary.
 
Last edited:
Originally Posted by eljefino
You are correct. When coolant gets hot-ish it's only the pressure cap that keeps it liquid. Once it passes that seal it flash-explodes to vapor inside that little 5/16 rubber hose going off to the overflow.

Now, that gas hits the overflow and bubbles up through it, which gives it a chance to cool down again. I muse that if one kept the overflow somewhat overfull, that gives the vapors more time to cool down and recondense into fluid. Possibly reducing boil-off consumption.

I mean, why do they give you all that space above the "hot" line?
laugh.gif
Worst case, you'll burp a little out, and find your new hot level.

This exactly. Any non pressurized overflow tank that vents to atmosphere will lose a little fluid to evaporation. Keep it topped up right to the MAX line or a little above. As a benefit, each time you add a few ounces of coolant, you're replenishing the additives in the coolant.
 
Since you installed new cooling components, the level may be dropping from air burping out the system. New hoses also stretch a little. Just keep an eye on it but this sounds normal.
 
Thanks for the replies. I have learnt something new today.

My previous ride (a GM product) had a pressurised overflow tank which did not require me to add a single drop of coolant or water for five years. In the sixth, the liquid level dropped maybe half an inch and I noticed that just before the water pump gave out. Hence my overreaction to observing the coolant level dropping in my Mazda.
 
Status
Not open for further replies.
Back
Top