Storing petrol in a metal jerry can - is it better to keep the can partly or fully filled?

[NY]

When I once had a flight in a little Cessna, the pilot explained his pre-flight checks. He approached each wing and held a measuring cylinder under a drain c*ck at the lowest point and let out a bit of the liquid. He checked whether there was a meniscus part the way up the liquid (change from water to petrol) and dipped a bit of paper into the liquid; having thrown the liquid on the grass, he lit the paper to prove that it was soaked in petrol, not water.

Most fuel tanks have the outlet slightly above the very bottom so there is space for any contaminating water to sink to the bottom and yet not get drawn into the fuel line. On a boat, the worst that can happen if you draw water into the engine is that the boat will stop - albeit maybe in the path of an oncoming vessel - but you don't want the engine of an aircraft to start trying to run on water when it's in mid-air...

[Chris J Dixon]

Isn't that why a dipstick with special paste is used to check how much water is at the bottom of the tank?

Chris

[Dave Liquorice]

But not a lot. Can't say I've ever found any water in any fuel can. Bottom of the plastic heating oil tank yes, lots but I suspect that's rain that gets blown in via the unsealed filler or vent caps. I've fitted a glass bowl filter on the outlet of the tank since to catch and see any water coming out of the tank. After water in low points of fuel line froze enough to starve the boiler of oil...

I suspect that is more down to an abundace of caution. If you you think you have x litres of fuel you don't want to find, once up in the air, you have x litres of fuel and y litres of water. Not that you should be flying that low of fuel in the first place. B-)

[nightjar]

I had a test cylinder with a plastic ball in it, which would float in water, but not in AVGAS. Much easier than trying to spot the dividing line by eye.

[jimzzr]

What these stories illustrate is the importance of a fuels flash point. In essence a sample of fuel is gradually heated and a test flame applied until an ignition occurs. I.e it is the temperature at which there is sufficient vapour pressure that the concentration of volatiles in the vapour space above the liquid reaches the lower flammability limit.

For diesel the flash point must be >60 deg, For Jet A between about 40-60 deg, and vegetable oil well over 100 deg. However, the flash point of petrol is around minus 40 deg. In fact the fuel is so volatile that at room temperature in a sealed can/petrol tank that the equilibrium vapour concentration is actually above the upper flammable limit (needing further dilution with air to explode. If it had been a diesel tank that overflowed at Buncefield there would never have been a possibility of a vapour cloud explosion. The regulations for petrol and diesel are different for good reason.

[nightjar]

...

Not on each individual occasion, but there is a cumulative effect.

It can get cold up there and even trace amounts of water can block a fuel line by freezing in it.

[newshound]

IIRC there was a modern commercial airliner disaster perhaps ten years ago that was attributed to ice in fuel lines.

[newshound]

Perhaps this one

although no fatalities.

[Clive Arthur]

The metal can has a long history and has been lauded in song...

"God rest ye, metal jerry can, let nothing you dismay."

[Dave Liquorice]

Would take an awful lot of suitable cycles for any apprecihable amount to build up in a 5 l fuel can. Water will slowly be absorbed by the fuel and carried away.

It can get cold enough at ground level to freeze the water in a 15 mm copper fuel line... Said fuel line now only has one low point and a water trap (glass bowl filter) at the feed end. B-) The globule of water in the bowl does shrink slowly as the 2000 odd litres per fill are used. It generally grows a bit after a fill up and the contents of the tank well stired.