Menu Sign In Contact FAQ
Banner
Welcome to our forums

Fuel icing and precautions

With the thermometer showing -0.1C this morning maybe it is time to consider this old topic…

If anticipating flight below about -20C, I am putting a litre of IPA (isopropyl alcohol) into my tanks, half in one and half in the other. That works out at roughly 0.5%. I put it in at the pumps, just before a fillup, so it gets a chance to mix well before the next flight.

Administrator
Shoreham EGKA, United Kingdom

I flew the Mirage all last winter with OATs at -40 and below and never added IPA. Jet fuel is more of an issue with cold temps than Avgas.

EGTK Oxford

Ii find that Avgas icing is less of a problem than it used to be. Perhaps the improved fuel storage regulations mean that less water is dissolved in the fuel.

That’s possible.

Avgas itself needs -58C to freeze, so any issues must be to do with dissolved water.

Also I am sure it is very aircraft type dependent, with perhaps two things affecting it:

  • whether any parts of the fuel system – where the fuel moves slowly – are exposed to the ambient air
  • whether there are easily blocked filters etc
Administrator
Shoreham EGKA, United Kingdom

The aircraft which has been mentioned as suffering severely from fuel icing is the Aztec.

This old document is interesting, and suggests that the design of the fuel system is defective and allows water to accumulate in some places, where it then freezes.

Otherwise, I completely cannot understand how one can possibly get a fuel icing related engine failure, given avgas 100LL freezes about -58C.

A recent flight test with an instrumented engine reveals that the temperature drop along the thin fuel pipes is negligible, and certainly under 2C under any possible operating conditions (fuel temperature, OAT, etc) in cruise.

Of course if you close the mixture then the pipes (and eventually the rest of the engine) will cool right down to the airflow temperature… but if the avgas is clean, there is nothing to freeze unless you are somehow flying at FL300.

Last Edited by Peter at 06 May 09:46
Administrator
Shoreham EGKA, United Kingdom

I have just got some data from a test flight done some months ago, on an IO540 engined aircraft.

Two PT100 probes were attached to points at the ends of a particular fuel injector line, and a third was reading the OAT. The flight went to FL120 for a bit and data was collected during climb, during the level bit, and the descent. The vertical scale is degC:

The result is certainly unexpected because it shows virtually no temperature drop across the pipe, whose length is about 40cm and which is fully subjected to the ambient air at about 150kt.

If fuel icing is happening, it is not happening in these very exposed pipes, but further back.

The brief rise in the OAT (of about 2C) during the coldest portion of the flight is unexplained. It may be real (maybe sunlight was shining onto the sensor during that part of the flight, which did have a number of direction changes) but it doesn’t have any impact on the validity of the test.

The probes

The four-channel PT100 data logger (USB connected to a laptop)

It is based on an old Pico PT104 product with the plastic box replaced with a diecast box for noise immunity and to fit a decent quality probe connector on it.

The logger was calibrated so all probes were within 0.01C of each other.

Administrator
Shoreham EGKA, United Kingdom

On the DA42 the fuel temperature has to be in the green for flight, I think it was +5C. I have only a few hours in the type, maybe those who fly them regularly can explain or correct me.

Darley Moor, Gamston (UK)

The DA42 fuel temp limitation applies primarily to non-Jet A1 (i.e. diesel fuel). In this circumstance you cannot start an engine at less than -5 and not take-off until it is +5. Take off is permitted if the fuel temp is in the yellow range (-30 to +4) and you are using Jet A1.

Peter, I was behing a pair of IO-540s the other day at 25 000ft with an OAT of about -38. Whilst we didn’t experience any fuel icing, one of the mixture levers froze which caused us think a bit when it was time to descend.

Fly safely
Various UK. Operate throughout Europe and Middle East, United Kingdom

Peter, I was behing a pair of IO-540s the other day at 25 000ft with an OAT of about -38. Whilst we didn’t experience any fuel icing, one of the mixture levers froze which caused us think a bit when it was time to descend.

Water in the mixture control cable

Administrator
Shoreham EGKA, United Kingdom

Probably. Still a bit of a PITA. Luckily judicious ‘waggling’ cleared the issue before the engine coughed and died.

Fly safely
Various UK. Operate throughout Europe and Middle East, United Kingdom
Sign in to add your message

Back to Top