The accident of a Continental Dash 8 Q400 in Buffalo raises some interesting questions about one of important weather hazards: icing. Note that I never said that icing is the cause of this accident – it is way too early to know about the causes – but it is a good opportunity to talk about icing.

Icing occurs when an aircraft meets liquid water at temperatures below freezing. Liquid water at temperature below freezing ? Yes, this can happen. Cooling water below freezing is not sufficient to turn water into ice. The crystallization process also requires a solid support to start. Water droplets found in clouds are made of pure water and no impurity can serve to start the process. If the air is cold enough these droplets can reach temperature below freezing and become “supercooled”. But supercooled is not super-cool (sorry – I could not resist) because when an aircraft meets such droplets, it acts as a perfect starter for the crystallization process. When flying in clouds at temperatures between 0 and -15 degrees Celsius there is a risk of ice build-up on the airframe, propeller, and any exposed part of the aircraft: landing gear, rudder, elevator, … Below -15 °C, the air can’t usually contain enough water to create icing hazard.

Ice affects aircraft in many ways – none of them being positive:

• airfoils shape is altered making them less effective (wings but also elevator, rudder, ailerons and propeller blades)
• weight increases
• air-intakes – including pitot and static – can be obstructed by ice
• in extreme cases control surfaces can be locked in position by ice accretion

Different systems have been developped to fight against ice. Anti-ice system aim at preventing ice formation. On light aircraft they are based on fluids dripping out of micro holes on the leading edges or on electrical heating for propeller blades or windshield. De-icing systems are used to remove ice after if it form. On light aircraft and some turboprops this takes the form of pneumatic boots: rubber tubes on the protected surfaces (mostly leading edges) that inflate and break the ice accumulation. On jet aircraft both anti-ice and de-ice use hot air from the turbines (bleed air) which is directed on the surfaces to be de-iced.

The black-boxes of the Dash 8 Q400 destroyed in the Bufalo accident revealed that the anti-ice system was activated during most of the flight. The crew discussed about the observed ice accumulation on the windshield but never qualified it of “severe”. The flight data recorder shown that the aircraft’s pitched up to 31 degrees, then down to -45 and the bank angle increased over 100° before the end of the recording. The autopilot was engaged when the things started to get wrong.

I don’t know about the causes of the crash in Buffalo nor if what follows is applicable but it is worth discussing it in a general way. Icing is not affecting all parts of an aircraft the same way. As all turbulent phenomena the impact of icing is not easy to predict. It can impact wings and elevator differently, or make a prop become unbalanced. You probably remember from your basic aerodynamics courses that the horizontal tail plane’s role is also to maintain the airplane’s pitch controllable. If the tail airfoil is badly impacted by ice the airplane can become uncontrollable.

The use of autopilot in icing conditions is generallly not recommended, and sometimes formally forbidden. I don’t know about the autopilot used on the Dash 8 – once again, I don’t have any idea about the causes of this accident – but here is an extract of the DA42 aircraft flying manual supplement about the “Flight In Known Icing” conditions. By default the DA42 is not certified to fly in icing but with some extra equipment this can be allowed (more about this in a later post). This supplement contains the following remark about the autopilot:

Use of the autopilot is prohibited when any ice is observed forming aft of the protected
surfaces of the wing, or when unusual lateral trim requirements or autopilot trim warnings
are encountered.

NOTE
The autopilot may mask tactile cues that indicate adverse  changes in handling characteristics; therefore, the pilot should consider not using the autopilot when any ice is visible on the airplane.

This is quite self-explanatory, but once again I don’t know if this applies to the Dash 8, this was a simple illustration how nasty icing can be. For light aviation icing is a very serious hazard and the only way to deal with it is to escape the icing area as soon as possible, even if this means turning around.

I’ll continue to read the information released by the NTSB about the accident in Buffalo. If you want to read more about it you can follow the following links:

Aero-news network about use of autopilot

AvWeb.com about icing

AvWeb.com about changes in attitude during the accident in Buffalo

Aero-news network about FAA recommendations on icing

I posted here about my personal experience and opinion about icing and light aircraft. Click here if you want to learn more about icing from NASA.