ATR 72 crash near Macao by the end of 2002

interesting read!

Brilliant Galdri , thanx

Hello eu copilot,

I´m a former ATR driver, so maybe I can give you some insight. There can be no doubt about it, in the early days of the ATR, they had some icing problems. Some of it can surely be blamed on the profile of the wing, I´m not going to pretend otherwise. But there is another aspect rarely discussed, namely the human factor in these accidents/incidents. It is not all the aircrafts fault, the biggest part of the blame must fall on the operation of the aircraft in these situations.

Before I go into specific accidents, I would like to explain what is happening to the wing of the ATR in Severe Icing Conditions. All of the ATR accidents/incidents related to Icing have happened in Freezing rain. What happens, supercooled waterdropplets strike the front of the wing and flow backwards over the de-icing boots before they freeze. Some of them make it to the area beyond the boots before they freeze. For the droplets to make it this far back they have to be of a certain size, IIRC they have to be bigger than 150 microns. In the certification criteria for Air Transport Aeroplanes dealing with flight into known Icing Conditions, states that the De-icing equipment of such an airplane must be able to deal with supercooled droplets of 40 microns. So it is evident that the aircraft getting into difficaulties because of Severe Icing are being operated outside their certification envelope. But we as pilots don´t have any equipment to measure the dropplet size, I get to that problem a little later.

When an ATR is flying in Severe Icing Conditions (freezing rain) with a dropplet size of at least 150 microns (flight tested at Edwards AFB) a ridge of Ice froms aft of the de-icing boots when they are operated. This ridge of ice starts to behave like a vortex generator, creating a very strong vortex behind it. Unfortunately the lowest pressure in this vortex is above the ailerons. Now, if the ice accumulation is completely equal on both wings AND the ailerons are both dead center, nothing happens as long as those conditions are met. Both ailerons have equal low pressure above them trying to suck them up. But this is not the perfect world. Ice accumulation is seldom absolutly equal and the autopilot is constantly making small roll corrections. As soon as the autopilot starts to make one aileron move up, the vortex grabs it and wants to slamm it to the stop. Initially the autopilot can handle the force, but a point is reached when it is overpowered by the vortex. The first indication the crew of the aircraft would have of that problem is a messeage on the ADU (Advisory Display Unit, the autopilot status sceen) saying…….AILERON MISTRIM. Moments later the autopilot disconnects and the bewildered crew is treated to a very fast aileron roll. From which they try to recover by pulling back on the yoke as the aircraft goes past inverted. Game over.

Now lets have a look at some accidents. The first accident caused by icing on the ATR was when an Alitalia (if I remember correctly) ATR crashed near Milan. The cause? The crew was climbing the aircraft in V/S mode on the autopilot through icing. The speed fell off, the aircraft stalled, autopilot disengaged and a quick aileron roll followed. They never pulled out of the resaulting dive.

The most famous ATR accident is the one near Roselawn Il. in the US where an ATR 72 crashed due to icing. They were trying to get into Chicago but the airport was closed due to weather. They were stacked in the holding over Roselawn at 5000 feet, on autopilot in freezing rain. They were holding in freezing rain for approx. 35 min. before it happened. Autopilot disconnected and a sharp aileron roll followed with a dive to the ground.

After this accident, ATR came under a lot of pressure from all and sundry. Boeing wanted to get a commercial exposure from the accident by trying to get the idea into peoples head that everything European was inherently dangerous. It was part of their early battle with Airbus. Various commercial pilot associations in the US jumped on the Boeing wagon and an intense lobbism was started to get the NTSB to be EXTRA careful in this case. As a resault the NTSB grounded all ATR’s flying in the US while the investigation was under way. The fact that the ATR that crashed had been flying in holding for over 30 min. In severe icing was completely lost on them. The ensuing investigation resaulted in the ATR becoming THE most studied aircraft in the world at the time with regards to icing. Nothing definate was found which indicated the ATR as a particulary dangerous in icing conditions. I´ve seen the uncut version of the tape shot on these experiments, and it is hard to understand how the aircraft could fly with all this ice on it. In the end the NTSB was forced to lift the grounding of ATR’s in the US.

Even though the test were inconclusive, ATR tried to fix the problem by extending the de-icing boots further aft with mixed resaults. Further more they issued a very compehencsive guidelines in the FCOM and the AFM on how to recognize and deal with severe icing. To recognize severe icing, it is recommanded that the pilots pay particular attention to water splashing or flowing on the front window when OAT is at or below 0° also pilots are advised to pay attention to unheated portions of the side windows for any sign of ice buildup. If one or both of these symptoms are present pilots are advised disconnect the autopilot asap and fly the aircraft manually. This is to prevent the autopilot from masking the effects the icing is having on the aircraft. Next step, according to the FCOM and the AFM is to leave and avoid conditions of severe icing, declaring an emergency if neccesary. Another point is never to let the speed go below 160kts.

Even after these guidelines were issued by ATR, we continued to receive one or two reports a year where ATR’s had departed controlled flight after flight into severe icing conditions on autopilot. Fortunately none of them resaulted in an accident. But it says a lot about the training and/or the attitude the people flying the aircraft. To have all these tips and recommidations and not following them is pure folly. I´ve flown the ATR in severe icing and by stricktly following the ATR procedures we didn´t have any difficaulties at all.

Now, back to you original question about this ATR that crashed late last year. It is very unlikely that this aircraft crashed because of icing IF the crew followed the correct proceedure.

Hope this answears some of your questions

Regards,
Galdri

thank you, I hope that he will read this, a maybe someone else who also operates ATR.

Galdi used to fly the ATR i think

Ice problem maybe.
I remember that ATR’s had some problem with ice in the early days.

By the way , welcome on board EU copilot 🙂