Effect of dual control inputs on elevator response

Flexibility in the ATR 72’s pitch control system between the control columns results in a change in the aircraft’s longitudinal handling qualities and control dynamics when dual control inputs are made. This could result in an aircraft-pilot coupling event where flight crew may find it difficult to control the aircraft.

Response by Avions de transport régional (ATR)

The flexibility in the ATR72's pitch control system between the control columns is not inducing a specific issue in the aircraft's longitudinal handling qualities. The design assumption on all large commercial transport aircraft is one 'pilot flying'. As provided in previous comments standard operating procedures are set up to preclude occurrence of dual input. 

ATR acknowledges the addition of new pages to the final report that relate to Longitudinal Handling qualities and are using "Etkin's" curves to substantiate the analysis. ATR has a different interpretation of the analysis of rigid system vs. flexible system. Indeed, the "Etkin's" curves will be modified in case of any dual input, whatever the configurations (rigid or flexible). 

For the event under investigation, the 3.34g load experienced in the longitudinal axis (before the declutch) have been obtained through dual application on the commands in the same direction. In this case, the "Etkin's" curve modification will be of the same order for flexible and rigid control column interconnection system. Moreover, the aircraft response was consistent with the flight crew inputs during the event. 

Aircraft-pilot coupling (pilot-induced oscillation) is a recognized phenomenon within the industry. ATR has addressed for example during the ATR operators Flight Safety Conference the in-service experience of large rudder inputs during landing roll resulting in larger than expected aircraft movements. ATR would be happy to further discuss APC for this event, considering the safety issue was raised in the latest version of the report. 

However, the consequences of dual input would need to be taken into account as one pilot will react non-linearly to the other pilot input. The system to consider is the system considering <Aircraft û Other Pilot> and not only the <Aircraft>. This is fully described in the document reference 98 mentioned in the provided draft report (Aviation Safety and Pilot Control û Understanding and preventing Unfavorable Pilot-Vehicle Interactions).Therefore, the note 99 of the draft report is also applicable to rigid systems and is thus independent of the location of the interconnection point. 

Finally, ATR would like to recall that the following actions have been taken at various industry levels: 

- ATR released the AOM42/72/2016/03 and revised the FCOM/AFM/QRH to raise crew awareness regarding the potential detrimental effect of uncoordinated crew input and/or large and aggressive flight control input at high speed. 
- EASA released the SIB 2016-20 rev.1 to highlight the risks associated to rapid and large alternating control inputs. 
- EASA added the "Inappropriate Flight Control Inputs" item to its risk portfolio in the frame of their risk management system, recognizing this is an industry concern. It will cover the issue of simultaneous inputs, as well as inputs of large amplitude or frequency inadequate for the flight phase at the event. 
- Paragraph 5.3 of the ICAO Airplane Upset Prevention and Recovery Training Aid revision 3 (AUPRTA ICAO revision 3 (AUPRTA).pdf. 

ATSB comment

ATR's response that the safety issue does not apply to their aircraft appears to be based upon two premises: "The standard operating procedures are sufficient to preclude dual control inputs". The referenced information and discussion on aircraft-pilot coupling applies to systems with rigid and flexible interconnection between control columns. 

Analysis in this report, based upon ATR supplied data, has shown that dual control inputs occur at a rate that were they any other system failure mode, would be considered 'probable'. As such, standard operating procedures have not been shown to sufficiently protect the aircraft from dual control inputs. 

The ATSB acknowledges that the referenced document relating to aircraft-pilot coupling applies to both flexible and rigid interconnected aircraft, as was detailed in the draft version of the report; however, ATR have not provided any argument to show that the differences between the responses to dual control inputs in the ATSB analysis does not specifically apply to their aircraft. Other than claiming that they have a different interpretation of the analysis, ATR do not identify any deficiencies in the ATSB's detailed analysis in either the safety analysis section, or appendix C. 

In summary, the response from ATR has not provided sufficient evidence and/or argument to convince the ATSB that the aircraft does not have a susceptibility to aircraft-pilot coupling events when dual control inputs are made. Thus, the ATSB makes the following safety recommendation.

In their response of 26 June 2019, ATR did not provide any new and compelling evidence to indicate that the risk of this safety issue had changed. Again, they were focussing the attention on crew resource management, which may reduce the probability, but has not been shown to prevent dual control inputs. ATR has not provided any further updates or responses to indicate that any further work to mitigate the identified risk has been taken. As such, the ATSB has decided to close this safety issue as not addressed.