All... I don't want to speculate on this particular accident... but the discussion about VFR night flight with little visual reference is very relevant. I don't think a lot of people appreciate the illusions that come into play during night VFR flying. Over DFW, and other highly populated areas, it's fairly easy to distinguish sky from ground at night. Same is true over completely unpopulated, dry ground, on a clear, starlit night. But I did my flight training in eastern Canada. Canada requires 5 hours of hood training to gain a night rating - for good reason. I was surprised, even when just outside the Ottawa area, how easy it is to lose your horizon. Reflections of stars from lakes and rivers, sparsely distributed ground lights that look like stars, combined with some cloud cover that can make areas of the sky look like ground, can easily conspire to turn the world up-side-down. I was lucky my instructor was one of the top Human Factors people in the business (Bob K.) and he gave me some great lessons on this.
I also don't think people understand how the current certification regulations stack against giving helicopters a better chance in these circumstances. There has been a lot of work in the FAR Part 23 (light airplane) world to lower the barriers to allow better technology in the cockpit. The Part 23 folks have recognized that "loss of situational awareness" kills way more pilots than misleading equipment and displays, yet the certification requirements for equipment and software keep going up, and the overall safety record keeps going down. Raising the bar on equipment is just not working.
For the most part, "complex" systems (ie: those with processors, software, and complex architectures) have to be certified via the air transport regs (AC25-1309 which leads to ARP4760, ARP-4754 and DO-178B for methods of compliance) Other directorates have simply adopted the Part 25 practice for complex systems, and so for primary flight instruments, the requirements look more like what you'd need in a 747 rather than a 407. That's why steam gauges live on.
The Part 23 folks (light fixed-wing aircraft) were the first to depart from air transport standards with AC23-1309-1C. They have proved their theory through the CAPSTONE project where new the technology (intuitive, synthetic vision displays, that don’t require the continuous currency training and mental gymnastics to interpret) were put in VFR aircraft. The result over the first two years of the study in Alaska (lots of water, lots of night, lots of VFR in near IMC, lots of stuff to run in to) was a 40% reduction in fatal accidents.
As a result of this Part 23 work we are seeing the new MFD flight screens from Avidyne, Garmin, Chelton(ex Sierra), Universal and the like that cost 1/4 of the heavy iron stuff. It's now going into the basic, fixed wing aircraft such as the Cessna 182, Mustang, Cirrus, etc.
Yet the technology barriers for Helicopters remain high. The big issue for Helicopter IFR is stability. Fixed wing aircraft tend to right themselves, most helicopters don’t. So in near IMC conditions, the helicopter pilot cannot afford to momentarily divert his attention (to retrieve a chart or similar) for fear to find the dirty side is no longer pointing down. The specific requirements for helicopter stability for IFR certification are detailed in Appendix B of FAR Part 27. Other than a few models (e.g. some Bell 206 models) few helicopters can meet the requirements without SAS or SCAS. With today’s technology, SAS or SCAS could be made cheaply. But now come the regulations.... Appendix B goes on to say that if a SAS is used to achieve that performance, then failure of the SAS has to be “extremely improbable”. The AC25-1309 interpretation of “extremely improbable” drives you to a triplex system and “level-A” software. For the most part, the certification authorities have accepted a dual system with “level-A” software, but that is because “they have cut us a break.” The DGAC noted in a recent meeting that they have been taking the more stringent interpretation of “extremely improbable” on recent certifications, and I suspect that is what led to the almost triplex implementation in the EC-135. So per the regulation, you either require nothing – if your aircraft is inherently stable, or you require duplex or triplex, level-A SAS (with dual hydraulics, etc). There’s no in between.
The point is that VFR helicopters would be safer in near IMC conditions if they were made more stable. They could be more stable if they had SAS. The bean counters don’t want to pay for SAS unless they get some kind of credit (i.e. IFR capable aircraft). For IFR credit, the regs require you meet “extremely improbable”, which means a high cost /redundant system. And so, the market for low-cost, VFR-only, SAS is soft. As a result, nothing gets done for light VFR helicopters. And accidents, such as possibly this one, continue to happen.
We need the same kind of thinking that went into the new Part 23 to be applied to Part 27. In the mean time, hold on... and good luck keeping the dirty side down.
Last edited by Avnx EO; 24th Oct 2003 at 02:52.