Shorrick_Mk2 wrote:
Is the software in a VFR box of a demonstrably lesser quality as in a IFR box?
It’s not really the quality of the software as such, it’s the design assurance and the functionality. For IFR, the business logic tree has a lot more branches than for VFR.
Shorrick_Mk2 wrote:
Is the software in a VFR box of a demonstrably lesser quality as in a IFR box? If yes – please demonstrate or document the technical differences.
I admit I don’t know anything specifically about avionics programming, but I do know two things about software engineering in general which may be relevant:
- It costs a lot to certify software. It may be as good in the uncertified (VFR) box as in the certified (IFR) one, but you wouldn’t know.
- You can’t get good quality software by testing and debugging. It has to be designed in from the start.
Shorrick_Mk2 wrote:
Is the software in a VFR box of a demonstrably lesser quality as in a IFR box?
Who knows? Considering it is enough with an old school vacuum driven horizon or two, I would think a new AHRS is much better in every single respect. But again, who knows? and that is probably the main issue. An instrument build and certified according to a TSO have fewer unknowns. Also, experimental stuff are updated regularly when bugs are found, improvements are made, but that could also create new bugs of course. What I mean, is the LAA should encourage people to get as much certified equipment as possible, include clever dissimilar redundancy etc, and otherwise do nothing, because IMO what they are doing now serves no one any good. How can LAA tell this stuff apart?
Ok, so all we get so far is that we think it’s better than vacuum but worse than “certified IFR” however there is no measurable data or assessment protocol.
For VFR non-certified aircraft, there are no specific equipment qualification requirements. However, in certified aircraft, electronic equipment including EFIS need to meet strict certification requirements. Software has to be compliant to DO-178B at level C or above depending on the aircraft weight category as stated in FAA AC23-1309-1C Complex electronic hardware must be complaint to DO-254. (EASA have adopted the same requirements).
The failure case for a particular type of equipment in a given airframe must be assessed using a failure modes and effects analysis and a fault tree is normally developed. In the case of an EFIS approved for primary attitude in IMC, failure cases such as total loss of attitude (EFIS and standby), or un-annunciated misleading EFIS attitude with good standby, must be shown to have a probability of failure better than 1E-7 (or even higher in heavier aircraft).
The equipment must meet environmental qualification per DO160 including vibration, temperature, emissions and RF susceptibility. Additional requirements also exist for compliance with HIRF and lightning – now addressed by EASA under AMC20-136 and 20-158.
It’ll be interesting to see how EASA approach approval of the Dynon and Garmin attitude indicators that have recently received FAA STC approval without having specific equipment qualifications.
wigglyamp wrote:
It’ll be interesting to see how EASA approach approval of the Dynon and Garmin attitude indicators that have recently received FAA STC approval without having specific equipment qualifications.
Maybe these are made according to applicable standards, but are not produced under a TSO Authorization? Similar to an experimental Lycoming. I guess stand alone attitude indicators are easier to certify than an EFIS also?
