82 Vol. 9 / No. 12/ December/January 2017/18 Global Aviator
Aircraft Technical
By Dr Guy Gratton
Reinventing the wheel
change was entirely understandable.
And in the last year FAA in America
and EASA in Europe have issued their
new versions of part 23: FAR-23 (which
no longer seems to carry issue numbers,
just an expectation we’ll use the version
online at the time) in the USA, and CS 23
Amendment 5 in Europe. So far as I can
tell, the European and American versions
are identical so one helpful aspect is that
we now have a global level playing field;
they also divide aeroplanes into eight
categories now: all of them potentially up
to 19,000lb (8,617kg) maximum take-off
weight, which I’ve shown in Tables 1 and 2- four certification levels, each divided into
low and high speed performance levels.
Doubtless like all previous standards
we’ll see whole generations of aeroplanes
designed exactly to the limits of categories.
After this bit of very sensible change
however, what we then see is some very
radical thinking, that I’m afraid worries
me. The legacy part 23 standards – and
for that matter every other major civilian
and military airworthiness standard in
the world contains a great deal of clear
information about minimum acceptable
characteristics, strengths, gradients of this,
safety margins of that. It’s often said that
every line of these standards are written
in blood, and with good cause – many
thousands of people over many decades
have worked to create those standards
and they have served us all very well.
It means that anybody managing an
aircraft, or designing a modification, or
keeping going a vintage aeroplane, has
had a very good idea of what standards
they were working to, and could keep
aircraft in compliance with that standard.
And here’s the change – all of that
detail in part 23 has suddenly been
moved out of primacy and relabelled
“Acceptable Means of Compliance”. The
intention here is that of course twofold.
Firstly any company who wants tokeep using the old version of part 23,
in which they are probably very expert,
have every right and ability to do so and
won’t be penalised by these changes.
Secondly, that there are a great many
safety standards that exist outside of civil
aviation and may prove entirely fit for
the purpose of awarding a Certificate of
Airworthiness. This might, for example,
include CE standards for electronic
devices, military standards for an aircraft
built originally as a military trainer but
now being proposed for civil certification,
or some of the new ASTM standards
for light sports aircraft. This new route
offers a great deal of potential flexibility.
But in my opinion this also offers an
enormous potential for abuse and future
problems. Think first about the process
of getting anything other than legacy part
23 (or maybe bits of part 25) approved for
use – it doesn’t take a lot of imagination
to predict companies shopping around
for the most lenient authority, or for risk
averse authorities covering their backs by
insisting on compliance with ever escalating
standards that are far more complex
than safety really demands. Also where
varying standards are going to be applied
to different aircraft it’s both going to be
increasingly different to easily transfer
modifications from one to another where,
for example, two aeroplanes now have
different g-limits or have used different
material safety factors. It is also tough
enough now, particularly with an aeroplane
or piece of equipment whose manufacturer
has gone out of business to track the
complete approval basis when keeping
an ageing aeroplane going or designing
modifications to update equipment. In
the future, even the assumptions we
can make now about approval basis
will apparently be pure guesswork.
There are ways in which we can hope
to resolve all of these concerns – detailed
documentation: perhaps within a more
complex new global Type Certificate
Data Sheet standard, for aircraft and
modifications aren’t out of court, and
nor is some joint European/American
directory of approval bases and minimum
standards which would presumably be
co-ordinated by ICAO. If that starts to
happen, then I will be a lot less concerned
about the apparent weaknesses of the new
part 23. However, right at the moment,
I can’t see any of this going on – and
instead we are seeing equipment suppliers
celebrating what they hope will be a much
improved ability to reduce certification
and modification costs: but may well prove
to be the start of a certification nightmare
creating a world full of uncertainties,
inconsistent safety standards and dubious
loopholes that could last indefinitely. I
hope that I’m wrong, but at the moment
I’m just not seeing the clarity of the future
of certification methods and approaches
that we really should have in the wake of
this step change in practices worldwide.
At the moment if, as an engineer,
somebody presented me with a new part 23
class aeroplane to certify, I would as far as
possible be applying the legacy version of
the standard, informed – but not replaced- by best practice and other standards from
elsewhere. Will my views change?, maybe –
but on present evidence, not anytime soon.•
For at least a decade there have been
complaints from across the aircraft design
and certification industry that the baseline
standard for most small to medium
aeroplanes – known to most of us as part
23, has become too inflexible for what it’s
supposed to do now. A lot of this criticism
was certainly valid – a single standard
covering everything from 550kg singles
like the Piper Cub to turbine twins like
the Citation Jet doesn’t make intuitive
sense. Also a lot of the basic thinking
behind part 23 when it was first created
needs to be questioned. Airframes have
proved to have much longer lives than
anybody expected they would back in the
1950s, but are seeing regular demands for
avionics upgrades: probably 4 or more
times in the life of an aircraft now. Off the
shelf technology is appearing from other
industries – materials, engines, electronic
devices – and the mandatory application
of often 50 year old aviation standards
to those new technologies has made it
extremely difficult to get those technologies
usefully onto aeroplanes. So a desire forCertification level: 1 2 3 4
Number of passenger seats: <3 2 – 6 7 – 9 10 - 19Table 1, New part-23 certification levelsVNO MMO
Low speed ≤ 250 kts CAS ≤ 0.6
High Speed >250 kts CAS >0.6Table 2, New part-23 Performance LevelsPart 23 aeroplane approvals – now built on sand?