Recommendation

We have studied the BASI air safety interim recommendations as
notified to us in ref letter. Before deciding upon any mandatory
continued airworthiness action, if any, we have addressed a number
of questions to Saab Aircraft AB for their consideration and reply
to us before Aug 2, 1999. Copy of our letter dated June 15, 1999 is
enclosed for your information.



Please regard this information as an interim reply to your
recommendations.

Further actions will depend upon Saab reply and our final position
will not be available within 60 days as requested but hopefully
late August.



SAAB 340. Request for investigation of modifications or procedures
to reduce the probability of stall in icing conditions



Ref.: BASI, Australia, "Release for Air Safety Interim
Recommendation IR990072, IR990073, IR990074, IR990075 and
IR990076".



LFV have studied the referenced recommendations, which you also
have been informed of. Although that SAAB 340 has been correctly
certificated according to the requirements and their
interpretations, applicable at the time, the concerns of BASI seems
well founded. It is evident from many other cases that pilots may
not follow correct procedures during high work load, when they are
subjected to unusual situations, when procedures in or
characteristics of another aircraft they have been flying earlier
are slightly different, when they are complacent, etc. It seems
obvious that the pilots involved in the Australian incident (to
VH-LPI on Nov 11, 1998) did not follow the AFM Normal Procedure: If
it is not certain there is no ice accumulation on the aircraft; or
if ice accumulation is observed on the aircraft, maintain an
airspeed .... not less than 1.4 times the stall speed in any
configuration." In this case the pilots apparently knew they were
in icing conditions.



Normally we are hesitant to require retroactive actions on one type
of aircraft that should also apply to other aircraft of similar
design. Such actions should be coordinated within JAA and possibly
with other authorities. We understand that such actions in the
icing area already are ongoing through the Ice Protection
Harmonization Working Group, IPHWG, with participation of JAA, FAA,
Transport Canada, and industry representatives.



However, notwithstanding that we believe that an action normally
should be coordinated within JAA to include all aircraft of similar
design and that the SAAB 340 AFM procedures in icing conditions
recently have been changed (i.e. a lesser amount of ice accreted
before turning on the de-icing boots, and to consider using
continuous mode to reduce pilot work load), we hereby request Saab
to provide the answers to the following questions. Our request is
based on the fact that at least one modification that possibly
might reduce the likelihood of a repeat occurrence of the
Australian incident already is available as an option, i.e. the
"ice speed button".



1. What modifications to Saab 340 or additional procedure changes
in order to reduce the probability of an icing incident similar to
that encountered by VH-LPI, taking into consideration pilot work
load and possible pilot situation misjudgement, could be rather
easily retrofitted to the aircraft? The "ice speed button" is
assumed to be one of them.

2. Do Saab believe that any of the studied modifications or
procedure changes might reduce the probability of an icing incident
similar to that encountered by VH-LPI? Could the modifications or
procedure changes still result in the same or other unwanted
effects? If so, what is the conclusion?

3. What modification or procedure change, if any, of those
practicable, would according to Saab constitute the highest degree
of improved safety, and the highest cost-effectiveness
respectively? Cost of retrofit to Saab and operators?

I hope you received my letter dated June 29, 1999 with copy of a
letter to Saab dated June 15, 1999. We have now received Saab
answer to our questions and are in the process of reviewing them
together with our Flight Operations section.



You have also received Saab comments on recommendation IR990072.
LFV supports in principle these Saab comments. (Letter dated 3 Aug
1999). Our goal is to take a decision with regard to the line of
action mid-September.

BASI Air safety interim recommendation IR9900765 has now been
considered by LFV and we have reached the following position.



Although the "ice speed button", if used, probably would have
contributed to preventing the stall incident on 11 Nov 1998, a
mandatory requirement of implementation of this modification, as
recommended in IR990076, is not supported by LFV. There are
operational drawbacks as shown in Saab Aircraft AB letter 72DSS0957
to BASI, dated 3 Aug 1999. This may result in the crew not using
the button in icing conditions assumed to be light. Then, if they
do not keep the AFM normal procedure in mind ("If it is not certain
there is no ice accumulation on the aircraft, or if ice
accumulation is observed on the aircraft, maintain an airspeed of
not less than VREF + 10 knots for landing and 1.4 times the stall
speed in any configuration."), this may still result in stall
incidents.



As you know, FAA NPRM 99-NM-148-AD proposes activation of the
deicing boots at first sign of ice build up on the airplane. This
eliminates the need for the crew to make judgements on when to
activate the boots and at the same time reduces the amount of ice
on the aircraft. This will reduce the likelihood of non-observed
speed reductions due to drag increase and stall without warning.
This in combination with the already existing AFM text cited above,
and pilot understanding of the reasons behind, would significantly
reduce the probability of future stall incidents in icing
conditions with the SAAB 340, we believe. The ice speed button,
although also an acceptable approach, does not appear to be the
ultimate remedy. Basic pilot knowledge and skills will always need
to be applied.



Increased stall speed due to airframe icing is intrinsic to all
aircraft. Any retroactive requirements that would significantly
increase safety for all types of aircraft in icing conditions
should therefore be coordinated world wide and not just be applied
to one type. This is another reason we do not support the "ice
speed button" as a retroactive requirement for the SAAB 340. On the
other hand, we welcome the FAA NPRM, which appears to solve a
problem for all modem aircraft with de-icing boots with limited
cost impact, mainly higher maintenance costs due to more frequent
use of the boots.



A change implementing the FAA proposal, based on the new knowledge
that ice bridging does not occur on modem type boots, will be
introduced in all LFV Approved Airplane Flight Manuals (AFM) and in
the manufacturers Airplane Operation Manuals shortly. On the FAA
side this change will most probably be addressed as an
Airworthiness Directive which in the U.S legal system mandates the
procedural change. Under the Swedish regulations operators are
required to apply the latest version of the AFM.