NTSB Factual Report

On June 9, 2001, about 1000 central daylight time, a Packwood Packwood 1, N2402F, piloted by a private pilot, sustained substantial damage on impact with a fence and terrain during a forced landing after takeoff following an in-flight loss of engine power encountered on climbout from runway 18 at the Monett Municipal Airport (M58), near Pierce City, Missouri. The personal flight was operating under 14 CFR Part 91. Visual meteorological conditions prevailed at the time of the accident. No flight plan was on file. The pilot sustained no injuries. The local flight was originating from M58 at the time of the accident.

During a telephone interview, the pilot stated "...the engine was running great during the takeoff. The engine quit about 300 ft AGL. He had tress ahead and performed a forced landing on the remaining runway. He went off the end of the runway and impacted fencing. He initially thought that the fuel pump had failed. He said that he did not have time to activate a back up fuel pump. After the accident he tested both fuel pumps and they both checked out ok. He shipped the Ellison throttle back to the manufacturer, Ellison, to be re-built. He said that Ellison indicated to him the unit was recalled 12 years earlier. He was not aware of the recall at the time of the accident."

The TBI manufacturer stated:

At your request we have completed our standard

preshipping acceptance test on Throttle Body Injector

Model EFS-2, S/N 1039.

Our pretest inspection of the unit revealed that the

regulator seal had been removed and the EFS-2 had been

completely disassembled and reassembled by someone who

lacked training in the correct procedure for installing

lock wire. During the bench check of its regulator

cracking pressure and idle setting, it was determined

that the inlet control valve was stuck in the closed

position, a condition not unusual for a unit which had

been run on automotive fuel, or one that has been stored

for a period of time with standing fuel inside.

Additionally the idle needle valve was found to be fully

closed. When the regulator cover was removed to free in

inlet control ball, water stains were observed over a

large sector of the control diaphragm. Removal of the

inlet fuel screen revealed that it too had experienced

substantial long term water contamination some time in

its past.

Following reassembly, the subject EFS-2 was installed

on our 1835 CC Volkswagen test engine which was

instrumented for measuring both fuel flow and air flow,

following which the test engine started and idled

normally. After a suitable warm up period the engine

was operated through its full range of throttle openings

with ten data points taken ranging from idle to full

throttle.

Data readings of fuel flow and airflow obtained during

this test were plotted to reveal a metering curve

indistinguishable from a new EFS-2. (See appended

performance chart)

At 0953, the reported temperature and dew point at Joplin, Missouri, were 24 degrees Celsius and 19 degrees Celsius respectively.

The temperature and dew point were plotted on a Transport Canada carburetor-icing chart. Their intersection fell in the moderate icing-cruise power or serious icing-descent power area. (See appended icing chart)

An excerpt from the TBI manufacturer's web site stated:

Contrary to common belief, the ELLISON TBI can accumulate

ice! Additionally, engines with cold induction manifolds

such as the four cylinder Continental engines, the

Continental O-470, and all Volkswagen derivative engines

are especially susceptible to the formation of manifold

ice. FOR THESE REASONS, ALL TBI INSTALLATIONS MUST

INCLUDE AN INDUCTION HEATING SYSTEM capable of

providing an inlet air temperature rise of 90 degrees F. Follow

engine or airframe manufacturer's recommendations for use of

induction heat. ...

The Ellison Throttle body Injector (TBI) requires clean

fuel for proper operation. A 70 micron filter is

required to keep contaminates out of the TBI. The TBI

does have a last chance fuel strainer installed on the

fuel inlet fitting. This strainer is there only to trap

large particles in the event of a primary filter

failure. The strainer is fine enough to reduce the

possibility of particles plugging the fuel metering

tube but particles which can pass through the fuel inlet

screen can keep the valve in the fuel regulator from

closing completely. This can result in an erratic idle

and fuel leakage after engine shutdown. Many throttle

body installations use a standard gascolater with

screen before the fuel pumps to prevent large particles

from damaging the fuel pumps and to provide a low point

water drain, but then after the engine driven pump, a 70

micron filter or finer must be installed.

This filter must be carefully selected. Some paper filters

will not pass water and could cause fuel system blockage if

filled with water. A drain must be provided and it must be

drained before each flight. If this filter is installed on

the pressure side of the boost pump it must have sufficient

strength to handle pump pressure without bursting or leaking.

The Glassair III uses a Fram HPG 1 with a low point water

drain added which is a very good filter housing and

perfectly capable of operating with pressurized fuel. As

with all fuel system components, the filter must be located

well away from sources of heat, preferably outside of the

engine compartment. If located in the engine compartment it

must be blast cooled to prevent vapor formation.

Subsequent to the accident, the TBI manufacturer, Ellison, stated that "we will be including information on our website about the factory recall of early-model EFS-2's as well as copies of all service bulletins which have been issued on any of our products."

NTSB Probable Cause

an engine failure during initial climbout and the fence struck during the runway overrun.