Brief Summary:
On two occasions (27 May and 21 June) my
Arcus-M engine has refused to start. It then failed to start on many
subsequent attempts the same day and on following days. Suddenly, without
explanation, it starts and runs properly.
Both times the failures occurred on the ground after landing,
when I tried to start the engine to taxi back to the hangar. I flew 13 times (26 starts) and had 250
minutes of engine time between the two failures, and have flown twice (4 starts)
since the second failure. The engine has
never failed after it starts running. It has only failed when I try to start
it. Since no one can explain what caused the engine to fail or to start
working, I do not have high confidence that I won't have a failure again.
Things
we did before the problem mysteriously went away and the engine started:
We confirmed a good battery charge and that we have
ignition. At my mechanic's instructions
we either unplugged and replugged connectors or confirmed other connectors
looked good, e.g. wiring connections at MCU & powerplant operating units in
panel. We confirmed the latest Trijekt
software version. We ran the Trijekt
program to confirm the throttle sensor is getting a good reading and the rpm
reading is reasonable and the air pressure and temperature reported by the
software match the ASOS. We did not
smell fuel, and the problem occurred even the next morning when the engine was
cold and battery fully charged. As
always I started the engine with the throttle at full idle, but also tried
moving the throttle forward a bit, and when I tried starting in redundancy
mode, I had the throttle slightly cracked and moved it forward as the engine
was cranking the way I would with a non-fuel injected two stroke engine.
Suspecting a fuel problem we confirmed fuel pump operation, drained
fuel from both the fuselage and directly from the wing tank (after shaking the
wing) and all looked good. We confirmed fuel pressure. Finally
we changed the injectors on the redundancy system. We did not
change the non-redundancy injectors because they are harder to access and we
thought we'd start with the easiest test. We then tried to start the
engine in redundancy mode. It started, but without explanation, the
engine ran on the non-redundancy system (with the original injectors). We subsequently had both sets of injectors
checked by a qualified mechanic who determined none showed signs of fuel contamination or
clogging. No one understands what changed to make it work.
The first time the engine failed (on 27 May) we changed spark
plugs (which were fairly dirty with carbon build-up) and drained fuel, which
showed some very fine silt like contamination. Still the engine would not
start. Then, at my mechanic's suggestion I tried to start in redundancy
mode and it worked. For unexplained reasons it then also worked in
non-redundancy mode as well. No one understands what changed to
make it work.
Summing
up:
The consensus is that the most likely cause is bad fuel, since I
did discover bad fuel after the first failure and it is the only thing we've
found "wrong". However, the
injectors were clean. It has been
suggested that I change the fuel filters, but that is a lot of work and there
is no clear sign that they are the issue.
It is hard to fix a problem that goes away without explanation. Any ideas are welcome.