From: "Stephen Izett stephen.izett@gmail.com" <flyrotary@lancaironline.net>
Received-SPF: pass
 receiver=logan.com; client-ip=209.85.215.178; envelope-from=stephen.izett@gmail.com
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.11\))
Subject: Update on EMI problem.
Message-Id: <CB6B7982-C89B-46ED-A50C-CEFD5D5851A8@gmail.com>
Date: Wed, 15 Jan 2020 12:23:18 +0800
To: Rotary motors in aircraft <flyrotary@lancaironline.net>

A frustrating few hours

I got down to the hanger this morning to look into the EMI in both the =
engine controller and EFIS.

First thing I did was reroute the radio coax so that instead of being in =
close proximity to many cables down the centre of the aeroplane instead =
down the very right hand side.
This increased the distance to various wire looms by ~450mm.

Prior to this change the symptom outside the hangar was from memory =
about a 300ft/min vertical speed on tx in the Skyview EFIS.

First test with radio tx after the reroute of coax inside the hangar =
yielded a "4 times worse" symptom in the EFIS! ~1300ft/min reading on =
tx.

I then tried the test outside and away from the metal hangar by 10m and =
the symptom was reduced to ~600ft/min and increase in symptoms of about =
100% on original problem.
I then applied a clamp on ferrite to the cable loom entering the ADHARS =
unit from where the symptom originates.
On tx outside the hangar again the symptoms are far reduced by over 1/2.

So if I had left the radio coax where it was, the ferrite may have fixed =
the problem in the EFIS.
I have more ferrites on order and they appear to be very effective, so =
will use these to attenuate the RF entering the various pieces of =
equipment.
EMI trouble shooting is indeed a black art!!

I also removed the Engine Control unit and soldered 0.1uF ceramics =
across the outputs to earth of the MAP sensors of both A and B computers =
as specified by Tracy.
On reinstalling the Engine Control unit I was not able to start the =
engine!
It would fire but not run on either controller.
So spark and fuel are present, but not to remain running.

So the mod appears to have messed up the MAP sensing.
My initial thought was that the caps were actually resistors packaged =
wrongly but they measure open circuit.

So the few hours of mods and testing has been exceedingly frustrating, =
going backwards at every turn except for the 1 ferrite showing itself to =
be very effective.

I=E2=80=99m hoping my next outing will be very much more fruitful.
In the mean time I=E2=80=99ll ponder what could possibly cause both =
Computers in the Engine Controller to be u/s.
If I cannot discover my deliberate mistake. I=E2=80=99ll remove the caps =
and retest the Controller on Friday.

Cheers

Steve