Second follow up:
Nothing heard from Dynon Support on Friday, so I decided to troubleshoot on my own. I assumed that Dynon would suspect a faulty harness, so I verified wiring of my homemade unshielded harness with a Fluke 70 digital ohmeter. Everything checked okay. opened all connector shells and reseated the machined pins then verified integrity again with ohmeter, and reinstalled. The system operated normally for one flight and then I got the REMOTE COMPASS NOT DETECTED message again just after making a radio transmission on 122.9. Rebooted the system, and found that on transmitting, the compass would spin wildly for about 4 seconds then the error message would appear. If I released the PTT button immediately when the message appeared, they system would recover. If the PTT not released immediately, a reboot would be required. The radio transmission was the single factor that caused the error message.
Unplugged the EDC-D10A while EFIS-D10A powered (I have a type 2 probe) and found the EFIS did not loose the OAT indication until the system was downpowered. THis would indicate that an OAT indication is not evidence that the EDC is still connected, only that it was connected since the last reboot long enough for the EFIS to capture the temperature, which then would not change, but would be used by the EFIS to compute TAS & DA. If the EDC/OAT not reconnected, the temp would be lost on reboot.
Tuned the radio to 118.9 and transmitted - no problem - the compass never moved and no error message. Same at 134.9 - no problem, but from about 120.0 to about 126.0 the problem would occur. Hardly noticable at these frequencies, but really bad in the middle where I had experienced the problem.
FYI - I'm using a Bob Archer dipole antenna needed in the wooden fuselage because no ground plane surface is available. The antenna is about 30" from the EDC. The VSWR is in the normal range and the King KY-92 puts about 8 watts into the antenna.
VSWR could not be the problem as it would be lowest at the frequency to which the antenna was cut, and should climb as the frequency became higher or lower. My problem was at mid band where VSWR is lowest
I became suspicious of the OAT probe wiring. I set the EFIS to "N" (no probe) and the problem remained. On my airplane, the OAT probe is mounted on a fairing that is removed for inspections. If I wired the probe directly into the DB9 connector on the EDC, I would "tether" the panel to the airplane since neither the OAT probe or the DB9 connector and harness could pass through the 3/8" hole for the probe.
I needed to a way to disconnect the probe. Called tech support and they affirmed the manual statement that the type 2 OAT cable can be extended. They did not state if shielded cable was a requirement. I installed three "handshake" type connectors and connected to the EDC DB9 using #22 tefzel unshieled wire. Used about 18 to 24 inches between the end of the shielded cable and the DB9. This allowed the probe wiring to be disconnected when the fairing was removed.
This length of the #22 wire happens to be very close to or at 1/4 wavelength for the frequency range that causes the problem. What to do? is the OAT wiring suseptable to RF if not shielded? Does the whole harnes from the EDC to the EFIS need to be RF shielde? (book says no) or do I have a bad component? Inquiring minds want to know?
Keith
