I've installed the Dynon EFIS in the panel, and it is wired up to the electrical busses. The normal supply is from the Main Bus. If the Main Alternator fails, I'll shed the Main Bus to reduce the electrical load below 8 amps, the rating of the Standby Alternator. The EFIS will automatically switch to its internal battery, which is supposed to be good for two hours. If desired, I have installed a switch that allows the EFIS to get power from the Battery Hot Bus.

The Dynon EFIS Installation Guide also shows a line for Keep Alive power that would be fed from the Battery Hot Bus. This power serves to charge the internal battery and power the clock when the normal power is not supplied. I elected not to connect the Keep Alive power. The internal battery will be charged any time the aircraft is operated anyway, so I see no need to have it charged by the Keep Alive power between flights. I don't care if the clock keeps time or not. I don't want any unswitched power feeding the EFIS, just in case it ever has an internal meltdown that bypasses its built-in power switch. I prefer to be able to remove power from any item, either by turning off a power switch that is external to the unit, or by shutting down power to the bus that feeds the bus.

I've wired up a 1/4 inch stereo jack to provide a way to connect a PC laptop up to the EFIS. Today I wired up a matching cable that goes from a stereo plug to a DB-9 connector that mates up to the laptop serial port. I hooked a borrowed Windows laptop up to the EFIS via the stereo jack and ran the Dynon EFIS-D10 Support Program. The EFIS Support Program recognized the EFIS right away, and I was able to configure the colour ranges on the airspeed tape.

Next I figured out how to record the data stream that the EFIS puts out. Dynon says to use the Windows HyperTerminal program, but it wasn't installed on the borrowed laptop. So I downloaded HyperTerminal PE. It took quite a bit of messing around with different settings to get it to work, but I finally chanced upon the right ones:

Connect using: Direct to Com 1
Bits per Second: 115200
Data Bits: 8
Parity N
Stop Bits 1
Flow Control: None

I captured the data to a file - it took about 240 KB of data for 1.5 minutes, which should be about 9.5 MB per hour. The data imported into a spreadsheet easily enough, once I had removed the first line in a text editor. There are 64 data records per second, and there doesn't seem to be any way to slow that down to a more reasonable data rate.