On Monday, November 19, 2018 at 1:47:34 PM UTC-8, Darryl Ramm wrote:

Andy Blackburn was in involved in that one (against my better advice/pessimism). He is in that thread you linked to. You could ask him where this is at, and details about USA ADS-B interoperability etc., if he does not pop up here to reply.


Thanks Darryl.

This thread already captures the high points of the NMEA Stratux project. Here is some additional color commentary:

I launched the project a couple of years ago and got some initial traction, but then had some issues with pinouts and left it for about a year. At that time I started getting queries from other glider pilots who had found the project on GitHub, so four of us (Andrzej Kobus, John Carlyle, Charlie Gillespie and I) went back at it. By then the fork in the Stratux code was dormant and the person who kindly did the initial coding had lost interest. He did one final round of edits and bowed out.

We did end up with code that successfully modified Stratux to output NMEA sentences more or less in line with the Flarm Dataport Spec. I don't know if this mod works with the current version of Stratux so beware. When I say 'more or less' I mean that the collision warnings don't use the Flarm collision algorithms at all, they are simple proximity warnings that (IIRC) go to the higher alarm values as the target gets closer. It also doesn't de-dupe targets so anyone carrying UAT Out and Flarm (probably no one) or 1090ES Out and Flarm (probably a lot over time - something like 40% of US contest pilots in the latest poll either have 1090ES Out or plan to get it in two years) will generate duplicate targets. As Darryl said, the best plan for that would be to use Stratux for UAT only or have a dedicated display for all ADS-B targets.

There are outer issues when you get down to actual implementation. Some people used a K6 MUX to combine Stratux NMEA with Flarm NMEA so it can be routed to a single display, which seemed to work. The problem with that is that the Flarm Dataport Spec does not contemplate having two traffic sources on the same serial connection, so all the status sentences, such as whether you have good GPS, will either be merged, which will generate unexpected results (e.g. alternating good/bad GPS indications - but no indication which one is which), or you can use the filter function in the K6 MUX to pass status for only one source and never know if the other source is working or not (unless you see a target, which is less than ideal).

There was some other funkiness in the traffic I saw (traffic changing between ADS-R, TIS-B and between 1090-ES and UAT. I suspect this was because they were going in and out of range of my receiver and/or different ground stations, SSR coverage or the service volumes of aircraft receiving ADS-R and TIS-B). We pre-pended two letter codes to each target to identify which band (u for UAT, e for 1090eS) and which service (direct, tIS-B, ADS-r) we were receiving, which was super useful for debugging. These only showed up in ForeFlight. Seems to me that knowing if a target is TIS-B in particular is helpful because the display ought to reflect a much higher position uncertainty.

Andrzej has the source code and, as he said, may take it up. Here is a link to some photos of output on my Oudie and Foreflight.

https://drive.google.com/open?id=0Bw...Eg5ME8zZmNJWGs

As for me, I'm waiting for the new Flarm-licensed devices (e.g. LXNav PowerMouse) that apparently will recognize ADS-R and TIS-B traffic. This will cover the vast majority of my use cases and I think is superior to having to keep PCAS turned on, since it'll give me a rough GPS location for transponder-only and/or Mode C traffic so long as they have radar coverage (or ground station coverage for ADS-R - assuming you have ADS-B Out) along with the proper collision warning and de-duplication that can only be done inside an integrated receiver. Darryl - please correct anything that's incorrect in the above. I just pounded this out and may have been sloppy about some details.

The FAA should've just mandated Mode-S for the US and avoided all this foolishness, but that's spilled milk.

Andy Blackburn
9B