ADS-B latency?
This thread on another forum: https://www.reddit.com/r/flying/comments/53ihoi/came_within_100_to_300_of_a_midair_this_weekend/
The poster says that there was a delay of several seconds (from the sounds of it, at least 10 seconds) between reality and the position shown on his ADS-B display.
Does ADS-B really have this much latency, as designed? Note this is in the US where to radiate ADS-B you must have a certified GPS source. If there’s this much latency in ADS-B equipment, how can it be useful for its intended purpose?
Or does someone have a bad installation/fault?
ADS-B specifies a maximum total latency of 2 seconds, but for TISB, which is generated by radar detection of mode C targets and broadcast from a ground station, it really depends on the number, type, and update rates of the radar systems, which can be 5 seconds in a terminal radar and twice that in en enroute radar. ADS-B is not a traffic system, it is a surveillance system that can aid in the visual acquisition of traffic.
NCYankee wrote:
ADS-B specifies a maximum total latency of 2 seconds, but for TISB, which is generated by radar detection of mode C targets and broadcast from a ground station, it really depends on the number, type, and update rates of the radar systems, which can be 5 seconds in a terminal radar and twice that in en enroute radar. ADS-B is not a traffic system, it is a surveillance system that can aid in the visual acquisition of traffic.
Which is not relevant to Europe where there is no TIS-B. I see almost no latency and while 2 sec is the total latency permitted that is corrected by the systems somewhat using predictive technology. However I would not be surprised if that is limited in aerobatic aircraft which for example may not have a consistent GPS fix.
I see no reason why sh*itty installations should be the sole province of active TCAS systems like mine, which has a range of 15nm (to spec) to the left and right, about 5nm behind, and about 1-2nm ahead (12 o’clock).
I reckon the loss of range ahead is due to the curvature of the roof ahead of the upper antenna
but it could be something else…
Similarly sh*itty transponder installations are endemic in GA. I nearly always pick up light jets etc at the full 10-15nm range at 3 or 9 o’clock but many old GA planes don’t show till say 2nm. Or less than 1nm at 12 o’clock.
And since ADS-B uses the transponder antenna…
ADS-B is not a traffic system, it is a surveillance system that can aid in the visual acquisition of traffic.
It does make me wonder whether it will ever be worth paying $5000 for the ADS-B IN upgrade to my TAS605. Surely it should work better – because it is picking up signals transmitted by other aircraft.
Peter wrote:
It does make me wonder whether it will ever be worth paying $5000 for the ADS-B IN upgrade to my TAS605. Surely it should work better – because it is picking up signals transmitted by other aircraft.
I can receive from 50-100nm away with a GDL39 in the back of the cockpit, not even on glareshield. The direct ADS-B signals work very well.
Peter wrote:
Similarly sh*itty transponder installations are endemic in GA. I nearly always pick up light jets etc at the full 10-15nm range at 3 or 9 o’clock but many old GA planes don’t show till say 2nm. Or less than 1nm at 12 o’clock.
You have a strange antenna positions in my opinion, that might cause this issues. The power output of a business jet or airliner is often higher then high end GA, and even more then low en GA. There is quite some dB’s difference between in tolerance low end with poor coax transponder and high end low loss coax transponder of an airliner. This is also why the educated guess of distance, based on signal strength doesn’t work well.
The only good thing is that fast aircraft generally have higher performance transponder, and show closer then a slow flying aircraft.
A better TAS605 antenna installation:
This doesn’t have the obstruction in front from your combined GPS/Com antenna
I went through that exercise, @wigglyamp, when debating with the installer where to put my antennae.
I have two GPS antennae already which N709AM did not and does not have. Hey, that makes him illegal because he doesn’t have duplicate nav 
See this … search for “scallop”. It appeared clear that there is no legal way to mount any antenna on the TB20 GT composite roof other than in the scalloped-out areas. When I was going N-reg in 2005, the DAR (the late Manfred Herweg) was very fussy about this and made us remove the ceiling trim and check. I dare say £2k to a DER would deal with it but that is another 2k the customer has to pay… and you have about 2cm of composite in that roof if you go elsewhere.
The correct solution thus involves moving the one VHF antenna (COM1/NAV1) to the underside.
I was dealing with a less than convincingly competent installer (read the whole article if you like) and wanted to minimise the re-wiring. Especially antenna re-wiring for which they wanted to use cheap coax so I bought and free-issued them with 50m of RG400!
You may well say I should have used the other company but they wanted 2k for a DER package for mounting the TAS box in the luggage compartment! Later they agreed with me that wasn’t actually necessary but it was too late. So I ended up with a bodge job, which seems to work but possibly the way the ground plane connection was bodged is not helping the range.
We are all where we are…
Anyway this is getting off topic for ADS-B. Am I right there is no justification for a late warning with ADS-B unless at least one of the two installations is bodged?
I have the TAS620 in our PA28 with a clear view forward from the top antenna and agree that the warning from head-on traffic isn’t great. Yesterday a target popped-up 5 miles directly in front but we didn’t get the TA until about 1 mile which is hardly the minimum 30 second warning when closing at 220 knots! Recent experience of the Garmin GTS800 in a military helicopter project shows it to be much more accurate on bearings and warnings.
That’s an extremely useful data point.
Maybe the propeller is affecting it. I have had numerous late warnings at 12 o’clock, at 1-2nm.
The bearing is accurate enough though.
Would ADS-B be any better?
