There is also a script available which checks the values every few minutes with different gain settings.
But as you never have a stable environment with same amount of planes and messages, i do not think it delivers reliable results compared to a check every 24 hours.
That’s nice, but it’s entirely consistent with what I said earlier. You can’t pass “max” or “agc” as a gain setting. The “–enable-agc” options enables the digital AGC in the RTL2832 which does nothing useful for ADS-B, and does not affect the tuner gain at all.
Again: pass a numeric gain to dump1090 if you are configuring it directly, there is no special interpretation of “agc” or “max”, and --enable-agc is not what you want.
7 Likes
My antenna is +/- 40 feet agl.
“Strong” is, of course, relative.
vb77
In my case the levels are very diverse. From planes flying almost overhead to the ones at 200 miles away, simultaneously. Sure, I tinkered with gain and found an optim setting but I was still loosing some samples.
The only thing that works in my case is a single with higher dynamic range. Airspy in my case.
Surely is much more expensive, so it doesn’t fit everyone’s budget. That’s the compromise one needs to make in that case…
6 Likes
I’ve used graphs1090 to find the optimum gain. This is a gain where the upper RSSI level is around -3 dBFS and the lower RSSI level is around -25 to -30 dBFS and the percentage of strong messages is around 5%. The graphs have been helpful in finding those.
I’m reinstalling PiAware soon and I’m not going to put the graphs back on because it’s not clear to me what local effects they themselves may be causing. With the fresh install and no graphs I think I can find the same information thus:
- Sort the aircraft table by RSSI and watch it for a few moments to see what the upper and lower levels are. I’ve used this approach before and found that the graphed and table upper and lower RSSI levels correlate instantly on any change of gain, so it’s reliable enough to use.
- Use the awk command to display the percentage of strong messages
What time period is the awk command using for its calculation? If it’s 24 hours, is there a way to reduce that to the most recent 2 hours only, akin to selecting the 2 hour view on the graphs? That will enable optimum gain to be found within a day, as one can do now with the 2 hour graphs if testing starts early enough in the day once traffic gets going.
It simply uses “now”. So it’s the same as the graph but only for the moment you use it.
Could you stop writing something when you actually don’t know the answer or you have nothing useful to contribute? Thank you!
It’s using all data since dump1090-fa was restarted.
Alternatively you could use the 15 min bucket in dump1090-fa stats.
There are no longer buckets.
grep last15min instead of grep total in the command line should do the trick.
If you want 2 hours, you’ll have to do subtract two totals two hours apart under the assumption there has been no restart.
2 Likes
Cheers, the 15 minute one is fine for this purpose. I spotted it anyway when I went into stats.json and noticed the epoch timestamps.
I can’t remember if it was this thread I originally posted them in, but I’ve updated the curve fitting equation used in my graph scripts to give more consistent results. They now use a logistic curve, and will only show for the range of data that is present, as extrapolating gives weird results:
https://raw.githubusercontent.com/caiusseverus/adsbcompare/master/daily.sh :
https://raw.githubusercontent.com/caiusseverus/adsbcompare/master/compare.sh :
9 Likes
I would like to open up this discussion again.
Now that i have the FA filter additionally mounted for several weeks, i was playing around with the gain setting.
Following the recommendation of wiedehopf to keep an eye on the messages > -3dBFS i realized that the this number is almost the same independent if the gain is at 42.1 or 49.6.
But what is different is the number of Tracks with single messages.
So shouldn’t it be also required to keep an eye on this number or is this something you cannot control?
Once setting the gain to 49.6 the single message track count is significantly higher while the tracks with more than one message is decreased.
Can someone confirm this or is my investigation just by chance based on the not yet fully operational number of flights?
I was just doing some more gain tweak tests today for the first time in a few months. I think the answer depends on your receiver location and characteristics. There is no one size fits all. For my situation, with the blue FA Pro Stick Plus that has amp and filter already installed, is already optimized for gain of 49.6 for most scenarios. 48 might increase message rate marginally at times for me with no significant impact to range.
In my tests today I came to the same conclusion as last January’s tests. Keep it at 48 or 49.6 to optimize for my situation. Anything below 40 had a immediately noticeable drop in range. My station antenna is surrounded by tall trees/forest and I’ve got hills and big mountains in many directions near and far and very little noise.
I have this stick operating outdoor with a Jetvision antenna. Setting the gain to 49.6 clearly overloads the decoder and i have a non acceptable noise level as result.
For this setup gain of 40.2 works best.
But later i have installed the additional 1090MHz filter between stick and antenna. Now i can go up to 49.6 again without the noise. However it does not give much better results than without filter and lower gain.
I decided now for myself to stop any tests until the major traffic from last year is back. The current situation is simply not reliable enough.
Later i will keep an eye on the single message tracks also trying to find any relation.
Last Feb I did a test with a second filter too… I only used it for a day. Less noise, but no benefit. I even tried a larger and heavier DPD Productions antenna in place of my FA 1090 antenna for a week. Also no benefit. Curiosity keeps getting the best of me…
Fun projects, but I suspect there’s nothing I can do to improve my station range or message count pending future receivers releasing with better hardware. Even that might not help considering the 150’ trees around me. My antenna is already at 35 feet. The forest limits my range north and west to about 100 Nmi and lots of message disruption. East and SE I get near expected range since the forest starts across the street that direction…
That’s my concern as well. I am already on max range based on geography and i think there’s not much hardware i can exchange to improve it.
The next step would be moving to a different location. Not sure what my wife is thinking about that 
3 Likes
Try moving only the antenna, don’t do it with wife.
But yes, sometimes you reach the limit and there’s nothing reasonably to do.
The easiest thing is to raise the antenna, but that brings other problems: lighting, static discharges, maybe even insurance company can say something about it, etc.
Still keep in mind that adding filters in a chain may add attenuation and even if noise is lower there is no benefit.
You may try a better radio with a better noise figure but we are talking about more and more money. It’s up to you how much you want to spend, of course.
No Antenna cable can be that long to move 50+ km which would be required based on my location. I am on my geographical limit. And moving myself together with the antenna would cause another trouble… with my wife 
But thanks for suggestions
2 Likes
If I ever move from here, can you imagine the information that I’ll insist is in the sales pack for this house? “Superb ADS-B location” 
Property includes 300 m radio tower.
3 Likes
If I had a 300m tower, you’d have to drag me out of here, kicking and screaming
1 Like