Performance Of Perfect Vision Amp PVLA-30 – Week 1 Results

FlightAware ADS-B Flight Tracking
1

Background & Hardware Change
For the last 4 weeks I have been running a stable basic FlightAware setup with Pi B+ and software from FA with SD card 1.2 download with no software mods, antenna changes, etc. which would impact performance.

On Wednesday 5/13 night after my 5pm PDT date roll, I added a dc blocker, the Perfect Vision PVLA-30 in line amp, 21v power inserter and dc blocker.

**Results – Week 1 **

Self Comparison (BRep007, Site 6930) To Prior Week
Compared my results from Week 1 with PV amp from Thursday 5/14 –Wednesday 5/20 against my results for the week prior without PV amp from Thursday 5/7 –Wednesday 5/13.

Positions: +0%; +2,285 more positions
Aircraft: +20%; +1,725 more aircraft
Positions/Aircraft: -16%; 16 less positions/aircraft

1024×471 83.6 KB

Peer Comparison To Prior Week
Next I looked at my increase relative to 3 nearby PiAware feeder sites (within 6 miles) to base my changes against the average of their changes in order to eliminate any week over week volume fluctuations.

1024×467 80.1 KB

The comparison of my performance week over week versus the average of the nearby 3 peer sites shows:

1024×444 64.2 KB

Positions: Net +2% (my change of +0% over their change of -2%)
Aircraft: Net +16% (my change of +20% over their change of +3%)
Positions/Aircraft: -11%; (my change of -16% over their change of -5%)

Starting on Saturday 5/16, I have been reporting a decrease in positions per aircraft of about -20% (orange shading). Peer sites have experienced a decrease in positions per aircraft of about -5%.

A Potential Causal Factor?
On Saturday 5/16, I claimed my PlanePlotter account (KHND 7653) using the sharing code. In addition, I started running Meinberg NTP in the background on my home pc (not the RPI) to synch network time as indicated by PlanePlotter. Prior to Saturday 5/16, I had been feeding PlanePlotter and have seen the stats on my FlightAware Nearby ADS-B Sites page but it was marked “Unclaimed.” The stats were not integrated into my master FA account before then.

So, I averaged 97 positions/aircraft BEFORE amp, 90 for the first 2 days WITH amp (Thu & Fri), and then 78 for the next 5 days WITH amp and PlanePlotter site claim (Sat-Wed).

Questions For FA ADMIN/Forum Members

1 – Did claiming my PlanePlotter site while running Meinberg NTP (network time protocol) and integrating the feed into my overall FA account somehow reduce my FlightAware tracked positions? Coincidentally my positions started to decrease on Saturday after making this claim.

#2 – Perhaps recent system related issues began impacting the data sent starting on Saturday? I’m grasping, but I was down and so were my peers…

#3 - Are the reported FlightAware stats on the “My ADS-B page” based upon all positions and aircraft I sent to FlightAware or are they based on what I am “credited” with after FlightAware determines validity, stability, etc. of the feeder via some algorithm?

#4 - Would the addition of new nearby PiAware feeders effect my tracked positions and aircraft following the same “credited” logic above if this was the case for Q3.

#5 - Did the Perfect Vision amp eventually overload the dongle (beginning on Saturday) to reduce positions? Seems unlikely after running smoothly for 2 days with amp, but who knows?

#6 - What is the value of sharing PlanePlotter info with FlightAware if I am already feeding data using PiAware? I like some aspects of PlanePlotter as there are some cool features, but their Help, detailed instructions and Forum are below par.

#7 – So am I better optimizing for aircraft or positions? Hmmm…

Any insight into this perplexing decline in positions from ADMIN or other members is greatly appreciated.

Thanks, Brian

2

I have that amp and it’s working well. After installing, I had to change my gain to 30. Otherwise, I was loosing positions, (compared to no amp with default gain setting). With the gain adjusted, I ended up with a massive increase in hits. You’ll have to adjust it and find the best setting for your setup.

3

Thanks, glad to hear! Do you have any step by step instructions to how exactly adjust this? I have never modified anything on the Pi. I just imaged and inserted the micro SD card, plugged it in and that was it.

Brian

4

I use dump1090-mut, and piaware-mut, so I wouldn’t expect the configuration to be the same. For dump1090-mut, the configuration is at:
/etc/default/dump1090-mutability

I’m not sure where the Flightaware provided stuff is configured, however, I do know you can change it.

5

Thanks, maybe someone else will know how to modify the gain settings for the traditional PiAware setup. I did a quick search and did not really understand what to do.

6

Arguments are passed to dump1090 via $PROG_ARGS in the startup script, /etc/init.d/fadump1090.sh

7

Thanks, but would you know how to exactly modify the script to adjust the settings? I am at a loss as to what to do.

Thx

8

Updated Results After Weeks 1 & 2 Combined

**Self Comparison (BRep007, Site 6930) To Prior Control Week **
Compared results from Weeks 1&2 with PVLA-30 amp from Thursday 5/14 –Wednesday 5/27 against my results for the Control (week prior without PV amp from Thursday 5/7 –Wednesday 5/13).

796×799 122 KB

Positions: -3%; -26,985 more positions
Aircraft: +16%; +1,396 more aircraft
Positions/Aircraft: -17%; -16 less positions/aircraft

**Peer Comparison To Prior Control Week **
Next I looked at my increase relative to 2 nearby PiAware feeder sites (within 6 miles) to base my changes against the average of their changes in order to eliminate any week over week volume fluctuations.

795×799 113 KB

Net Change my performance versus the average of 2 Peer sites shows:

795×124 24.6 KB

Positions: Net +9% pts (my change of -3% over Peer change of -12%)
Aircraft: Net +18% pts (my change of +16% over Peer change of -2%)
Positions/Aircraft: -6% pt (my change of -17% over Peer change of -10%)

After Monday 5/25 date roll, I modified the R820T dongle gain changing from “-10” to 30 to help increase positions. After 2 days, my positions/aircraft have increased to ~88 from the prior 5 days ~78.

Background & Hardware Change
I am running a stable basic FlightAware setup with Pi B+ and software from FA with SD card 1.2 download with no software mods, antenna changes (4 leg spider externally mounted), etc. which would impact performance. On Wednesday 5/13 night after my 5pm PDT date roll, I added a dc blocker, the Perfect Vision PVLA-30 in line amp, 21v power inserter and dc blocker.

**
Questions:**
I arbitrarily set the dongle gain to 30 based upon another members recommendation within another post with similar amp setup. Is there a more scientific way to identify optimal dongle gain for my setup versus trial and error?

What’s left to optimize performance other than moving the amp to the base of the antenna which (entails a trip to the attic and laying in insulation)?

Thanks, Brian.

9

User tjowen suggested:

Just a couple days ago, I added back in a filter and started adjusting as he suggested. I ended up settling on a gain of 44. I’m getting about 10% in the above -3db and I’m happy with it. Lowering the ratio cost too many positions overall. Every system is different, so you have to find what your system needs. There is also a thread on adding graphing to gain better feedback for adjustments.