Where next~? (Hardware upgrades)

One more thing about only looking at bits is that this can be very misleading when comparing radio systems. The bits refer to the size of each sample but if the radio has a lot of internal noise you can’t use some of the bits. In radio terms the extra bits are noisy are not effective.

Another way of thinking of it is that you have ENOB number of bits that are measure your signal and the rest of the bits measuring noise. These noisy bits are bouncing on and off randomly due to noise and your radio is measuring the noise accurate to 1 to 5 bits :). (There is a little part of signal in the noisy bits but require special processing or tricks to get some signal out of them.)

The RTL dongles are 7bit ENOB with the last bit bouncing around due to noise.
The other radio systems that say they are 12 bits or 14 bits might actually only have a 8-12 bits of ENOB.

All the bits in the sample have to be processed. Having many noisy bits increases the CPU requirements while also not giving you anything in return. This is why you sometimes see 14 bit radio perform poorly since they have 4-5 noisy bits with only 10bits ENOB.

1 Like

AirSpy ENOB: 10.4
RSP1A ENOB: 10.4

Both 12 bits, but the RSP1A sampling below 8MHz is 14 bits. “The 10.4 ENOB was specified for 8 MHz sampling and includes non-linearities, and spurs as well as quantisation noise.”

ENOB Calculation:
ENOB is based on the equation for an ideal ADC’s SNR: SNR = 6.02 × N + 1.76 dB, where N is the ADC’s resolution. A real world ADC never achieves this SNR due to its own noise and errors. You can rearrange the equation to calculate an ADC’s effective N, or ENOB as we commonly call it: ENOB = (SNR – 1.76)/6.02 dB.”

1 Like

This is my airspy feeder.
It uses the hab/nevis amp and a cavity filter.

It is only in my attic so doesn’t have the antenna height of my other feeders.

310,326 positions in the <50 miles? Wow… Lots of “birds” there in NY.
I barely pull 40k.

Yes. This one get 500k

It is my radsrcape…

Well, that’s a… beast :smiley:

I get well over 600k usually within 50k range. I live within 50 miles of SFO, OAK and SJC. I suspect though that most of the count within 50k is grounds crew circling the airports in their cars. I get much lower counts on FR24 on the same feeder. Wonder if I am doing something wrong. Only been doing this for a month and a half or so. still learning

Have you looked at the threads for optimising the gain settings?

Other threads regarding the need for a filter. Maybe try the new, narrower band FA filter.

Maybe state your setup so we can have an idea of that it looks like.
RPI1/2/3, Dongle type, Coax type, amplifier, antenna type, Filter, if any etc.

Can you move dongle closer to the antenna?
etc

Hardware upgrades include:-
A better dongle- airspy, FA dongles, Mode S Beast, Radarcape etc
A better antenna and antenna system. LMR 400 coax, FA or DPD antenna.
An amplifier/filter like the hab/nevis or over inline amps.
Better filters, New FA filter or Cavity filters (lower loss, narrower bandwidth but much more expensive).
RPI upgrades (obvious)

This looks very good to me.

FR24 has different statistics. I wouldn’t see that as a reason that something is wrong.

Anyway if you are asking for hardware upgrades you should in as much detail as possible describe your current setup.

My words, you have 1.1 million now!

You have 210 airplanes at 250+ miles away and 6900 planes in the 200-250 miles range. I think your system is OK as it is. I would be happy to get those numbers.

You probably just have a tremendous amount of planes on ground.

What improved my reception sensibly where:

  1. AirSpy Mini receiver. Awesome Dynamic Range, you might benefit form that.
  2. Antenna height, to clear some of the trees around me. You seem to do fine in that aspect.
1 Like

My setup is as follows:

RPI 3 to Orange FA dongle, to a Uputronics SAW Filtered Pre-amp. Then 5 meters of cable to a roof-mounted antenna Jetvision A3. The gain is set at 22.9. Still get about 20% of messages > -3 dBFS over a 24 hour period. Planes I can see up till about 240 nm to the north. The south is mostly blocked due to hills and roofs (including my own).

Generally I am happy with how it is running now. If the count for FA is different than for FR24 then this setup should be okay. A Cavity filter I would like to get but Jetvision seems to be out of stock.

1 Like

Unlikely that a cavity filter will provide much improvement.
The Uputronics is already quite good at filtering unwanted signals.

You could always change to a blue Flightaware ProStickPlus which has yet another SAW filter :wink:
Or you could try a rtl-sdr blog v3 dongle. Those might be just a tiny bit better compared to the Flightaware sticks but that is just my personal speculation. (You don’t need the internal LNA of the Flightaware sticks as you already have a strong external LNA)

Instead of investing into a quite expensive cavity filter, the airspy mini for sure would be a better upgrade.
With the rtl-sdr derived dongles you will always have to trade reception of strong for weak signals. That is quite a bit better with an airspy.

@wiedehopf Thank you for sharing your thoughts. That is helpful.

An airspy would require a 3B+. My 3B can’t cope with it.
It would have a much greater dynamic range and you already have the amp to go with it.
I was using an Odroid XU4, however, it has problems and I haven’t had time to fix it.

1 Like

You have more aircrafts/messages than are at his location, this changes CPU load quite a bit.

I vaguely remember we chatted about that already but i don’t quite remember the exact config options you ran with it and if it was the most current binary.
Also if you follow the instructions on the airspy web page and limit the RPi to 800 MHz that could explain the limitation.
If you just put on a heatsink or use a fan and not throttle it, it works much better.

1 Like

I had a look at your nearest neighbour feeding FA

and it is apparent that about 1/3 of the positions you report are to your North West and less than 40NM. Is this the ground traffic you mentioned?

You also have almost no positions South of East whereas @gecarey has quite a lot.

If it is possible, I would suggest re-positioning your antenna so that your roof shields that very high count from the NW and enables reception from the South East without shadowing the North.

Are you trying to maximize position reports or number of aircraft?

If you are trying to maximize number of aircraft I would ignore the Still get about 20% of messages > -3 dBFS over a 24 hour period and increase the gain a bit. All the positions that are then not counted because they are too high a signal strength will have already been counted when they are further away. The danger is that there will be too many high level signals close in that will block the receiver. It is a bit of a black art juggle to get the balance just right.

Moving the antenna to the other side of the roof will then use the roof to attenuate those very loud close in signals.

As suggested, using a receiver with better dynamic range may also help but there is more you can do to optimise your current setup.

S.

1 Like

Okay.
So, I will get an AirSpy. I have a 3B+. Just need to tune it then to the Airspy. It appears I don’t need a separate power source for the Airspy but correct me if I’m wrong (the Uputronics is plugged into a Apple iPhone charger that delivers 5v).

My location is such that I have the hills to the south and to the west limiting my view options (for those of you familiar around here, I live several blocks south of the Hillsdale mall in San Mateo).

This forum has been greatly helpful for my learning. Thank you all for your comments on this and other topics of discussion.

2 Likes

The Airspy will be able to power the uputroncics amp via bias-t.

And it is apparent that about 1/3 of the positions you report are to your North West and less than 40NM. Is this the ground traffic you mentioned?

I was being a bit facetious. There are definitely plenty of lorries driving around the SFO airport. They show up with Flight number OPS23 or some other number at altitude GROUND. At busy times there are up to 5 or so but at those times I have 200 planes on my local radar. The point is, that the ground operations lorries do count towards positions locally. Not sure they get uploaded though to FA.

@jonhawkes2030
I would not need a separate bias tee then, and put the uputronics up the mast near the antenna and power directly from the Airspy? Interesting. Then I would need to find a pole mount box to fit the uputronics.