Here is your Flower Pot Spider-Stick
The Spider was NOT built by me. I found the photo of Spider by Google image search.
In this photo, I added PL259-SMA adapter and the Pro Stick, using Photoshop
Current setup is RPi B with Nooelec 820T2, and a 15dB LNA powered from the RPi GPIO header. Have also had a AusPi wireless dongle on the RPI at the same time with absolutely no issues.
Increased ‘gain’ in the RTL dongle doesn’t translate to any significantly higher power consumption. A common cause of power issues is voltage drop from long USB runs or the power supply simply doesn’t have enough amperage to run the RPi and the RTL dongle. That is well documented.
Coat the antenna and the coax connector in liquid electrical tape, hide the coax a bit better, and it would be nearly impossible to see from more than a few feet away. Even just the spider mounted up on the roof would be pretty hard to see if you weren’t looking for it. http://i1296.photobucket.com/albums/ag9/kc0rzw/Mobile%20Uploads/image_zpsofcwtohz.jpg
As I mentioned in my next reply to that discussion, I was just relaying what I’ve read over and over again on multiple sites for over a year now. Checking for future replies first never hurts.
From what I gathered, many of those folks were using 5v / 2000-2500mA power adapters with no long USB runs. Some things related to Pi USB power that works for some folks doesn’t always work for others. That’s also well documented.
I never tried PoE or long USB cable.
I will first try different lengths of USB cable between Dongle & Pi, and see beyond what length trouble starts.
Next I will plugin dongle directly in the Pi, and try different lengths of CAT5 with PoE, and again see at what length trouble starts.
This will be only an experiment. I dont need long runs of anything. My indoor Cantenna is about 8 ft above the floor, and RPi is just below it on the floor. The router is about 15 ft from the Pi. Power Socket on wall is 3 ft from the Pi.
My biggest problem is the lack of precision from MLAT due to the difficulty of precisely timestamping when Mode-S signals arrive. Using ADS-B signals when present to synchronize multiple PiAware stations that can see the same ADS-B plane at the same time is ingenious, but with CPU clock drift and variability in buffering by the SDR dongle the solution seems inherently limited. Does the ProStick somehow address this basic limitation?
What would be really cool is a Pi Hat type of board with both the SDR chip and a GPS chip, like AdaFruit’s “Ultimate GPS” (adafruit.com/products/746), so that the signals could be hardware timestamped and you get precise station location as a bonus. It seems like such a board could be produced <$100 in quantity. I’d even buy two extra for other PiAware stations in my area just to have the better MLAT results.
GPS-timestamping incoming samples is a good way to do it (e.g. the Radarcape does this) but it is not just a case of attaching a GPS chip to an existing receiver. You are looking at a FPGA or a custom chip for the digital side of the receive chain, which is immediately quite a bit more expensive than using a RTL2832 where the design/fabrication overheads are a sunk cost that someone else has already paid. I would be surprised if you can do it for <$100.
The prostick is no different to a regular rtlsdr dongle when it comes to how mlat is done.
Planefinder are releasing a new FPGA based receiver, and that’s going to cost £300. It does have good performance for mlat though - this is apparently unfiltered:
It’s really not ‘black art’ and is quite repeatable. I choose to go with what I myself have tested to work, rather than anecdotal reports. RPi B uses about 600 ma, and by default provides about 500 ma through the USB if fed by an appropriate power supply.
The Nooelec RTL dongle uses about 280-300 ma. My 15dB LNA uses about 30 ma. Not sure what other ‘power hungry’ devices one might hang off a dedicated ADS-B RPi, but if it impedes performance, then the issue lands purely with the ancillary device and the user, not the RTL dongle or tiny amplifier.
RTl dongle receiver gain factor in itself is not an appreciable variable power consumer in the scheme of the system. Possibly, a properly set gain might result in more planes tracked, in turn requiring more work by the RPi CPU if certain parameters are enabled, but this is hardly a ‘fault’ of the dongle or RPi, and would not result in excessive power being drawn if all other power supply parameters are met.
Inline satellite amplifiers are noisy and not designed for weak signal boosting, but instead, to bring up the level of already optimum LNB output to feed a whole house system or long coax run to a distribution amp.
Glad someone posted this, and that I have been following this thread closely, didn’t want to be waiting any longer than needed to get my dongle and filter… Didn’t get a notification either, but not hassle
UK users need to be careful of the post office adding an import tax (VAT?). I got clobbered when I ordered a couple of FlightAware filters from wifi-expert.
I’ll wait until they are available from a European supplier.
Not sure where that came from, but I’ll go with it…
That’s all true - although many people see a considerable spike in coverage area and overall performance when using one at the base of their antenna. I’ve messed around with a Perfect Vision in-line satellite amp in the past and saw phenomenal results vs. no amp at all. That was many months ago, though. I’ve been using Mini-Circuits’ ZRL-2400LN+ amps for quite some time now… They’re $140 a piece new - I picked up four of them un-used in a surplus eBay auction for $200 total.
The pictures below show how I am going to use the “Pro Stick + FA Filter” when I purchase it.
Since the Pro Stick has an RF Amplifier, It is advantageous to add a filter between antenna & the stick.
I am lucky that I live i an apartment, with no balconies.
As a result, my installation is strictly indoor and need not be weather proof.
Finished photos are done in Photoshop, by combining the individual photos of Filter, Pro Stick, Can, Whip, Adaptors, and Spider.
Spider is NOT made by me. Found Spider’s photo by Google image search.