Would 2 pi’s run off 1 dongle ok using something like this?
What about a regular usb hub. Dongle plugged into the hub, type A male to the RPi3 and micro usb from the hub to the zero??
No, you can’t use a dongle (or other USB device, for that matter) connected to two different USB hosts.
1 Like
I get better performance from my Pi0 W with the MakerSpot Accessories 4-Port Micro USB OTG Hub than with the USB OTG Cable which was included in the CanaKit starter kit.
Ok, so I’ve decided I’m going to set things up to run the RPi3 and the zero at the same time, it is the only way to truly try and figure out why my results differ so much from everyone else’s. A splitter is easy, but what to do with a 2nd dongle??
I have been considering upgrading the dongle to a FA one. Ideally I’d go for the pro stick (orange) + a filter bc I"m not entirely sure I need a filter given the ranges I am getting. I do have a giant radio tower nearby though so I suspect I need something. My thinking is the the orange stick gives me the most options. Anyway, getting 2 pro sticks + filters or even 2x pro stick plus’s is an expensive solution at the moment.
The dongle I have is cheap, it is only R820T, but I am getting good ranges but is it part of the problem?? Does anyone know how I can check the specs of this dongle to confirm it is only a R820T? Is there a command to do this?
you do not need a filter that matches your range but a filter that matches the noise at your site. over there in Australia you have cell phone 960 MHz - if there is some in your range the standalone flightaware filter is really really bad at this frequency. the problem with the dongle/amplifier/filter combos like fa pro stick plus is - they amplify/filter at the wrong end of antenna run and the position their filter is used is suboptimal.
before buying new equipment first see what kind of noise is at your site - see this post: Filter not producing desired result - #11 by obj
p.s. if you are looking for an excellent filter/amplifier then go for this one: 1090MHz ADS-B Filtered Preamp Anthony from uputronics ia very friendly person - maybe he sells you one without case - this way shipping to Australia is cheaper and faster. and what I saw in your pictures you are a master in homemade cases anyways 
p.p.s. within rtl_power you can see some specs from your dongle too Rtl-sdr - rtl-sdr - Open Source Mobile Communications
p.p.p.s. best is one of the cheap dongles with R820T2: Rtl2832u R820t2 FM SDR DAB Digital Dvb-t HDTV TV Tuner Receiver Stick for sale online | eBay
1 Like
Thanks Tom,
I just tried running rtl_power but it looks like maybe my dongle isn’t compatible. I had tried gqrx before too and couldn’t get that to work either. This is what I got;
pi@raspberrypi:~/rtl-sdr/build $ ./src/rtl_power -f 800M:1200M:100k -i 30 -c 50% -e 30m -g 30 -F 9 >scan.csv
Range: 800M:1200M:100k
Mode: hopping
Number of frequency hops: 500
Dongle bandwidth: 1600000Hz
Downsampling by: 1x
Cropping by: 50.00%
Total FFT bins: 8000
Logged FFT bins: 4000
FFT bin size: 100000Hz
Buffer size: 16384 bytes (5.12ms)
Reporting every 30 seconds
Found 1 device(s):
0: Realtek, RTL2838UHIDIR, SN: 00000001
Using device 0: Generic RTL2832U OEM
usb_claim_interface error -6
Failed to open rtlsdr device #0.
pi@raspberrypi:~/rtl-sdr/build $
EDIT
Re; dongle/filter vs range, what I meant was that I’ve been reading that R820T2 has much better performance but I seem to be doing ok with the R820T. I don’t know whether to get 2x T2 dongles or just get another one the same as I’m using.
Also, the interface error -6 has got me intrigued. Trying to find out more about that.
try this before rtl_power:
sudo killall dump1090-mutability
or dump-fa - don’t know what you are running and how flightaware dump is named find dump name with one of these
ls /etc/init.d
service --status-all
systemctl -l --type service --all
sudo systemctl -r --type service --all
Of course! That makes sense, the device is already in use!
sudo killall dump1090-fa worked a treat.
Scanning now.
perfect here is my 24h scan and one that shows that the fa-standalone-filter isn’t a good choice when you have strong signals at 960mhz
why you wanna buy two now? just buy one t2 and find out whether it performs better. in addition for these huge differences you see between zero and pi3 there is no need for a splitter. make 2 identical sd-card images and run one for five minutes and then the other one. this is how i’m doing 99% of my tests - and be assured it works
Well I figure to run a true comparison I should run them both at the same time over an extended period, connected to the same antenna and using the same dongles. That way I know that the only difference is the zero and I can keep my main site feeding without interruption while I mess around getting the zero to work properly.
I got rtl_power to scan but I couldn’t get the display to work. Working on that now.
there is zero difference between zero and pi3 in receiving mode-s via standard 820t2 dongle! i approved this using two different zero w while receiving up to 300 aircrafts. period
scan image:
cd rtl-sdr/build
wget https://raw.githubusercontent.com/keenerd/rtl-sdr-misc/master/heatmap/heatmap.py
chmod +x heatmap.py
nohup ./heatmap.py scan.csv scan.png &
I know that is what is supposed to happen, perhaps it is the difference between an R820T and an R820T2?? What do you suggest, just get a T2 for the zero? Problem is getting a pigtail locally, MCX doesn’t seem to be a common connector here.
wget gave me a Unable to establish SSL connection error which I fixed by modifying /etc/resolvconf.conf but the next problem is I can’t get the png to generate.
Good news is I did get gqrx working though, I just have learn how to drive it!
NO this NOT supposed this is ASSURED by one-to-one tests i did. but you can stab around in the dark until the end of time - if this is your plan the other option is to straight forward to best possible results you can get at your site.
did you email anthony about the saw-filter amp without case and what the costs then were inclusive transport to australia?
the result of your scan is that your mode-s receiving is very faint - and like i supposed the 960mhz noise makes the fa standalone filter obsolet.
buy a 820t2 with mcx or smc connector → use the uputronics amp/filter as close as possible to your antenna → buy reasonable 50 ohm antenna-cable. that’s it.
what i was wondering when i wanted to send you links what and where to buy - i just found books at amazon and even on ebay nearls every single piece comes from china. do you koalas don’t have any electronic mail order over there?
p.s. in addition what is a bit strange that there is nearly zero signal at 1090mhz but instead a little more at 1110mhz. don’t know whether there is a real signal at 1110 or the frequency of your dongle could be so much off. anyways simply go for the dongle all other use - the 820t2 with smc or mcx connector.
But it’s not, is it. My one-to-one tests show a difference with my system. Why am I not able to simply unplug the RPi3, plug in the zero and see exactly the same results? Am I missing something obvious here?
The whole point of this post was because I simply wanted to replace my RPi3 with a zero which, by all reports, gives the same results. It’s not working for me hence having to ‘stab around in the dark’ in an attempt to figure out why. My aim, at the moment, isn’t to improve my system to maximise results. That can come later.
If the reason I’m seeing a difference between the RPi3 and the zero is simply due to the dongle I am using isn’t very good or that the RPi3 can compensate better for it, then Great! I’ll order one right now. If it’s not that, then what is it?
One thing it can’t be is antenna, coax, connectors, amps and filters (or lack thereof) because, as you say, the pi’s have been demonstrated to be the same. While these things would no doubt improve performance overall, they don’t explain the difference unless the zero is more sensitive, less capable or doesn’t play well with the RTL-SDR dongle I have, which is definitely a possibility.
I highly suspect the reason the frequency is off is due to the dongle. Either due to heat or it is just rubbish, maybe both, I don’t know. RTL-Power did state during the test that the frequency wasn’t locked, so that could be it too??
The reason I was hoping to source items locally is not wanting to wait 3-4 weeks for stuff to arrive from China or the US.
I’ll order one of those ones you link to and see what happens.
perfect decision but again is it really that hard to get electronic stuff over there in australia? and does amazon australia really just sell books and nothing else?
Out of curiosity, have you run any tests without amps and unfiltered?
Yes, only books. We can get almost anything we like, it is just a matter of insane shipping costs and delivery times. e.g. Getting something delivered from the U.S usually takes 3-4 weeks and the shipping costs more than the item.
bummer! seems to be a way to get a rich man - selling electronic equipment fast to aussies