Click to view our Accessibility Statement or contact us with accessibility-related questions
Drop Refurbished
Like-new products you can trustDrop Rewards
Get $5 for every 500 points you earn! Learn more
Drop Keyboard Club
Become a member and expand your keycap collectionCollaborate With Us
For Brands & DesignersFollow Drop
- The web interface is pretty slick, but that's all you get. No API at the moment, no way to channel raw IQ data anywhere. - There is no hardware gain control, for cost reasons. The frontend can and will overload if you're close to even a single AM broadcast station. And you'll have no choice but to either attenuate the entire signal (not very desirable!), build a selective antenna, or build a notch filter to attenuate just that band. You're on your own if you need to build a filter, it's not documented anywhere on the site, however you'll find a link buried in the support forum. - 12 bit Analog-to-digital converter is MUCH better than the 8-bit ADC's found on RTL-SDR dongles. 16 would be even better, the 12-bit does present a problem when overloaded with broadcast stations. Without a hardware gain control, the entire board is useless, if that overloading signal creates intermod. - The KiwiSDR includes a software GPS that's used for disciplining the main oscillator. Clever! you'd think, because GPS time sources are exceptionally accurate, being that they all have Cesium and Rubidium atomic clocks. He remarks that the GPS frontend was basically free it was so cheap. My problem with this is that his software GPS approach (of demodulating and decoding the GPS navigation data) is not 'free'. The BeagleBone is spending a massive amount of time chewing through that data, as is the FPGA. This isn't nearly as efficient as a dedicated GPS chip would be, nor as simple as a TCXO or other precision clock source. There's no room left for modification, and bandwidth is limited between the FPGA and the BBB because of the software GPS. It's VERY clever, but I'm going to disable mine and make some modifications to the hardware, which will give me more CPU and bandwidth to play with. - The bandwidth is actually pretty deceiving. Yes it will receive 0-30 MHz, but not all at once. It can generate a waterfall across 0-30, but when it actually decodes a signal, it's only using a few KHz of bandwidth, and can only decode a few channels. It's clever, but in reality, it's got less real-time bandwidth than an RTL-SDR dongle. But because it has a full bandwidth waterfall (unlike the RTL-SDR), you won't care. As I mentioned above, more bandwidth is theoretically possible, but would take some work / modification. - The Active MiniWhip type antennas are highly recommended for this or any other high bandwidth SDR. Rather than being resonant at one frequency, they pick up everything. They're small, cheap, and can be placed anywhere outside. Just make sure to install a ground rod at least 8 feet deep and ground it. Otherwise it's useless. You can pick up kits or fully built active MiniWhip antennas on eBay. Mine shipped from Ukraine and costed me $18.00 USD. I had to fabricate an enclosure for the antenna and an enclosure for the power splitter, which probably ran another $20.00. A ground rod will be another $15.00 or more. All in all though, these will actually perform better than a dipole or vertical. It's hard to beat a dipole, but the Active MiniWhip does, hands down. Here's an example of one if you don't believe me: http://websdr.ewi.utwente.nl:8901/ - SDRplay is a better deal all around. Higher bandwidth (1 kHz to 2 GHz vs KiwiSDR's 0-30 MHz). Larger playback bandwidth (10 MHz vs a few KHz x 4), 8 built-in selectable filters which can block out broadcast stations that would otherwise overload the frontend. Sixteen receive channels vs 4. Includes two separate antennas, one of which has a power source for an LNA. Has a hardware RF gain conrol (KiwiSDR does not). Supports more SDR applications and includes an API. MUCH cheaper, at under $200. - That being said, KiwiSDR is fully open source, which has my full respect. I think the creator needs to take a hard look at SDRPlay and implement some of those features. Or hell, maybe I'll fork it and modify it if he won't. I don't buy the 'low cost' argument because somehow the SDRplays are cheaper.
My experience is that for general listening and spotting new signals the Kiwi SDR is an excellent performer.
The SDR Play has it's own problems with dynamic range and it requires either a lot of RF gain adjustment or activation of the broadband AGC to prevent ADC overload as you tune around. The HF pre-amp (required on the HF bands to get the best sensitivity) also degrades the dynamic range of the ADC. I had an SDR Play V1, but sent it back for a refund as I found the Kiwi behaved better and got much more use at my location, but your milage may vary.
Take a listen to my KiWi and judge for yourself what it is capable of when a good antenna is connected.
http://southwest.ddns.net:8073/
Regards,
Martin - G8JNJ
I sold my KiwiSDR and have been much happier with my SDRPlay RSP2. Have you actually tried the RSP2? I find it confusing because my experience is the opposite. The KiwiSDR doesn't even have a gain control, nor filtering. How could the RSP be possibly any worse than the KiwiSDR in this regard? The KiwiSDR by its very design will overload and cause artifacts, whereas the RSP2 can completely eliminate out-of-band interference, and allow the gain to be placed anywhere you'd like.
The RSP2 also comes with an API, whereas the KiwiSDR is just a constrained listening device (albeit, I concede, internet-connected which wins lots of points there). With the API, I've already been able to get my feet wet writing decoding algorithms and whatnot. The RSP2 has a higher bandwidth as I mentioned, so it's more useful for monitoring and scanning. Not to mention that it goes up into the UHF and VHF range, even up into the L band if I recall correctly. KiwiSDR just can't compete in those areas. It's really good at exactly one thing -- putting 0-30 MHz onto the internet for listening. As I said, it wins a lot of points there. It's very slick. Love OpenWebRx.
I think it's not so easy to compare the two SDR's because they are intended for different purposes.
For those that are interested, the background information relating to the points I made earlier in this thread can be found in the comments section of this review.
http://www.rtl-sdr.com/review-sdrplay-rsp2-db-gain/
Regards,
Martin - G8JNJ
The Kiwi is still NOT an "IQ generator" like the RSP2 and most other USB-output style SDRs. That was the entire point of the project. But a lot has changed since flyinglotus' comments nearly a year ago.
There is now an external control API and IQ output, although only at the current 12 kHz "audio" bandwidth. See the "kiwiclient" project on Github for Python code that talks to a Kiwi and can save IQ data to a file. And people are using virtual audio cable (VAC) programs to link Kiwi IQ output to DRM and C-QUAM AM stereo decoders and other external IQ processing programs.
The latest development is that GPS timestamped IQ samples are being used to do crowd-sourced VLF/LF and HF direction finding. See: http://hcab14.blogspot.com and https://github.com/jks-prv/Beagle_SDR_GPS/issues/130 It's early days but this is a very exciting development. And I'm quietly chuckling at all the people that called me an idiot for spending the additional $8 to put an SD-GPS on the board in the first place.
TDoA location solution for German time station DCF77, 77.5 kHz using GPS timestamped IQ data from three cooperating KiwiSDRs. by Christoph Mayer hcab14.blogspot.com