Three new SDRs in the shack

[Terry Fox]

I have recently purchased three new SDRs for testing and playing with.  Up 
front:  I have NO RELATIONSHIP with any of these companies, other than being 
a customer, who paid full-price for everything.

1/2:  BooyaSDR
The first two are BooyaSDR units.  These are different than almost all other 
SDRs available to RF enthusiasts at the moment, in that they sample the RF 
input directly, then send ALL those samples to the host computer.  No FPGA 
or other device to reduce the amount of samples.  The host computer then 
does all the computations to show the FULL bandwidth of their respective 
input.  The software shows this spectrum display in a series of waterfalls 
stacked on top of each other.  I purchased one of the 100MHz full SDR & 
antenna, and then a 16MHz digitizer set, less antenna.  I also have a 64MHz 
oscillator that I can put in place of the socketed 100MHz oscillator, so I 
can test the boards at that lower rate.

There are three models of the BooyaSDR currently available.  They are:
16MHz/14-bit sampler, uses USB-2, $55
64MHz/16-bit sampler, requires USB-3 on host computer, $235
100MHz/16-bit sampler, requires USB-3 on host computer, $265

Each also has an option to purchase an active antenna (+$30), which seems to 
be based on the PA0RDT active antenna.  The antenna uses CAT-5 cable between 
the antenna itself and the power injector board.  The injector uses a 12VDC 
wall-wart for power.  I'm not sure if the antenna is available separately, 
or only with one of the above SDRs.

I am very interested in whether the 16MHz BooyaSDR can be used for LF 
operations.

The BooyaSDR requires the use of their own software.  Since there is no 
Digital DownConversion (DDC) in an FPGA or other device, the host computer 
MUST be really fast.  I have three i7 computers that I've tried with the 
BooyaSDR.  The latest, a Dell 8900 just over a year old, generally runs ok 
with the BooyaSDR.  A less-than year-old Microsoft Surface Pro 4 has trouble 
keeping up at the 100MHz sample rate, but is somewhat better at the 64MHz 
sample rate.  An older HP dv7 laptop has a lot of sputtering at either rate 
on the faster SDR.  All three computers work fine with the 16MHz SDR.

I should mention that this samples-direct-to-very-fast-computer is new for 
amateur SDR systems as off-the-shelf computers are just now getting fast 
enough to handle the high sample rates.  The openHPSDR people have been 
leaning in this direction within the last year or so, but BooyaSDR seems to 
have beat them to it.

The Windows-only BooyaSDR software seems a little awkward to use initially, 
compared to other SDR-based software packages, such as HDSDR, SDR#, 
SDR-radio/SDRConsole, etc.  The source code for the BooyaSDR software is 
available, and compiles with the no-cost CodeBlocks C compiler.  I have 
rebuilt the software on all three computers mentioned above.  There are a 
couple tricky things that I can explain if you are so inclined.  You CAN 
zoom into specific frequency ranges, using somewhat awkward controls at the 
top of the GUI.

There is a lot of potential with these BooyaSDRs, since you can see the 
complete spectrum (up to 1/2 the clock sample rate) all at once.  Trying to 
zoom into particular signals can be difficult, however.  I need to play with 
the hardware and software more before I can give an honest review.  My 
initial reaction is:
1.  Seeing this much spectrum can be a real trip.  This is impressive!
2.  Computer MUST be fast.
3.  Software could be tweaked some to improve user experience.
4.  Use of Cat-5 (UTP) cable between active antenna board and power injector 
(bias-T) may allow RFI.
5.  My boards seem to overload rather easily.  I have an AM broadcast 
station nearby, and have this problem with some SDRs.
6.  There is NO filtering or protection in front of the ADC inputs.  Of 
course, to see the whole spectrum, you may not WANT filtering....
7.  No matching transmitter.
8.  BooyaSDR is to be commended for thinking outside the FPGA!  As host 
computers get faster, more SDRs will be doing this.

If interested, check it out at http://booyasdr.sourceforge.net/


3.  LimeSDR
Last summer, I succumbed to a Crowd Supply offering, called the LimeSDR. 
Crowd Supply is another Indiegogo or Kickstarter.  It finally arrived this 
May.  The LimeSDR is supposed to cover 100kHz to 3.8gHz, with two receiver 
and two transmitter channels, making the LimeSDR a MIMO device 
(Multi-In-Multi-Out).  It uses an LMS7002M chipset, providing 12-bit 
sampling.  THe LimeSDR board only costs about $289.

Funny thing happened on the way to the final design.....  Some filtering 
components were added to the receiver front-end that drastically limits 
reception on LF and HF.  100kHz to approximately 100MHz were MIA. 
Apparently the LimeSDR designers decided that even though the main chip 
could receive down to 100kHz, and the LimeSDR was advertised to operate down 
to 100kHz, going down that far MIGHT cause problems in VHF and UHF 
operations.  So they added a little filtering.  No biggie.  And no LF or HF.

Finally, some clever people saw that a small (0402 SMT) inductor was right 
across the input.  Removing that inductor (with the flick of tweezers) 
allows HF operations!  I removed said device, and I can now receive HF 
pretty good.  LF is still wanting, however.  In the LimeSDR Blog or Wiki or 
somewhere, there are additional details on how to change the rest of the 
components in the ADC input filter to further improve LF/HF operations.  I 
plan to order the tiny 0402 capacitors to accomplish this.

The input and output connectors on the LimeSDR are VERY TINY and not for the 
faint of heart, U.FL.  Fortunately, adaptor cables to SMA can be found 
easily and cheaply on ebay and Amazon.  The connection to the host computer 
is USB-3 (like the BooyaSDR).  The LimeSDR DOES have an FPGA onboard with a 
DDC, to reduce the samples to/from the computer.

The LimeSDR actually has four receive input connectors and two transmit 
output connectors, with software selection possible.  The LimeSDR can be 
used with many SDR programs, with SDRConsole by Simon Brown being the most 
popular.  There are also other programs being developed for the LimeSDR, 
such as a Python-based Vector-Network-Analyzer (VNA).

If you are interested in VHF/UHF/low microwaves, the LimeSDR seems to be a 
good choice.  There is a comparison chart showing the LimeSDR compared to 
some of the ETTUS and other hardware, which might be of interest.  HF is 
also adequate once the stupid inductor is removed.  LF operations is still 
TBD at this point.

For more info, see the following links:
https://www.crowdsupply.com/lime-micro/limesdr
http://www.limemicro.com/
https://myriadrf.org/projects/limesdr/

Once again, I am NOT trying to sell any of these.  And, I still like my Red 
PItaya.
73, Terry, N4TLF