Comparison between Lime SDR Mini and ELAD FDM-S2

Lime SDR is a very smart project supported from Myriad RF .
It is mostly based on a big multifunction chip LMS7200M that allows both RX and TX over a wide frequency range (10 MHz - 3.5 GHz) and bandwidth till 30 MHz with a quadrature double mixer conversion and 12 Bit DAC .

This are the valuable points .

https://wiki.myriadrf.org/LimeSDR-Mini_v1.1_hardware_description#RF_Frequency_Range

Regarding the RX section it is based on a double baseband conversion in quadrature generating I&Q signals that are then applied to two different AD converters at 12 bits and then to an FPGA that drives data to an USB3.0 interface .
Elad FDM-S2 is a well known high performance receiver covering from 10kHz to 54 MHz  the FM international band 74-108 MHz and the weather satellite and Radio Ham 2m band 135 -148 MHz  .

It is a direct sampling receiver at 16 bit with 122.88 MSPS using the first three Nyquist bands and filters to cover the range specified .

The performance are well listed in this table :

The full manual where thay was extracted can be found here : 

http://sdr.eladit.com/FDM-S2%20Sampler/ELAD%20FDM-S2%20User%20Manual%20Rev%201.08.pdf

 Both Lime and ELAD can be driven from HDSDR .
For SDR Lime Mini , this are the info :

https://wiki.myriadrf.org/LimeSDR_ExtIO_Plugin_for_HDSDR

I have decided to perform some comparison using this software and a Marconi 2022E RF generator  searching for a "rough rule of thumb evaluation" of the absolute dynamic range and the SFDR  (Spurious Free Dynamic Range ) of the two receivers in the FM Band .

Lime SDR has a lot of variables on various stages of the architecture ( at least three different gains ).
This is typical of this architecture and unluckily not a good sign ( think at similar architectures from Mirics or Rafael Micro ) .
It is not clear , but it also looks that there are no RF filters at all .

Trying to put Lime SDR in the best conditions ,thinking at spectrum analyzers that use directly mixers for having the minimum of conversion spurious , I have reduced at the minimum the control of the RF gain on the panel of HDSDR .

Without any input I have get the picture of the whole 20 MHz FM spectrum ( with center at 98 MHz ) like in the picture .

Even after calibration ( as suggested from the instruction ) , the center line at 98 MHz was found to be at a level that we will consider -70 dBr ( relative to the 0 of HDSDR ) .
The noise density in a band of around 600 Hz was found to be around -97 dBr .

After and applied to the input of the receiver a signal at 97 MHz tuning the level till I read on the HDSDR a comfortable level for the A/D of around 10 dBfs .

The input level was remarkably 0 dBm ( seen as around -10 dBr) , but a lot of spurious was seen on the band .
The highest was spur at a level of -40 dBc .

I repeated something very similar with the Elad FDM-S2 of course with bandwith of around 6 MHz .

The first image is without signal applied to the input .

In a band of 750 Hz ( the most similar to that of Lime SDR , that HDSDR allowed me ) , is not visible even under -100 dBr and I can insure that it is the same on all the band .

I applied then the 97MHz carrier at the input till getting more or less the same level at the A/D ( -10 dBfs ) .

The level applied in this case was -7dBm and what was read was - 10 dBr ( this means that the scale in this case is quite precise ) .
The very small carriers seen around 97 MHz are spurious from the generator .
The increase of the noise is again the wideband noise generated from the amplifiers of the generator and on the spectrogram can be seen also many FM stations whose signal are so high here in Italy that are able to skip the shield of the cables and /or generator .
The same can seen on all the band , bot nothing else like a carrier that will be a spur of the 97MHz generated from the Elad .

 Of course I could be wrong , but my impression is that , regarding the FM Band , we have two different things :

- A good consumer integrated receiver
- Something very similar to an instrument

I am waiting to check an ELAD FDM-S3 that is able to cover the full FM band with the same specs of the FDM-S2 .