QS1R RF input modification


1. Introduction

The QS1R is an excellent SDR receiver that covers 0-62MHz in one shot. The sampling rate is 125MHz and ADC has RF input bandwidth up to 700MHz, so one can use this receiver in other Nyquist zones beside the first one. My partucular interest is the third zone that holds 2m amateur radio frequency band (144-146MHz). 

The QS1R has the ability to bypass the input LPF filter by a simple rework. Now one needs to use external antialias filters.

In case one wants to retain the internal LPF (HF port) AND have a dedicated VHF port, this can be achieved with a relatively simple modification. The LPF needs to be replaced with a LPF/HPF diplexer. Fortunately there is an opportunity to reuse the current LPF and PCB lines that enable LPF bypass so that the most troublesome work is drilling the hole in the front panel for additional connector (VHF port).

2. Modification

The modification is shown on pictures below.

Original input LPF (with bypass components indicated):

Original input LPF

Input LPF modified to DPLX (remove last shunt 62pF cap and add HPF components - 2 x 27pF + 15pF + 2 x 47nH):
Modified schematic

Rework step 1 (remove 62pF cap; solder first 27pF cap; remove solder mask and cut PCB traces; use existing GND vias for inductor ground connections; second inductor is placed to already present pads for C67):
Rework step 1

Rework step 2 (solder the rest of HPF components):

Rework step 2

Rework step 3 (drill additional hole into  front panel for your connector of choice):

Rework step 3

Rework step 4 (make a coax connection - first remove more solder mask and cut the trace just behind the coax solder point):

Rework step 4


- use 0402 or 0603 components only (if at hand use shielded 47nH inductors - as the ones in the LPF section)
- be carefull in soldering coax central pin not to create shortcut - remove some ground copper around pcb trace at that point
- really use coax connection not just a piece of wire and hoping pcb top copper plane and front panel will act as a gound

QS1R after modification:

Front panel after modification

3. Results

As the aim was to cover 2m/70cm amateur band the secod Nyquist zone was used as a transition band for LPF/HPF filter sections. This means that this modification makes QS1R unusable for 70MHz :(

The frequency amplitude and return loss response is shown on the picture below:


I made carefull SFDR test prior and after modification. About 10 new spurs appeared all below SFDR specification level (-103dBm) except one at 28.08MHz with -96dBm level.
At around 42, 56 and 60MHz the USB interface noise bumps were/are present just on the SFDR limit and can be made invisible with grounding the USB connector to its opening in the panel - I used piece of chocolate paper, not very profesional neither repetitive but those frequencies are not of my interest anyway...

The modified QS1R was then used as test instrument for measuring IMD performance of one 2m and one 70cm FET power amp (no external bandpassantialias filter was used) and many other ham applications ...

2m FET PA (f = 144 - 2x 62.5 = 19MHz)
2m PA

70cm FET PA (f = 432 - 6x 62.5 = 57MHz)
70cm PA