Ah-so Dave, no differential amps in your SDR kit. Jack; if memory serves - if you want "general coverage" try to find an SDR with low-noise differential amps instead of opamps. With a differential amp, the gain is constant regardless of the antenna impedance, whereas often with opamps, the antenna impedance forms part of the input circuit to the opamp, which in conjunction with the feedback impedance determines the gain of the SDR. In other words with an inverting opamp, your SDR gain will vary with antenna impedance. Something that might not be good for a general coverage setup.
You might get around this with a non-inverting opamp setup instead of inverting, but I've never seen a radio like that and I seem to remember reading somewhere that it is a bad idea; can't remember why. BTW, low noise differential amps have higher noise than low-noise opamps. The best differential amps I've seen have at least a couple nV/rtHz noise with low-noise opamps these days running significantly below one nV/rtHz. But this difference probably would not matter much on the HF bands though. The venerable Burr Brown (now TI) INA163 comes to mind as a decent low-noise differential amp, 1nV/rtHz and a single gain set resistor. For general coverage you might use a low-noise pre-amp ahead of the QSD to prevent the antenna impedance from affecting the SDR bandwidth. But as I stated before, this is a contentious idea as it will reduce dynamic range, for one thing. These two issues (diff amp and pre-amp) are explained in the excellent "SDR for the Masses" 4-part series of QEX articles. Can't remember the author's name - and I don't have access to the papers at the moment. Then there's the issue of an input filter. I'm looking at a simple manually tuned preselector for a general coverage SDR. I might put the tuned circuit through a pair of step-up RF transformers to improve Q. Finding cheap 365pf variable capacitors is becoming increasingly difficult though. I'm also going to look at a high-impedance input pre-amp using a FET low-noise op-amp. I have to study the noise budget carefully first though to make sure I'm not creating a problem rather than solving one. It's been a long time since I wrestled with opamp noise calculations. GL, David --- In soft_radio@yahoogroups.com, [EMAIL PROTECTED] wrote: > > On 31 Mar 2008 at 4:18, drmail377 wrote: > > > Hi Jack, > > > > Do have a look at Dave WB6HDW's fine DDS SDR and filter kits. I see he > > already posted a reply to you. If you want general coverage, then you > > want to use a DDS (see detail below). Plus, you need to use > > differential amplifiers on the QSD output instead of op-amps to reduce > > the affect of antenna impedance changes. A purist might also want to > > put a preamplifier/preselector on the front end of the radio to > > further isolate the affects of changing antenna impedance on > > performance. But many will argue that a preamp is not necessary and > > may actually be harmful as the dynamic range of the QSD is > > compromised. I don't believe the WB6HDW SDR has a preamp, but I think > > it does use differential amplifiers. > > > No, it uses a regular OP amp. The circuit is basically a clone of the SoftRock with very high > frequency dividors in the quadrature generator to handle the max frequency of the DDS and > faster switches in the QSD. > > Dave - WB6DHW > <http://wb6dhw.com> >
