[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Proposal: SDR-DSP receiver

The receiver is a dual-conversion superheterodyne with a 350 MHz first IF and a
10.7 MHz second IF. One PCB handles the analog signal processing and second PCB
generates the local oscillator (LO).

The input to the analog signal processing board is 28-2450 MHz RF and the
bandpass filters are external (to be taken from previous projects). The RF is
mixed with a 378-2100 MHz LO in a high-level diode DBM. The first IF is at 350
MHz and has two 2MHz bandwidth SAW filters and about 15 dB of gain provided by
SiGe MMICs with a +36 dBm OIP3. The 350 MHz signal is then converted to 10.7 MHz
in a high-level Gilbert-cell mixer. Here it is filtered again by 7.5 or 30 kHz
wide 4-pole monolithic crystal filters, a 230 kHz wide ceramic filter or a 2 MHz
wide LC filter and then amplified and applied to the ADC. AGC is applied to the
10.7 MHz IF amplifiers which can provide 10-60 dB of gain. Lots of gain is
required to bring narrow-band signals like PSK31 above the ADC spur level. Wide
band signals need little gain. The trick is not to degrade the noise figure when
the gain is reduced so CFB op amps with variable feedback are used.

The LO board uses a 1-2 GHZ low-noise VCO whose output passed through an ECL
divide by 1/2/4/8/16 chip to get lower frequencies. The VCO is phase-locked to a
VCXO with a sigma-delta PLL so the frequency resolution is less than 1 Hz.

This will be the subject of another article for QEX in the near future.



----- Original Message -----
From: "Achim Vollhardt" <avollhar@physik.unizh.ch>
To: "John B. Stephensen" <kd6ozh@amsat.org>
Cc: <kaj.wiik@iki.fi>; "AMSAT Bulletin Board" <amsat-bb@amsat.org>
Sent: Tuesday, 30 March, 2004 09:16 UTC
Subject: Re: Proposal: SDR-DSP receiver

> Hello John and Kaj,
> the thing with sampling 144 MHz directly was just meant to be an
> example.. to do it properly one would convert this down to about 10 MHz
> and sample it there, there I agree.
> I am certainly interested in your files ( I can handle quite some
> MBytes..:) . What I don't understand up to now how frequency selection
> is made.. with a VCO before sampling or a digital down converter like
> the AD6620 or similar?
> 73s Achim, DH2VA
> John B. Stephensen wrote:
> > Hello Achim and Kaj,
> >
> > The dynamic range is determied by the level of the signal artifacts produced
> > the ADC and these get worse as the RF frequency goes up. The AD9244 has a 90
> > dynamic range when the input signals are limited to 45 MHz. By 240 MHz the
> > dynamic range has dropped to 70 dB.
> >
> > I was planning to convert the RF to 10.7 MHz and digitize there to get the
> > performance out of the ADC. I've got a partially assembled receiver board
with a
> > 350MHz 2MHz-wide first IF and an assembled but not yet tested 60-2150MHz LO
> > board. Transmitting is easy. You can get quadrature modulator chips for
> > 50-3500MHz so all that is needed are baseband I and Q signals. That's why my
> > board has two DACs.
> >
> > 73,
> >
> > John
> > KD6OZH

Sent via amsat-bb@amsat.org. Opinions expressed are those of the author.
Not an AMSAT member? Join now to support the amateur satellite program!
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org