On Sun, Oct 26, 2008 at 9:08 PM, rafael2k [email protected] wrote:
I’d like to know the output power I can get from a usrp 1, and if someone has
any experience with this kind of experiment, like what amplifiers to use,
what filters, and things like that.
The maximum USRP1 output using the LFTX board is approximately 2V P-P
into 50 ohms, or about 10 milliwatts rms. Depending on your waveform
peak to average ratio, your rms power will be less.
On TX, to get to say, a max of 5W for QRP operation, you’d need a PA
with at least 27 dB of gain. For operation with SSB, this would have
to be a linear amplifier. For CW only, you could use a non-linear PA
which would be easier to design.
The LFRX board has no gain, and essentially just provides
anti-aliasing filtering and impedance matching between the input at 50
ohms and the analog-to-digital converter input circuitry. You can
think of it as having roughly a 55dB noise figure, so you’d need at
least that much external low-noise amplification to get any sort of
useful weak signal reception. (The 55dB figure was from an
empirical measurement with a signal generator; YMMV.)
Since the ADC is digitizing the entire 0-30 MHz spectrum, it would be
essential to have a band filter ahead of the amplification, in order
to avoid strong out-of-band signals limiting the amount of gain you
could use. In addition, you’d want some sort of analog AGC.
Anyway, it’s obvious the LFRX and LFTX were designed as baseband
interface cards, not as an HF receiver or transceiver. Still let us
know what you come up with. It’s certainly possible, given the open
design of the USRP1/USRP2, that you or someone else could design
customer daughterboards that incorporate some or all of the above.