Re: [Discuss-gnuradio] Max Voltage swing for WBX boards

[Vincent W]

On 09/14/2010 01:29 PM, Marcus D. Leech wrote:
>>
>> Hi,
>>
>> I'm trying to do some signal reconstruction and in testing and developing a
>> concept design, I wanted to attenuate a test signal and hook it up directly 
>> to
>> the RX port of my USRP2 through either the Basic RX or LFRX daughter board 
>> (and,
>> eventually through my WBX board). My intention was to attenuate the signal
>> through either a resistive attenuator or inductively through a transformer.
>>
>> What are the absolute and preferred input voltage ranges for the RX, LFRX, 
>> and
>> WBX boards?
>>
> The BASIC_RX and LFRX boards accept a maximum power level of roughly 0dBm.
>
> I wouldn't put more than -20dBm into the WBX--since it has gain/mixer stages.
> Anything  with more power than that risks damage to the LNA.
>
>
>> What are the dangers I should be looking out for?
>>
> Too much power into any of the Rx boards can do damage.  But *before* you 
> reach
> the damage level, you end up with distortions caused by clipping, and driving
stages into
> non-linear operating regions.
>
> The BASIC_RX and LFRX boards are essentially gain-free boards, so they'll take
> higher input
>   power levels.  But all the other boards have several stages of gain in front
> of the mixers, and
>   there's usually a low-noise-amplifier (LNA) "out front".  Signal levels of
> -50dBm  to -80dBm are
>   pretty normal low-noise gain chains.
>> Are there specific precautions I should be taking?
>>
> Make sure that you don't inject too much power is the main thing.
>> What is the best way of attenuating the signal?
>>
>>
> Standard resistive attenuators--you can buy them ready-made in a whole bunch 
> of
> different attenuation ranges, or build them yourself with connectors/discrete
resistors.
>
>
> --
> Marcus Leech
> Principal Investigator
> Shirleys Bay Radio Astronomy Consortium
> http://www.sbrac.org

I had an opportunity to test this out, and everything worked out quite nicely. I
made a resistive attenuator using discrete resistors, and kept the voltage swing
to .15Vpp for the WBX, and 1Vpp for the LFRX board.

It worked quite nicely, and allowed me to save the signal to a file where I was
later able to process it using octave.

Everything seems to have checked out, so I'm pretty sure I didn't break
anything, and I didn't notice any signal clipping.


Thank you very much for your help,


Vvv