[Hackrf-dev] HackRF problems

[McDonald, J Douglas]

I want to emphasize that the RF chain in mine is not defective.
I tested it again last night. The three different gains all
work exactly as expected.

When connected not to an antenna but to a wired system
the number of spurious signals is greatly reduced but is nowhere near
zero and not as low as the dongles.

I think that most people for over the air receiving purposes need
to get really good bandpass filters to pick out only the signal of interest.

And remember that 8 bits is a very poor dynamic range for a system
without a really good, designed to purpose,  ANALOG AGC, so you have to,
by hand, get the various gains exactly right. I do it by looking at the
spectrum display. In general when working with weak signals
you want to leave the amp off and slowly raise the two other
gains in parallel from all the way down. At first if your is like mine
you will see big jumps up and down in the level of weak signals,
spurious signals, and noise. The spurious ones come and go
as you raise gain. At some point the apparent noise level
stabilizes and remains the same for a while as the real signal
levels rise up out of the constant noise floor. Then the noise
floor rises up. When it has risen about 5-9 dB you have reached the
best sensitivity and more important dynamic range and
lowest quantization distortion. I am fairly sure that this gizmo
has no built-in dither at the 1-LSB level. If it does there is something
bad going on, perhaps in the DC level area.

One way to look at it is the HackRF is NOT a receiver: its effectively an
IF strip with no AGC.

Today I'll try it as a transmitter feeding my old spectrum analyzer
looking for spurious signals. Currently for most spectrum analysis
I use either it (>2MHZ bandpass) or one of the dongles (better
for narrow band).

Doug McDonald

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://pairlist9.pair.net/pipermail/hackrf-dev/attachments/20141127/374fbbfe/attachment.html>