[Hackrf-dev] Is my HackRF One broken?

[Giovanni Mascellani]

Hi everybody.

I am not very experienced with SDR, but I believe that my HackRF has
problems when receiving. Here is what I have tried so far: I would be
very grateful if you could tell me whether this is normal or not and
possibly other tests to understand what is going on.

First of all, I am using an up-to-date Debian sid. The hackrf software
is the one distributed there. The HackRF One is connected directly to
the computer USB port and I am using the default antenna ANT500 shipped
with it.

$ hackrf_info
Found HackRF board 0:
Board ID Number: 2 (HackRF One)
Firmware Version: git-44df9d1
Part ID Number: 0xa000cb3c 0x00674352
Serial Number: 0x00000000 0x00000000 0x457863c8 0x2563771f

Gqrx opens it successfully and shows a plausible spectrum and waterfall
(I see very FM distinct stations, if I move to 900 and something MHz I
can see less distinct but still clear GSM channels; I can see DVB-T
channels and WiFi channels). I can tune and listen to FM stations, so
receiving works at least partly. But FM receiving is easy: the problems
begin when I try to decode something more delicate.

For example, if I try to use kalibrate-hackrf I get inconsistent results:

$ ./kal -g 16 -l 32 -s GSM900
kal: Scanning for GSM-900 base stations.
GSM-900:
	chan:   23 (939.6MHz + 24.296kHz)	power:  166951.26
	chan:   24 (939.8MHz + 9.106kHz)	power:  231138.37
	chan:   25 (940.0MHz - 23.307kHz)	power:  169991.90
	chan:   26 (940.2MHz - 38.398kHz)	power:  173509.28
	chan:   27 (940.4MHz - 34.286kHz)	power:  169435.45
	chan:   51 (945.2MHz + 31.693kHz)	power:  203465.57
	chan:   52 (945.4MHz + 37.553kHz)	power:  199520.76
	chan:   53 (945.6MHz - 9.242kHz)	power:  189952.00
	chan:   55 (946.0MHz - 18.520kHz)	power:  219433.97
	chan:  101 (955.2MHz + 16.068kHz)	power:  224264.62
	chan:  102 (955.4MHz - 8.907kHz)	power:  231327.13
	chan:  103 (955.6MHz - 33.901kHz)	power:  232225.30

For reference, here is the result of kalibrate-rtl running at the same
time on a random cheap RTL-SDR receiver:

$ ./kal -g 38 -s GSM900
Found 1 device(s):
  0:  Generic RTL2832U OEM

Using device 0: Generic RTL2832U OEM
Detached kernel driver
Found Rafael Micro R820T tuner
Exact sample rate is: 270833.002142 Hz
[R82XX] PLL not locked!
Setting gain: 38.0 dB
kal: Scanning for GSM-900 base stations.
GSM-900:
	chan: 1 (935.2MHz + 35.902kHz)	power: 152588.67
	chan: 19 (938.8MHz + 35.669kHz)	power: 79649.74
	chan: 33 (941.6MHz + 35.259kHz)	power: 96929.05
	chan: 34 (941.8MHz + 34.809kHz)	power: 446793.20
	chan: 94 (953.8MHz + 33.924kHz)	power: 95281.88
	chan: 98 (954.6MHz + 33.869kHz)	power: 302184.33
	chan: 100 (955.0MHz + 33.952kHz)	power: 117813.97
	chan: 106 (956.2MHz + 32.767kHz)	power: 90438.32

Results for HackRF have erratic channel offsets and rather constant
power levels, which to me means that it is just detecting random things
that by chance pass the detection threshold. Instead results for RTL
have consistent offsets and random powers (which is sensible, since
channel power is adapted to what is actually needed and towers may be
more or less far from me).

If I launch detection on a specific channel, RTL terminates in a few
dozens seconds with a consistent result (consistent both between
different invocations and different channels). HackRF (even when using a
channel that works with RTL) never terminates, which I think means that
it never finds any good reference.

What do you think about this? Do you have any suggestions?

Thanks, Giovanni.
-- 
Giovanni Mascellani <g.mascellani at gmail.com>
PhD Student - Scuola Normale Superiore, Pisa, Italy

http://poisson.phc.unipi.it/~mascellani

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <https://pairlist9.pair.net/pipermail/hackrf-dev/attachments/20160719/9d720a0d/attachment.pgp>