Signals appear multiple times

Having issues with the site, hardware, source code, or any other issues?
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Signals appear multiple times

Post by pe2bz »

Hi. After having a power issue with my first BladeRF I ordered and received an exchange BladeRF. Currently I am taking my time to install and use the BladeRF but I do not think that what I receive is what I want.

All is tried on both SDR Console and on SDR Sharp software and the same issue appears on both software.

No matter what samplerate and no matter what gain settings I use, signals appear multiple times on the waterfall. For example the ship traffic in the 162 MHz band with the AIS transponders on 161.975 and 162.025 MHz appear all over the band. Also on the airband signals appear multiple times. I have seen NOAA satellite transmissions 4 times on my waterfall today for only one satellite pass.

I have the BladeRF connected to an USB3 port. Changing it to an USB2 port did not solve the problem.

I attach some images and I hope there is a logic explanation and solution for this problem. I allready re-installed the software and had the BladeRF updated with no improvement.

Thanks for your replies,

Ben
bladerf_sdrsharp162MHz.JPG
bladerf_sdrsharp124MHz.JPG
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi there,

Could you confirm what your samplerate & bandwidth settings are for each of these configurations? We'll want to ensure your bandwidth is configured to be less than your sample rate to avoid aliases. Note that the lowest LPF setting on the LMS6002M is 1.5 MHz wide.

Best regards,
Jon
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Re: Signals appear multiple times

Post by pe2bz »

Hi Jon.
Thanks for your reply,

For all the posted images I did set the samplerate to 1 MS/s and the bandwidth to Auto.

This morning I tried for example the airband 124 MHz with 6 MSPS and auto and with 6 MSPS and 3,84 MHz bandwidth but the only difference is that I receive more signals in that range, all with their unwanted mirror signals.
See the attached file:
bladerf_airband.JPG
my BladeRF info:
bladeRF> info

Serial #: c7a6dce3cfffc7bfa8a296644f679837
VCTCXO DAC calibration: 0x90c0
FPGA size: 40 KLE
FPGA loaded: yes
USB bus: 0
USB address: 2
USB speed: SuperSpeed
Backend: Cypress driver
Instance: 0

Edit: Tried the same setup with the WinUSB driver , no change. Then I removed the Transverter board and also that makes no difference (except that reception starts @ 300 MHz)


Regards,

Ben
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi Ben,

In the case where you were using 1 Ms/s and "Auto" you would certainly be at risk of seeing aliases, given that the lowest LPF setting is 1.5 MHz. For a 1.5 MHz bandwidth, I'd recommend a 2 MS/s to 3 MS/s sample rate to ensure that the LPF reaches full rejection before hitting your sample rate. (Figure 6 in the LMS6002D datasheet shows the LPF roll-off in detail, if you're interested.)

However, your experiment with 6 MS/s and 3.84 MHz of bandwidth should be sufficient.

If a signal at (Fc + N) MHz is mirrored at (Fc - N) MHz, it's generally the case that some IQ imbalance is present. libbladeRF provides access to some IQ balance adjustment parameters provided by the FPGA. Unfortunately, I don't know if these GUI programs expose that functionality to the user. If not, you may be able to make changes in the bladeRF-cli, run the GUI, and repeat as necessary. However, this will likely be a bit of a painful process -- generally I like to use something like GNU Radio with osmoscom_fft or the bladeRF-cli in conjunction with baudline (in Linux) to view samples while making adjustments in the CLI.

If I were in your shoes, I think some of the things I'd try investigating/consider are:
  • Are you able to record and share IQ samples? In general, looking at IQ samples in the time domain can help note issues such as clipping, significant IQ imbalance, significant DC offset, or any other strange behavior. (The frequency domain is only half the picture ;) )
  • What is your antenna rated for? (It is a 50 Ohm load, correct?) Do you see this with other antennas or in other bands? (e.g., some NBFM voice channels in 462 MHz - 467 MHz)
  • Do you have access to a signal generator? Another SDR that you could use with something like osmocom_siggen (be sure to use an attenuator if connecting them directly)? It might be nice to see if you see the same results with a "clean" signal under your control, such as a tone or FM signal.
  • For the screenshots, could you list exactly what you expect to see where? I'm wondering if we're seeing a combination of intermod products and then mirrors due to IQ imbalance. However, I'm not certain what's supposed to be where in your waterfalls.
Hopefully something there helps. If not, keep letting us know what you've tried and we'll keep trying to offer better suggestions.
- Jon
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Re: Signals appear multiple times

Post by pe2bz »

Hi Jon,
thanks again for your reply.

I did make a record with SDRSharp of s DMR repeater on 438.375 MHz. On the screenshot below you see the signals disappearing after the repeater (at 12 miles distance) switches off. I expect only to see the main carrier but I notice 5 "strange" signals around. https://dl.dropboxusercontent.com/u/110 ... 0Hz_IQ.wav
438375mhz.JPG
It does not make a difference which antenna I use. I have it connected to a Logper 100 - 1300 MHz vertical, after an FM trap followed by a -3dB splitter. I also tried it with the Diamond V-2000 and with a mobile roof antenna, All stronger signals appear multiple times.

In the jpg image on 124 MHz with 6MSPS and 3.84 MHz bandwidth I expect to see carriers appearing and disappearing on the airband frequencies. However, just like in the 438375 example above, multiple carriers for each airplane appear and disappear. Setting vga to 3 dB (vga1 and 2) and LNA off does not solve this, the stronger signals still behave the same.

I have a HackRF but currently no Linux running. I did have a radiosonde with 100 mW transmitter at 405 MHz. I disconnected the antenna from the BladeRF and moved the radiosonde 10 meters away on the ground floor. 2 times concrete between the bladerf and the radiosonde. Still the main carrier shows up 5 to 7 times on the waterfall. Same transmitter on Airspy or SDR dongle only show up once.

As a try I did run the cli and used the calibrate command, I paste the before and after values, hope that's of any help.

Code: Select all

bladeRF> calibrate lms show
  LPF tuning module: 31
  TX LPF I filter: 31
  TX LPF Q filter: 31
  RX LPF I filter: 31
  RX LPF Q filter: 31
  RX VGA2 DC reference module: 31
  RX VGA2 stage 1, I channel: 31
  RX VGA2 stage 1, Q channel: 31
  RX VGA2 stage 2, I channel: 31
  RX VGA2 stage 2, Q channel: 31

bladeRF> calibrate lms
  Calibrating LMS LPF tuning module...
    LPF tuning module: 25
  Calibrating LMS TX LPF modules...
    TX LPF I filter: 45
    TX LPF Q filter: 33
  Calibrating LMS RX LPF modules...
    RX LPF I filter: 35
    RX LPF Q filter: 33

  Calibrating LMS RXVGA2 modules...
    RX VGA2 DC reference module: 27
    RX VGA2 stage 1, I channel: 37
    RX VGA2 stage 1, Q channel: 29
    RX VGA2 stage 2, I channel: 33
    RX VGA2 stage 2, Q channel: 29
bladeRF> calibrate dc rx
  RX DC I Setting = 543, error ~= 0
  RX DC Q Setting = 506, error ~= 0
bladeRF> calibrate dc tx

  TX DC I Setting = -132, error ~= 0.148199
  TX DC Q Setting = -28, error ~= -0.063094

bladeRF> calibrate lms rxlpf

  Calibrating LMS RX LPF modules...
    RX LPF I filter: 35
    RX LPF Q filter: 35

bladeRF> calibrate lms txlpf

  Calibrating LMS TX LPF modules...
    TX LPF I filter: 45
    TX LPF Q filter: 33
Kind regards,

Ben
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi Ben,

Just wanted to let you know we've been having some discussions on this on IRC and folks have been able to reproduce it. Playing with some of the mixer bias currents appears to calm these a bit, but at the end of the day it looks like there's a 250 KHz spur (frequency dependent) that we need to squash. We're looking into whether we've left something enabled (e.g., a comparator or cal-related clock) in the LMS6 that might cause this.

I'll write back when we have more info, and hopefully a solution.

- Jon
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Re: Signals appear multiple times

Post by pe2bz »

Thanks again Jon, you are taking support very seriously !

Kind regards,

Ben
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi Ben,

Brian (bpadalino) has tracked this down to a ferrite bead on the board. Removing it and shorting its pins appears to work around the issue; a number of kind folks on our IRC channel have made this modification and reported that they too see that the spurs no longer occur.

As I don't know FB on the board needs to be removed off the top of my head, I've asked bpadalino to add this to a wiki page, which I will link from here for you.

Best regards,
Jon
Serge911
Posts: 5
Joined: Wed Nov 04, 2015 1:13 am

Re: Signals appear multiple times

Post by Serge911 »

I have the same problem with mine bladerf. I am tried to make auto calibration as written in wiki. But it doesn't help. I don't have Linux at my PC, so I cannot using GNU radio for making manual correction. To me is not clear reason, why I should seen mirror signals at 444 mhz instead real at 436 mhz. In few replies above somebody promised to help to solve this problem. But no answer since 2013.
jump
Posts: 58
Joined: Mon Mar 03, 2014 5:31 pm
Contact:

Re: Signals appear multiple times

Post by jump »

I think you're looking at the wrong timestamp :)
2013 is the year jinik joined this forum. But this thread is pretty recent (initial post from Nov 04, 2015)
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Re: Signals appear multiple times

Post by pe2bz »

jynik wrote:Hi Ben,

Brian (bpadalino) has tracked this down to a ferrite bead on the board. Removing it and shorting its pins appears to work around the issue; a number of kind folks on our IRC channel have made this modification and reported that they too see that the spurs no longer occur.

As I don't know FB on the board needs to be removed off the top of my head, I've asked bpadalino to add this to a wiki page, which I will link from here for you.

Best regards,
Jon
Hi Jon,

any link available for me ?

Kind regards,

Ben
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi Ben,

I sincerely apologize for for the delay. Brian hasn't posted anything up yet regarding board mods, as he's still investigating further.

However, so far it seems that FB12 is one major culprit; there's still some analysis ongoing WRT FB11.

We've found that replacing FB12 with a 0 ohm resistor (i.e., a short) addresses these 200 - 250 kHz spurs. This is shown below:
bladeRF_fb12_shorted_mod.jpg
Here's a couple captures with and without the modification (with no calibrations or corrections applied). Note that these are two different boards, so there may slight deviations between them. (i.e., differences in the noise floor or other attributes are not necessarily related to the change.)

Without the modification:
without_mod.png
With the modification:
with_mod.png
Let me know if you have any questions!

Best regards,
Jon
pe2bz
Posts: 12
Joined: Sat May 02, 2015 11:49 am

Re: Signals appear multiple times

Post by pe2bz »

Jon,

I have had email contact with Brian who helped me out of my problems.

I replaced FB12 and Z6 and Z7 with a drop of SN40PB60 ;-)

Now, I don't see the "spurious" anymore and I can see weaker signals near stronger repeaters instead of 4 times the strong repeater.

It bothered me a long time but now it's gone.


Thanks for the support !

Ben - PE2BZ
jynik
Posts: 455
Joined: Thu Jun 06, 2013 8:15 pm

Re: Signals appear multiple times

Post by jynik »

Hi Ben,

Sorry for the trouble, but thank you for sticking with us and writing back to confirm your results! The team's working to have these components changed moving forward.

Cheers,
Jon
kk6rum
Posts: 6
Joined: Fri May 20, 2016 3:37 pm

Re: Signals appear multiple times

Post by kk6rum »

Does this issue affect all boards? And is the remedy the same if so?
Post Reply