Hello,
Due to some alias and image problems I had been experiencing (and still am) it was suggested to me in another thread that I try to calibrate the bladeRF using the "calibrate" command at the cli prompt, which I have done. I'm still trying to puzzle out if it is helping. But my question is pertinent to warning messages I receive when running calibrate at the command line, as shown below:
bladeRF> calibrate
[WARNING] Never converged - DC_CLBR_DONE: 1 DC_LOCK: 0
[WARNING] Never converged - DC_CLBR_DONE: 1 DC_LOCK: 0
Can someone provide an explanation of exactly what this means? Also, are there any other commands that may further assist with reducing imaging and aliasing? I see a "correct" option but I'm not sure what "arguments" to provide, or if it would even help; it is pertinent to IQ imbalances.
Finally, is there a "for dummies" in depth explanation of these commands available anywhere, and also for the various commands available to me via SDRConsole to include VGA1, VGA2, LPF, and LNA settings? I was pointed to the website for the chipset that bladeRF uses but unfortunately for me the data was above my level of comprehension for the most part.
Phew! That's enough for now....thanks to anyone in advance!
The "Calibrate" Command and a few other ??'s
-
- Posts: 83
- Joined: Tue Jan 21, 2014 6:38 pm
- Location: Cape Cod, Massachusetts, USA
- Contact:
-
- Posts: 303
- Joined: Mon Mar 04, 2013 4:53 pm
Re: The "Calibrate" Command and a few other ??'s
What that means is that there was some internal calibration registers that were written, and DC_LOCK was supposed to be active when the DC_CLBR_DONE was also set. We are actively working on the calibration routines to build them into the frequency tuning and gain settings as well as remove that DC bias in general.bladeRF> calibrate
[WARNING] Never converged - DC_CLBR_DONE: 1 DC_LOCK: 0
[WARNING] Never converged - DC_CLBR_DONE: 1 DC_LOCK: 0
Can someone provide an explanation of exactly what this means? Also, are there any other commands that may further assist with reducing imaging and aliasing? I see a "correct" option but I'm not sure what "arguments" to provide, or if it would even help; it is pertinent to IQ imbalances.
When it comes to the correction parameters, the doxygen documentation is the best to take a look at the bladerf_correction enum.
If you're looking for help on the various CLI commands, typing 'help' is a good start. If you want to try different calibration routines, then you can try just typing 'help calibrate' and you should see:Finally, is there a "for dummies" in depth explanation of these commands available anywhere, and also for the various commands available to me via SDRConsole to include VGA1, VGA2, LPF, and LNA settings? I was pointed to the website for the chipset that bladeRF uses but unfortunately for me the data was above my level of comprehension for the most part.
Code: Select all
bladeRF> help cal
calibrate [<module>]
Calibrate the DC offset for both RX and TX
Available modules for calibration:
tuning
txlpf
rxlpf
rxvga2
Leave blank to calibrate all of the above.
You should be able to do tuning, rxlpf and rxvga2 with no problem. Usually it's the txlpf that seems to give me a little trouble - though I am not sure why. We're working it.
This is a lot to digest. Let me know what other questions you may have.
Brian
-
- Posts: 12
- Joined: Fri Mar 01, 2013 4:32 am
Re: The "Calibrate" Command and a few other ??'s
I get a similar response:
I've also tried running the kalibration-bladeRF GSM calibration routine, but after it's given me the CAL value, I can't seem to write it to the device:
Is the VCTCXO DAC calibration value meant to read "0x9c7a" after that command?
Code: Select all
bladeRF> calibrate
[WARNING] Never converged - DC_CLBR_DONE: 1 DC_LOCK: 0
Error: An unexpected error occurred
Code: Select all
average [min, max] (range, stddev)
- 45Hz [-130, 14] (145, 41.442207)
overruns: 0
not found: 0
Found lowest offset of -8.418392Hz at 944.400000MHz (-0.008914 ppm) using DAC trim 0x9c7a
Code: Select all
bladeRF> info
VCTCXO DAC calibration: 0xfffc
FPGA size: 115 KLE
FPGA loaded: yes
USB bus: 4
USB address: 3
USB speed: SuperSpeed
Backend: libusb
Instance: 0
bladeRF> flash_init_cal 115 0x9c7a
[INFO] Erasing 1 blocks starting at block 3
[INFO] Erased block 3
[INFO] Done erasing 1 blocks
[INFO] Writing 1 pages starting at page 768
[INFO] Writing page 768
[INFO] Done writing 1 pages
bladeRF> info
VCTCXO DAC calibration: 0xfffc
FPGA size: 115 KLE
FPGA loaded: yes
USB bus: 4
USB address: 3
USB speed: SuperSpeed
Backend: libusb
Instance: 0
-
- Posts: 455
- Joined: Thu Jun 06, 2013 8:15 pm
Re: The "Calibrate" Command and a few other ??'s
After the flash_init_cal command, a power cycle is required for the new settings to take effect. The calibration settings are read from flash by the firmware when the device boots. I'll add a message to this command that states something to this effect.rowan.doherty wrote:Is the VCTCXO DAC calibration value meant to read "0x9c7a" after that command?
As far as the calibration command goes, it's broken in master and we're actively working on DC calibration over the next few days. These changes will be pushed to the dev-cal branch on GitHub. I'll post back here when something usable is up there.
In the meantime, you could quickly dial things in manually, if need be..