Hello,
It seems that LimeSDR local oscillator level depends on frequency. It seems that every 30kHz LO level is 6dB higher. The same is for TX and RX. As a result transmitted power is 6dB higher(measured using spectrum analyzer). Also received signal is 6dB higher on those frequencies(external test signal was used as source). What could be the reason for that? Is there any way to overcome that? Is it normal and/or expected?
EDIT:
I am using SoapySDR API. In gqrx I don’t see this effect.
That’s brilliant!
Do you have a starting frequency where you observed this effect so I compare what you got to mine?
There is something wrong with how I understand this. If every 30 kHz you increased 6db then you would be up 60db by 300kHz. That would take a milliwatt to a kilowatt. I will hook mine up to a radio-communication tester and check this out with sdrangle on tx and rx. I can generate and analyze calibrated signals up to 1GHz.
I what way are you using SoapySDR? Did you write some code?
It’s not linearly growing. Every multiple of 30khz has 6dB higher level than not 30khz multiple.
EDIT:
I narrowed it down to CMIX bypas bit in register 0x208. When LO frequency is multiple of 30kHz the CMIX bypass gets automatically enabled. It results in 6dB higher signal. It also can be seen in LimeSuiteGui during selftest. Set NCO to 0 MHz and try switching CMIX bypass on and off.