Spur in AD 9364 for TX

If you change the LOL frequency by say 500KHz or 1MHz, are you seeing any shift in these spur frequencies? Hope you are operating the REF_CLK at 1.3Vpk-pk.

Yes, We make sure it using REF_CLK at 1.3V Pk-Pk, Even we checked it by driving from signal generator. We checked with several frequency, As a result of we are getting same offset of 186 KHz.  It is shift in spur.

Since, you are not seeing a shift in the spur offset frequency with LO change, so it is not integer boundary spur.

Spurs at KHz offset range happens mainly because of power supply. The phase noise looks good from the output. All the spurs for this chip are rated at 50dBc from the tone power. You can try adding bulk capacitors to bypass the high frequency noise from power supply and see if there is a reduction in its levels.. Also, try giving the power supply from external source and check.

We checked it with driving from an external power supply and also placed decaps as much as possible but still the same effect. Are you saying, this AD9364 Chip maximum will get spur at 50 dBc from Carrier? Do you think, it caused by AD9364 IC itself causing spur ?.Kindly confirm ASAP !

Hope you are not setting the RX and the TX LO to the same frequencies, as there is a VCO pulling issue internal to our chip that can cause undefined spurs at the output.

Otherwise, it cannot be generated internally. When you are switching to external power supply, are you disabling the internal supply source? Can you share with us the power supply schematic section?

Yes, We disable the internal supply while driving it outside. Please see the below schematic.  

These 186KHz close in spurs are integer boundary spurs. If you shift the LO by say 1MHz, you should see a significant reduction in these spurs amplitude. Refer to the below link for more details:

https://www.analog.com/en/resources/analog-dialogue/articles/analyzing-integer-boundary-spurs.html 

You can frequency plan accordingly to get a reduction in these spurs.

 srimoyi 

Okay. Thanks, 

As you mentioned we shifted by 1 MHz from LO frequency (2 GHz), we can see a significant reduction in these spurs amplitude. But its working for TX channel 1 only when we check the same frequency for Tx channel 2 No improvement in spur level.

TX1 Channel- 53dB

TX2 Channel-47 dB

Whats the reason for that getting different amplitude levels in spurs in TX1 & Tx2 ?

Setup 1: channel -1 Tx frequency = RX frequency = 1.999 GHz, spur level is -53dBm

Setup 2: channel -2 Tx frequency = RX frequency = 1.999 GHz, spur level is -47dBm

setup 3: channel -2 Tx frequency =1.999 GHz , RX frequency = 1.998 or 2 GHz, spur level is -53dBm

Does this behaviour is due to VCO pulling in channel 2 while the TX & RX freq are same ?

If it is VCO pulling why its not happening for channel 1 ?

Yes, it is due to VCO pulling issue. It will happen for both the channels, and is not specific to a single channel. VCO pulling depends on the way the phase of the RXLO  gets added to the TX LO and causes issues, so for TX1 it may not happen for this specific frequency but will happen for other frequencies, if the LO's are kept same. So, its recommended to perform the tests with some offset between RX and TX LO.

Hello Srimoyi,
We have got different spur levels observed while changing the different frequency to intended frequency, let’s say 3 GHz.
Frequency of tone: 2.5 MHz
TX: 3 GHz, this is our desired frequency
Observed in the spectrum analyzer (SA)
Up-Band: -3 dBm at 3.00 GHz
Spurs: -59 dBm carrier offset of 186 KHz (Delta Marker)
When we shifting the frequency to 2.9 GHz for some time and then to 3 GHz
Spurs: -58 dBc carrier offset of 186 KHz (Delta Marker)
Again, we shifting the frequency to 3.1 GHz for some time and then to 3 GHz
Spurs: -49 dBc carrier offset of 186 KHz (Delta Marker)
Again, we shifting the frequency to 3.5 GHz for some time and then to 3 GHz
Spurs: -49 dBc carrier offset of 186 KHz (Delta Marker)
Why we observed 11 dB is difference when we change the different frequency to desired frequency? What is the reason for that?
And also as you suggested to look for integer boundary spur (IBS),It is absolutely works if we shift to 1 MHz offset, let’s say 2.999 GHz
We don’t see any spur above -60 dBc, even 186 KHz is gone to below -60 dBc. 

srimoyi  Look for your reply, Is it possible to connect with you via phone i would be happy. We are struggling this issue for a long.