Post Go back to editing

ADF5356-Issues with frequencies above 4 GHz

Hi there,

I was just testing out a custom ADF5356 board and was experiencing significant power loss after 4 GHz.

The custom board is essentially a copy of the evaluation board, with a few minor differences to fill a different need. There are 2 points I am interested in:

1--> @4.8GHz

2--> @roughly 9.5GHz

The power reading @ 4.8GHz is similar to that experienced by the evaluation board, but the power reading @ 9.5GHz jumps from -9DBm (eval board) to -24dbm (custom board). I have been trying to perform a number of troubleshooting procedures to try and identify what the problem is and whether or not the chip used is faulty or if there is a problem on my end.  

I have already ensured that the register 6 settings are as they should be and was hoping you might be able to give me more insight on this problem. I can include more details on the board itself if you are interested. 

I have gone through various forums and have found others with 'similar' issues and I have tried to ensure that those were either tested by us or had already been accounted for.

Thank you very much,

Any help on the matter would be greatly appreciated

  • Hi,

    I have two questions:

    - Are you using same cable while measuring the output power of evaluation board and custom board? 

    - You indicated that custom board is a copy of evaluation board. Does this mean you used same BOM as well, particularly 10 pF series cap and SMA connector on RFOUTB?


  • Hi Kudret,

    1st Q- I am using the same cable while measuring both the Eval Board and the custom board, but it after comparing your reference design (Eval board) to our custom board, it has come to my attention that we are using Johnson SMA connectors as opposed to the Rosenberger connectors (ref design) for all the SMA's. The Johnson connector is rated for up to 18GHz (we run max 10GHz) just like the Rosenberger connector, but I do realize that the landing pad for the Rosenberger is much narrower than the Johnson footprint.

    2nd Q- If I am not mistaken we are using the same series cap, but as just mentioned, the SMA connectors are not the same, I am currently trying to figure out if it was a case of unavailability of the part itself, or a decision made for other reasons.

    With this in mind, I was wondering if you think this alone should account for the drop in power read at the RFoutB @ approximately 9GHz (or anything over 5GHz).

    ---I was wondering if ADI has dedicated FAE's that I could get in contact with regarding the possibility of reviewing our custom layout in case there were other issues I may have missed.

    Once again, thank you very much

  • Hi,

    If you changed the trace width to match the pad size, then line impedance can vary which can lead high reflection at the output. Can you calculate the impedance of output trace that you use?



Reply Children