Post Go back to editing

ADF4155 as replacing part for LTC6947

Category: Hardware
Product Number: ADF4155

Dear support,

I adopted the ADF4155 as replacing part for LTC6947 (obsolete) in a 4.2 to 4.5 GHz synthesizer design with less than 10 us channel switching time.

I faced some restrictions in the chip functionalities that suggest me maybe the ADF4155 is not the correct choice for the replacement, in particular:

  • the double buffered registers of ADF4155 increase the register writing time with respect to the LTC6947 that is not double buffered;
  • the 32-bit registers of ADF4155 compared with the 8-bit  registers of LTC6947 further increase the register writing time;
  • the RFout port frequency is limited to the half of the operating frequency of the PLL chip (8 GHz), therefore the RFout port is not suitable for mute-till-lock operation above 4 GHz. 
  • if I attempt to change the register configuration in order to switch the CP from the locking value to a lower tracking value, the output frequency shows unlock events. In order to overcome this limitation I programmed the CSR functionality; this solves the unlock events, but still spectral purity and the phase noise level of this CSR solution was not as good as with LTC6947 CP switch. Moreover, using the CSR functionality further decrease the PFD reference frequency to 75 MHz maximum.

Could you suggest some possible solution with the ADF4155 to the following issues:

  1. Switching the CP value from the locking (5 mA) to the tracking PLL phase (0.313 mA).
  2. Increasing the PFD frequency up to 100 MHz.

Eventually, could you suggest a different chip for a a 4.2 to 4.5 GHz synthesizer design with less than 10 us switching time.

Thank you in advance for your reply,

Luca

  • Hi,

    Apologies for the long delay on this. 

    I'm not sure if the summary below has answered your questions. 

    Just to add additional comments

    the RFout port frequency is limited to the half of the operating frequency of the PLL chip (8 GHz), therefore the RFout port is not suitable for mute-till-lock operation above 4 GHz. 

    I was unclear on this statement as the ADF4155 does not contain a VCO. Is it possible that this is being confused with another part?

    If you just need 4.2 to 4.5 GHz, then a discrete solution might be a better approach as you could select a narrow band VCO. One option would be using an ADF41513 PLL + HMC586 VCO. No calibration would be required. So the tuning voltage would be the dominant factor. 

    While you would still have the double buffering requirement, a faster lock time could be achieved. 

    Regards,

    Kieran