• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 55 subscribers
  • Views 324 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC1120EMK-868-915: Why plenty of 47nF capacitors used instead of 100nF caps in the reference design?

David Byrne
Prodigy 150 points
Part Number: CC1120EMK-868-915
Other Parts Discussed in Thread: CC1101, CC1120

Hi.

I am completing a PCB design using this eval board as a reference. I would like to know if all the 47nF capacitors can be replaced with 100nF capacitors on the CC120RHBR? They look like they are just filtering DC levels. Will using 100nF X7R ceramic caps cause any misbehaviour? It seems odd that 47nF caps were needed for filtering, and not 100nF suggested. From a technical perspective, why 47nF?

Also,C61 is 220nF. Can I used a 1uF cermic cap instead?

This will be a high volume PCB of around 20,000+ units per annum initially and growing thereon. I want to reduce to the BOM item count as I am using 100nF caps and 1uF capc elsewhere on the PCB. 

regards,

David.

  • 0
    Mastermind 6160 points

    Generally, decoupling on VDD pins on RF reference designs may have a given value to filter a out a spur at a frequency (if you look at the CC1101 reference designs the decouplings have different value dependent on pin) 

    The CC1120 ref design have only been tested with the values given and the behaviour with different values are therefore not known. But it's my impression that the values are not important on this one based on an experiment I did at one point. I removed the decoupling caps one by one and looked at the output spectrum while doing so. The only one that made  a difference was the decoupling at AVDD_XOSC. I would produce a handful of boards using 100 nF decoupling and see if you see any performance difference.

     C61 is the load on an internal LDO and changing this value to 1 uF could risk making this regulator unstable so  I highly recommend keeping this value. 

  • +1 in reply to TheGhostOf
    TI__Genius 13580 points

    Hi David,

    TheGhostOf has several good points here (as usual) - TI have tested the CC1120 design extensively with the 47nF capacitor value(s) and have not tested it using 100nF, so the effect is unknown. As 47nF produced good performance and this is the design being recommended there has been no reason to test the other values (as the advice is always to follow the reference design).

    So, you can mount a value of 100nF but accept the accompanying risk (and decide upon your level of required testing based upon that).

    The C61 value should be kept as 220nF, also as TheGhostOf states.

    Regards,

    Zack

  • 0 in reply to TheGhostOf
    Prodigy 150 points

    A BIG thanks for your excellent and quick response. The manufacturing engineer was who wanted the change to reduce his BOM item count, but I was reluctant to do so based on what I suspected which was verified in the answers here. If you ever visit Melbourne in Australia, I owe you a beer. Thank you!

    regards,

    David

  • 0 in reply to ZC
    Prodigy 150 points

    Thanks Zack. Your finger is on the pulse  - very much appreciated. I think we will just keep with the recommended values. The time and effort to test (including EMC emissions testing) is not worth the effort for a couple of extra BOM line items. Have a great day and once again thanks!

    regards,

    David