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.

TPS650864: Efficiency and power sequencing

Part Number: TPS650864
Other Parts Discussed in Thread: TPS650860EVM-116, TPS650861

Power team,

We have a customer considering the TPS650864 to power a Zynq Ultrascale+.  Can you help with a few questions?

If they get the TPS650860EVM-116, can they configure the device to behave exactly like a TPS650864?

They will need to have one additional buck in the system, and they want it aligned with Buck 2 in the power sequencing.  It looks like this is possible using the PG signals for Buck 1 and Buck 2.  Is there a practical way to make Buck 6 wait for Buck 2 AND the external Buck without adding external logic?

How do customers calculate the efficiency of this PMIC?  Do we have a spreadsheet or other tool?

On page 109, there's an ITRAN value used to calculate the cap.  Do you have a value you recommend for this?

Thanks,
Darren

  

  • Hi Darren,

    We have a large amount of support right now but I will provide a response to your questions within the next 2 business days.

    Regards,

    James

  • Hi Darren,

    • The TPS650860EVM will come with a preset OTP configuration, but the registers can be adjusted according to the methods found in this programming guide:

      TPS65086100 Non-Volatile Memory Programming Guide

      There are different versions of the TPS650864 PMICs. They each have an OTP Generator that can help configure the PMIC registers to match the desired version. These OTP Generator versions can be found on the TPS650861 product page. Scroll down to the section called "Design Tools & Simulation" and you will find all the available OTP Generators. 

    • The TPS650864x PMICs are pre-programmed, so if you want the BUCK6 enable to be based off an external buck, I would recommend you use the user-programmable TPS650861. This PMIC can be programmed to match any of the TPS650864x versions using the OTP Generators I mentioned earlier. Once you set up the registers to match your desired variant, you can go in an make changes to the CTLx and PGOOD sequencing to fit your needs.

      For example, you could set up BUCK6 to be enabled by one of the CTLx pins on the PMIC. Next, use either the VOUT or the PGOOD of the external buck to pull the CTLx pin high and enable BUCK6. As additional security, you can also tie the BUCK6 enable logic to other internal PGOOD signals to make sure it does not activate before certain rails. If the external buck has a PGOOD output, I would recommend you use that instead of VOUT to make sure that the external buck is stable before BUCK6 is enabled. Note that the CTLx pins usually operate on 1.8V logic.

    • We don't have a spreadsheet for calculating efficiency but we have this application report that might be of use: Calculating Efficiency (Rev. A) - Texas Instruments
      Section 7.13 in the TPS650864 datasheet also has some efficiency figures.

    • ITRAN is dependent on the customer use case so we don't provide a recommended value. The figures in Section 9.2.3 of the TPS650864 datasheet might work as an example but I would try to customize ITRAN based on your specific system.

    Regards,

    James

  • Hi James,

    Thanks for the thorough responses.

    There isn't a lot of information in the datasheet on how to figure out what value to use for Itran.  Is the peak current during ramp up the best value to use?  Any other guidance you can give on it?

    Thanks,

    Darren

  • Hi Darren,

    I believe ITRAN is more related to any load steps you expect to see during operation. If you have components that will turn on or off and significantly change the load on the buck you will see the current step up (or down) in a short period of time. The ITRAN value should be the max step size you expect to see on the buck (max transient size). If you have multiple peripheral devices that could all turn on/off at the same time, I believe it would be best to plan for the worst case current step with the ITRAN value.

    Regards,

    James