• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 891 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS65094: The Jitter issue of TPS650942 VCGI power rail(Buck 2)

Alec Chen
Intellectual 730 points
Part Number: TPS65094

Dear Sir,

I am Alec Chen form WT and our customer is currently using TPS650942 in their design(please see the schematic as below). And we discover that the Jitter for the Buck 2(VCGI) power rail is over RD's spec (within 25%) . Please also see the Jitter waveform that we measure shown as below. The first one is the 4A load waveform and sec one is the 21A(max load). We check TPS650942's spec and it seems that there are not much that we can change to improve the Jitter . Therefore , Could you kindly give us some advice about improving the Jitter issue? Thanks in advance

Above, please kindly help us and thank in advance. if there is any question,please feel free to contact with me

Cheers

Alec

  • 0
    TI__Expert 4155 points
    Alec,

    Are the lengths of these pulses actually changing or is only the frequency changing? It looks like only the frequency is changing, which wouldn't make sense if the load was 21A. If you try putting the BUCK in forced PWM mode does this go away?
  • 0 in reply to Nick Jakse49
    Intellectual 730 points

    Dear Jack,

    We did double checked when we measure the wavform, it is not frequecy changing and it is the COT mode Jitter. we make sure that the PWM is out of low-load skip mode(that is why the first waveform is 4A load). We will try to use "forced PWM" mode and see if there is any improvement. and also, please kinldy let us know if there is a any other solution that we can improviee this jitter issue(RD can't go on the next stage testing without fixing this). thanks in the advacne

    Hope to hear form you soon

    Cheers

    Alec

  • 0 in reply to Alec Chen
    Intellectual 730 points
    Hi Nick, Sorry for the type error as you as Jack. sorry about that!!
  • 0 in reply to Nick Jakse49
    Intellectual 730 points

    Dear Nick,

    We have checked the forced PWM mode for the VCGI as you suggested and it seems no improvement or the Jitter. besdie that, we also mearure the VNN power rail and it seems also have large Jitter as VCGI power rail, please the schemaitc and waveform as below. please kinldy guve us some suggestion about impvoin ghte Jitter issue. thanks in advacne

    Above, please kindly help us wih the issue and hope to hear form your soon

    Cheers

    Alec

  • 0 in reply to Nick Jakse49
    Intellectual 730 points

    Dear Nick,

    We have visited RD for the Jitter issue and it seems the solutions that we suggested didn’t improve the Jitter and here are the solutions that we have tried so far

    1. Change to PFM into the forced PWM mode ==> no improvement for the Jitter
    2. Add a capacitance between the Sense+/ Sense- of VCGI (we tried 0.1nF and 1nF)  ==> no improvement for the Jitter
    3. Try to place PC122 0.1uF on VNN FB pin(we also try 1nF, 0.1nF) ==>  no improvement for the Jitter

    RD would like us to fix this issue soon as we can or it will delay the next built schedule .Therefore, please kindly give us some suggestions soon as you can. Thanks in advance

    Above, if there is any question ,please feel free to let me know

    Cheers

    Alec

  • 0 in reply to Alec Chen
    TI__Expert 4155 points
    Alec,

    I have looked into this more and I believe that the reason there is this jitter is because the BUCK controllers use the DCAP2 design, which gives the BUCK improved transient performance at the cost of a more consistent switching frequency. As such I don't believe there is going to be a way to get this jitter to go away. Perhaps the customer will be okay with this if they know it is not unexpected behavior?
  • 0 in reply to Nick Jakse49
    Intellectual 730 points

    Dear Nick,

    According to your explanation, It might not be not able to get the jitter to go away because of the DCAP2 mode, But  is there any solution that we could try to reduce the jitter?please let us know if there is.

    And if the jitter of TPS650942 cant be improved , could you kindly give us your input on the questions as below?

    1. the current Jitter of TPS650942 is around 30%-35%(we measure the power rails other than VNN and VCGI, and it seems other power rail have about the same jitter). Is this jitter considered ok still by TI design?

    2. Will there be any rick or impact to their power design with this jitter performance?Will the jitter impact the Intel CPU performance?

    Above, please kindly help us with those so we could try to explain to the customer. Hope to from you soon as possible

    Cheers

    Alec

  • 0 in reply to Alec Chen
    TI__Expert 4155 points
    Alec,

    I don't think there is a way to reduce this jitter.

    However, this is to be expected given the design of these BUCKs, and should not be considered an issue. This also should not have an impact on the CPU performance or pose any risk to the power design. The benefit of increased transient response is more important than having a consistent switching frequency.
  • 0 in reply to Nick Jakse49
    Intellectual 730 points
    Dear Nick,
    Thanks for your expaination, we will dicuss with RD about the advice that you provide and if there is other question form the cusorme , we will let you know.

    Cheers
    Alec