• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 1646 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.

TPS54160 Not Regulating

Bill Keith
Intellectual 300 points
Other Parts Discussed in Thread: TPS54160, TPS54040A, TPS54040

I need some help right away.  We are using your TPS54040A in a design with a rechargeable battery.  It works great unless the battery is low and needs charging.  Then we sometimes have a startup issue where it locks up around 1.5V.  When the battery needs charging the in-rush is close to 1A so I understand why it hangs up.  We dropped in the TPS54160 expecting the problem to go away.  It comes up all the time even with an added 250mA load so it solves that problem but now the output has a 1V ripple on it where the older design had 50mV.  I have used the switcher pro designer and adjusted my values but I can’t get rid of the ripple.

Attached is the 5404 design
My requirements
Vin Min = 7V at the chip
Vin Max = 34V at the chip
Iout Max = 0.750A
Iout Min = 0.050A 

Thanks.

  • 0
    TI__Guru**** 191445 points
    Can you post the conditions where you see 1 V ripple? Waveforms would be helpful. Please show Vout, PH, Vin and COMP.
  • 0 in reply to JohnTucker
    Intellectual 300 points

    It is much worse when charging the batery at 0.5A load, otherwise the ripple is about 0.4V at 50mA load.

    Captures were taken at 12V in and 0.55A load.

    Vin, Vout, COMP, PH, PH Expanded

  • 0 in reply to Bill Keith
    TI__Guru**** 191445 points
    Looks like your ripple is very low frequency, not at the switching frequency. It seems you have rather a lot of ripple on the input as well (and the output ripple may be tracking the input ripple). I would try shorting R39. Maybe add 100 uF electrolytic in parallel with C8. It may also be your load current is fluctuating. Can you try measuring the ripple into a purely resistive load? Can you try using a DC power supply as the input directly across C8?
  • 0 in reply to JohnTucker
    Intellectual 300 points
    I did try using the DC supply directly across C8 prior to posting my question with the same results. The load isnt fluctuating as it is essentially a resistive load at this time, i have striped pretty much everything else out. Why would the TPS54040 work absolutely great and the TPS54160 work so horribly, when they appear to be the same part except one can drive more current?
  • 0 in reply to Bill Keith
    TI__Guru**** 191445 points

    Did you re-compensate your circuit?  The transconductance of the power stage is different between the TPS54040 and TPS54160.  Otherwise the performance should be very similar.

  • 0 in reply to JohnTucker
    Intellectual 300 points

    Yes I did per the program.

  • 0 in reply to Bill Keith
    TI__Guru**** 191445 points
    In the last two screen shots, you can see that the current is dropping to zero. Is it possible to measure the inductor current directly? Do you have access to a current probe?
  • 0 in reply to JohnTucker
    Intellectual 300 points

    No I don’t have a current probe for something like that so I don’t have a convenient way to measure that.  What could cause the noise to be 1V on a 5V output?  Could the compensation network be that critical or is it more likely the inductor.  Should I use a smaller inductor if it appears the current is droping to 0?  I have never had this much trouble with any regulator I have worked with in the last 15 years.  Seems like I am missing something basic for this type of problem.  If I was trying to get the ripple down from 150mV to 50mV I could understand tweaking the compensation  network and some other things, but that isnt where I'm at.

  • 0 in reply to Bill Keith
    TI__Guru**** 191445 points
    its often difficult to troubleshoot this type of issue over a forum. It looks to me like the output voltage ripple is tracking the input voltage ripple. The ripple frequency appears to be about 6 kHz. That is right in the range of possible crossover frequencies. The p-p inductor ripple current during normal DCM operation is about 114 mV. The time for the current to drop to zero after the last on time is about 7.5 usec. That implies the current was at about 1.72 A peak when the high side shut off. It seems just below current limit, so maybe you are reaching current limit. In any case, your load current is definitely varying. You could also be unstable. It would great help if we could see the actual inductor current. Alternately, can I see COMP and SW node on the same scope shot? Both 100 usec/div and 5 usec / div. Also can you post your final schematic with TPS54160?

    Who do you normally work with at TI? Do you have an FAE you typically work with?