• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 515 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.

TPS54525/TPS54625 problem with Vin min

Lars Mikaelsson
Prodigy 50 points
Other Parts Discussed in Thread: TPS54625, TPS54525I made a Webench design for TPS54525 with Vin = 7 - 13V, Vout 5V, Iout 4A. Later on I wanted to reduce the Vin min to 6V. The batteries I'm using can be drained to belov 7 Volt and in order to get maximum runtime it seemed to be a good idea. This change seemed to be impossible to do, but the similar chip TPS54625 seemed to work. So I changed the design and recalculated it and webench gave me the design I wanted. I ordered new chip's ant tried it. I was a bit disappointed when I realized that this design didn't work belov 7V either. It behavied exactly the same as the original design with TPS54525. I checked the the dutucycle ant it doesn't go higher than about 80%, after that Vfb(pin 2) starts to drop. If I enter the same values into Webench I can't create the same design as I did last week, it refuses to allow this chip for this design. My board is ready for production so there is no possibilty to change to another type of chip. I can't see the limitations here. I even tried to reduce the current to 3A without success. Is there anyone that have clue of what to do? And why does webench behave like this
TPS54625 5V out 6V in.pdf
  • 0
    TI__Guru**** 191445 points

    TPS54525 and TPS54625 use DCAP2 control mode.  It is adaptive on time, hysteretic off time control.  The maximum duty cycle is limited by the minimum off time  The absolute maximum duty cycle for your condition will be between 81% typical and 76% worst case.  So your minimum input voltage not including FET losses is between 6.16 and 6.57 V.

    Besides this hard limitation, there are two other factors to consider.  Ideally, the TPS54X25 should regulate linearly with input voltage until the maximum duty cycle limit is reached.  in real life, there is a "soft corner" so that line regulation degrades as the maximum duty cycle is approached.  Also, these devices are designed to allow for fast transient response.   When operating near maximum duty cycle, that capability is impaired (the device responds to load step increases by increasing duty cycle.  It cannot do so when maximum duty cycle is reached).  So Webench limits the designs to duty cycles less than 65 %.  Older IC's may have had this limitation added recently.

    So far as I know there is no way to get around the maximum duty cycle limitation.

  • 0 in reply to JohnTucker
    Prodigy 50 points
    So you mean that changes have been made to Webench since last week when I made the attached design with TPS54625? This design allowed dutycycles up to 86%, which then must be completly wrong? And not even my original design with 7V input will be produced safly, even though it seems to work OK on my prototype board. Duty cycle then reaches about 81% at Vin = 7V, and according to the design report it should be possible to reach 86%, I feel that I have been screwed up here! So the only possible solution then is to redesign with Another kind of chip?
  • 0 in reply to Lars Mikaelsson
    TI__Guru**** 191445 points

    I am not sure about the timing of Webench releases.  I do know that duty cycle limitations have been a discussion topic recently.  The 65% limit for Webench is somewhat artificial as I pointed out previously.  These device will regulate properly at duty cycles above 65%, but there is not much headroom for transient response.  Those designs should operate reasonably well down to Vin = 6.57 V, depending on the output current.  I don't use Webench very often and I am not sure about every aspect of it.  if you absolutely need the input voltage down to 6 V, then these devices are not a good choice.