• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 367 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.

TPS40131: Discrepancies in Checking Calculation for τs and DCR

Roy Zhang
Prodigy 90 points
Part Number: TPS40131
Other Parts Discussed in Thread: TPS40322

Hi TI Expert,

I'm looking for a 24V input, 2-phase, non-droop controller as the Vcore solution for non-X86 processor based application while TPS40131 comes into sight. Can you please help to clarify the following cases as well as the question that come out of my checking calculation when studying the documentations (i.e. the datasheet slvs635 and the EVM slvu198a)?

Case 1:  By Equation 42 in the datasheet, direct calculation results τs = 4.518 μs, referring to the attached Excel file. But reverse derivation of Equation 44 comes up with τs = 3.437 μs.

Case 2:  By Equation 15 relevant description in the datasheet and Table 3 BOM in the EVM, the DCR of the selected power inductor presents 1.6 mΩ (nom) / 1.9 mΩ (max), referring to the attached PDF. But by reverse derivation, it can be figured out that "From Equation 52 |Tv(j2πfc)|=1 the compensator gain is solved as 2.09x10^4" is actually basing on the condition of DCR = 1.26 mΩ.

Question: How come there are such discrepancies? Did I understand and/or calculate something wrong or whatever?

Thanks and regards,

Roy

TPS40131_τs.xlsx

ihlp-5050fd-01.pdf

  

  • 0
    TI__Genius 11095 points

    I agree these looks like a typo errors. A few of the equations do not add up correctly. Also I believe we should use Leq = L/Nphases for small signal modeling considerations. 

  • 0 in reply to Matt Schurmann
    Prodigy 90 points

    Matt, thanks for your feedback.

    Well, let me make it clear to avoid mis-understanding —— in multi-phase application, when applying Equation 42 to calculate τs, which is a component of the small signal modeled COMP-to-Output Transfer Function Gvc(s) i.e. Equation 41, Leq = L/Nphase shall be taken into account.

    Is my above understanding correct? Besides, should Rout and DCR be left intact, or divided by Nphase?

  • 0 in reply to Roy Zhang
    TI__Genius 11095 points

    Hi Roy, 

    No, you should not scale DCR or Rout. The quantity Ac*DCR represents the current sense gain. ROUT is the total load (Vout/Iout). 

  • 0 in reply to Matt Schurmann
    Prodigy 90 points

    I got and read it, Matt. I can understand L/Nphase is common in small signal modeling for multi-phase case, and similar treatment also appears in the compensation calculation of TPS40322, though it is a VMC part. Thanks and take care.

  • 0 in reply to Roy Zhang
    TI__Genius 11095 points

    Hi Roy,

    I discussed it with someone who knows this device a little deeper than me, and we decided both L and DCR should be halved. The reason its not shown in the equation, is that DCR/Nphases / L/Nphases, cancels out in the eq 42. 

  • 0 in reply to Matt Schurmann
    Prodigy 90 points

    Hi Matt,

    Greatly apppreciated for you clarifying the understanding of Gvc(s) transfer function, up to now I have no doubt of Equation 42 any more. Actually the idea of “equal scaling cancelation” did flash across my mind when I was staring at the equation, and eventually I've realized there is no comparison between TPS40131 (CMC) and TPS40322 (VMC) at this point. Sorry for my mis-leading.