• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 455 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.

LM25141-Q1: Extremely slow startup (~20s) and weak stability under load

Arndt Kritzner
Intellectual 720 points
Part Number: LM25141-Q1
Other Parts Discussed in Thread: LM25141

Hi,

with the below LM25141 converter we experience unexpected behaviour. Input voltage is 10..30V, output should be 5V (~10A). The circuit has a load-side capacitor with 3F capacitance and about 50mOhms.

We would have expected, that the output voltage raises within about 2 seconds to 5V (at current limit) and then stable 5V is continuously available at the output.

Instead of this the output voltage increases very slowly - within about 20 seconds - to 3V and then raises with about the expected rate to 5V. Once 5V is reached this is weak under load and drops to 4.2V at 3A load. If at power on an electronic load with e.g. 2A preset load current is connected the output voltage does not raise at all.

Picture 1: Circuit schematic (PDF):

PS-5V10A.pdf

Picture 2: Voltage drop by 0.8V as load steps to 3A (from no load):

red: VIN, white: VOUT, green SS

Picture 3: Startup period with very slow voltage increase:

Any ideas what's mistaken and how to solve this (or what additionally to check).

Kind regards

Arndt

  • 0
    TI__Mastermind 22301 points

    Hi Arndt,

    Your 3F load capacitance is causing these issues.

    During startup the 3F capacitance causes the controller to be in current limit for more than 512 cycles, and the controller goes into hiccup mode as described in the datasheet. This is why soft start drops and recharges. You need a soft start capacitor to match the expected charging time to prevent this from happening.

    Also, the 3F is in parallel with the 220uF designed output capacitance, so the controller effectively sees 3F output capacitance. The compensation network designed seems to be designed for 220uF, so this control loop error is most like causing your load regulation and setpoint issue.

    I suggest using the design calculator to choose a softstart cap appropriate for 2 or 3s. You also need to use the calculator to redesign your compensation network with an output capacitance of 3F. Your loop bandwidth will have to decrease to get reasonable values for the compensation network.

    Please try these suggestions and let me know if you still have issues!

    -Orlando

  • 0 in reply to Orlando Murray
    Intellectual 720 points

    Hi Orlando,

    thanks for your reply. We changed the circuitry (see below) of our test board after the values calculated by the design calculation tool (see attached). This solved the startup issue but not the steady operation problems. But so the capacitor charging is time-controlled instead of beeing current-limited.

    From the design calculation tool we would expect an overall efficiency of above 90% but instead the circuit is running at about 65%. The voltage still drops under load (3A) from 5V to 4.2V.

    As I understood the design calculation tool does not calculate a parallel capacitor for the compensation network, but without (C51) the efficiency drops further, below 60%.

    Any ideas to stabilize this circuitry or is this chip just not intended for use in this manner?

    Kind regards

    Arndt

    0677.LM5141ADESIGN-CALC.xlsx

    8182.PS-5V10A.pdf

  • 0 in reply to Arndt Kritzner
    TI__Mastermind 22301 points
    Arndt,

    I would try Rcomp = 1Mohm, Ccomp = 1.5uF, and CHF = 150pF.

    The output capacitance is 3F, so actually 3,000,000 uF.
    Entering that in the quickstart calculator, and lowering the bandwidth to 150Hz suggests a Rcomp of 1Mohm.
    The bandwidth has to be reduced because with such a high 'C', C*dV/dt is limited by the (output) capacitor current Ic.
    Otherwise a high bandwidth would suggest unreasonable Rcomp in the quickstart.

    That then suggests an Ccomp of 1500nF, or 1.5uF.
    And with a higher Rcomp CHF should be reduced accordingly.

    Please try this compensation network and let me know how it goes.
    -Orlando
  • 0 in reply to Orlando Murray
    Intellectual 720 points

    Orlando,

    thanks for your suggestion. We changed our test circuitry accordingly, but this only worsened the behaviour and the output voltage stops at 2.6V.

    Kind regards

    Arndt

  • 0 in reply to Arndt Kritzner
    TI__Mastermind 40040 points
    Hello Arndt,

    I reevaluated and I calculate the following.

    I would try Rcomp = 1k, Ccomp = 4.7nF, and CHF = 470pF and retest.


    Hope this helps?
  • 0 in reply to David Baba
    TI__Mastermind 40040 points
    Hello Arndt,

    We did not hear back from you. I will close out this thread. If you still need assistance, please reopen so we can help resolve.

    Thanks.

    David.