• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 41 subscribers
  • Views 214 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LMR38025-Q1: Weird stability issue simulating LMR38025-Q1 using PSpice

Ivan
Prodigy 10 points
Part Number: LMR38025-Q1
Other Parts Discussed in Thread: LMR38025, LMR38020EVM

Tool/software:

Hi,

 

I’m currently working on designing DC/DC buck converter to step down 24 V input voltage to 15 V.

Simulating transient response I get great results when converter is loaded with higher current approx. 400 mA or higher, not sure of the exact value.  But when lightly loaded like 10 mA, 100 mA, I get this weird “startup” behavior. After some time everything goes back to normal. Can somebody please explain this behavior, is converter unstable? What should I do?

I’m attaching pspice schematic and simulation responses.

 

I have another question which is less important but it would be nice to know.

How likely or is it even possible to obtain values of internal compensation components like gain of the EA, Rc, Cc of internally compensated PCM converters e.g. LMR38025?  

 

Thanks and best regards,

Ivan

Buck_Sch.pdfTI_Resp.pdf

  • 0
    TI__Mastermind 22340 points

    Hi Ivan,

    This is a peak current mode device, which means that the high-side current is being used in the control loop. In this case, the converter starts up at no load. When the device first issues a large load transient, the output will at first drop, and then the device will react accordingly to generate the right error voltage to meet the regulation target. You can see that as soon as the transient returns back to light load, the output voltage returns back to the regulation target. The reason you don't see this at the 1.25A - 2.5A transient is that the sensed current is already added in the control loop. 

    Please see the attached quickstart calculator if you need to check loop response for the LMR38025.

    Quickstart_tool.xlsm

    Thanks,

    Richard

  • 0 in reply to Richard Fu22
    Prodigy 10 points

    Hi Richard, 

    thanks for the reply. But I fear thats not issue in this case, at least I see it that way. I hope Im wrong, but let me add more information. I have simulated same scenario again, same schematic just changed time delay of output load. Again I have some weird voltage drop at 6,2 ms with output loading starting at 7 ms. So before output current even starts to rise. 

    Could this be some kind of pspice model issue, this would not happen with real circuit?

    Im just asking for second opinion should this be concerne or not?

    Again Im attaching pspice schematic and simulation responses.

    TI_Resp2.pdf ; Buck_Sch2.pdf

    Thanks and best regards,

    Ivan

  • 0 in reply to Ivan
    TI__Mastermind 22340 points

    Hi Ivan,

    I believe that issue should not occur with the real circuit, and is a limitation of the PSPICE model. Let me look into this and I'll get you a response by next week. Please keep in mind that Monday is US holiday so response will be delayed.

    Regards,

    Richard

  • 0 in reply to Richard Fu22
    Prodigy 10 points

    Richard,

    take your time and thanks.

    One more thing regarding Quickstart_tool.xlsm, I had multiple runtime errors but I made it work so if somebody tries to use it, I would like to leave my steps of fixing it. First thing I enabled macros the way described in instructions. After that runtime errors persisted and I had to change windows regional settings to English (US). Then it started working. Apparently, something that has to do with decimal separator. You use dot/point, we use comma and it makes excel crash. After that I had couple more crashes, I had to change multiple values (e.g. L, Cout) in small increments to get it to my design values and not crash. Now it works perfect.

    Thanks and best regards,

    Ivan

  • +1 in reply to Ivan
    TI__Mastermind 22340 points

    Ivan,

    Appreciate it. I went into the lab and took a startup waveform with no load on the LMR38020EVM. The input here is 15V, output is 5V, with no load. As you can see, there is no issue with startup. 

    Thanks,

    Richard