Connect with us

output difference in tps43000 ic with and without load

Discussion in 'General Electronics Discussion' started by selva, Apr 27, 2010.

Scroll to continue with content
  1. selva

    selva

    25
    0
    Apr 26, 2010
    Hi all,

    I am using tps43000 IC in my circuit (PWM controller). The input voltage to this IC is from a battery (3.7V 4250mA). The output voltage from the PWM controller should be 3.3V and 5400mA. The output of PWM controller is connected to different loads.

    The problem is the PWM controller doesn't provides the required voltage (3.3V) when its connected to load, without a load connected to PWM controller it gives 3.3V as desired.

    i don't understand why happens. Can any one explain me the cause?

    replies much needed.

    warm regards,
    selva
     
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    The output from a PWM controller is a variable duty cycle square wave.

    How are you measuring the voltage? On an oscilloscope, or using a voltmeter?
     
  3. selva

    selva

    25
    0
    Apr 26, 2010
    we are using oscilloscope. i wanna mention here Ic43000 is a PWM controller IC. I did a little mistake in mentioning it above.

    warm regards,
    selva
     
    Last edited: Apr 27, 2010
  4. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    Just looking at your figures in the original post. It appears that you are aiming for greater than 100% efficiency.

    Is this the device you're using?
     
  5. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    If that is the device, it is quite likely that there is some minimum required load to ensure regulation.
     
  6. selva

    selva

    25
    0
    Apr 26, 2010
    Yes that is the kind of device i am implying.

    Actually i am providing 3.7V (avg) battery output to the IC43000 and i am expecting 3.3V to drive my circuit. I don't think thats greater than 100% efficiency expected, do i?:p
     
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    but you said

    By my calculation that's just a tad over 113% efficiency required there.

    I'm certain at least one of those figures is wrong :)
     
  8. selva

    selva

    25
    0
    Apr 26, 2010
    yes, But in my circuit i have connected the BUCK pin to ground, so that the PWM controller works either in boost or SEPIC topology. In such case its capable of producing 17.82W power (3.3V 5400mA) as expected.

    The total required power of the load given to the PWM controller is 9.03573W.
    yet it throws the same problem.

    warm regards,
    selva
     
  9. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    OK, let's just get this clear.

    You are expecting a power output of 17.8W with an input power of 9.04W?

    I may be misunderstanding you.
     
  10. selva

    selva

    25
    0
    Apr 26, 2010
    Ok done.

    1) The output power from my battery is 15.725W (3.7V 4250mA) thats the input to PWM controller.
    2) I am making the PWM controller to work in either boost or SEPIC topology and i am expecting its output to be 17.82W (3.3V 5400mA).
    3) Finally i am connecting a load to PWM controller that requires 9.03573W.

    What actually happens here is, the PWM controller works perfectly producing 3.3V without a load. But when its connected to a load its fail to produce the desired voltage required.

    The PWM controller used is TPS43000 IC

    Hope i am clear now :)
     
  11. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    Ah, I see. It should be capable of 17.8W and your load requires approx 9W

    Have you measured the output voltage at various loads to see when (i.e. at what current) it falls out of regulation?
     
  12. 55pilot

    55pilot

    434
    3
    Feb 23, 2010
    There is no way to tell without a schematics of the actual circuit you are using. Are you using a demo board, have you done the layout yourself or are you using a breadboard?

    For a DC-DC converter, layout is as critical as the schematic. A DC-DC converter can look good on schematic but not work at all because the layout is so messed up.

    ---55p
     
  13. selva

    selva

    25
    0
    Apr 26, 2010
    yes its done, i replaced the original load with the 2 ohms load. The same problem arises.

    note: whenever the problem arises the board gets heated with its connectivity boards (if connected).
     
  14. selva

    selva

    25
    0
    Apr 26, 2010
    Actually its one of my company project's and it has been dropped due to some reason. They had used LTC3785EUF in replacing TPS43000. I took this out to look for an actual reason why it happens.
    yes i do have the board, layouts designed by my seniors.

    Can you mention what kind of problems will be there in layout?
     
  15. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    Stray inductance

    Incorrect grounding

    They would be my top 2. I bet 55p will come up with a heap more :)
     
  16. 55pilot

    55pilot

    434
    3
    Feb 23, 2010
    Have you considered that the reason it was dropped was because it did not work? Before you waste a lot of your and others' time, you should find out why the project was dropped.
    Obviously not senior enough to get it to work. There is not much anyone can do for you without your posting that information.
    I am not going to spend the next several hours writing a thesis on the subject. Almost every manufacturer that makes DC-DC converter chips has several application notes on the subject. If there are specific issues in your design, I will point them out.

    ---55p
     
    Last edited: Apr 28, 2010
  17. selva

    selva

    25
    0
    Apr 26, 2010
    Have you ever consider why i won't, but i do. Due to lack of time they jumped to other IC and Remember every problem has a solution only thing you should take some effort to sort out things.
    i working on this on my spare time.

    So do you expect me to sit ideal till i reach that position.

    i am not expecting you to explain me the entire subject. I am working on it and i need a bit guidance to make myself clear.


    thats what i needed and thanks for that.


    Thank you steve i'll look at it.
     
  18. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,267
    2,715
    Jan 21, 2010
    I'd look at the datasheet for the chip in question, and pay especial attention to any suggested component layout the manufacturer gives. They may also recommend particular types of components in certain places.

    You should verify your design follows those recommendations and where it doesn't, ensure that the part/layout/technique is in no way inferior to the one recommended.

    Although you've been anything but specific about the project itself, check to see that the IC you're using is actually capable of the output you want. One that I looked up was specified for a far lower output voltage, and it may not be able to sustain a high enough mark space ratio to achieve the output you desire at the current you're after.
     
  19. davenn

    davenn Moderator

    13,424
    1,803
    Sep 5, 2009
    looking through the actual datasheet for the IC, I would be suprised if it is able to
    supply the levels of current that you are expecting from it.

    apart from that the component selection, component layout, the oscillator freq, to name
    3 things are quite critical to the operation of the device

    datasheet... http://focus.ti.com/lit/ds/symlink/tps43000.pdf



    cheers
    Dave
    VK2TDN
     
  20. selva

    selva

    25
    0
    Apr 26, 2010
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-