• State Resolved
  • Locked Locked
  • Replies 19 replies
  • Answers 1 answer
  • Subscribers 68 subscribers
  • Views 4381 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.

TPS61090 shorts on the input depending of the L1 coil

user4413718
Prodigy 60 points
Other Parts Discussed in Thread: TPS61240, TPS61090, TPS61230

Hello,

I have a short on the input pin probably because of a voltage spike on the SW pin. The device is damage when the battery feeding the IC is really low.

It's probably because of my layout (my coil is at 1.5cm of the IC) but it's work (most of time) with this coil :media.digikey.com/.../VLC5045.pdf but it's damage the device with this one : http://www.farnell.com/datasheets/1735537.pdf 

Can you explain me why it's working with one and not with the other coil ? it's because of the different Isat ? can you explain me the layout consequences ?

Here is the layout : 

Thank you in advance for the help and the explanation

  • 0
    TI__Mastermind 46905 points
    the layout is not good, the ceramic capacitor must be very close to the VOUT and GND pin . please follow the user guide to layout the PCB. i think the inductor is not the root cause of the damage in SW pin of this device.

    what is the input voltage and output current in your application?

    thanks
    Jasper Li
  • 0 in reply to Jasper Li
    Prodigy 60 points
    I use two different type of input voltage : USB 9V to 5V via a DC/DC Converter fr.farnell.com/.../2280247) or Lipo (3.7v depending the charge level) and each of this input pass through a MCP73871 battery management IC (this IC output the battery voltage or the 5V usb input) . The output is 5V @260mA.
  • 0 in reply to user4413718
    TI__Mastermind 46905 points
    so the input voltage will be not lower than 3.0V. if the output current is only 260mA, maybe you can have a look at TPS61240, simpler in application.Please follow the user guide for the layout.

    thanks
    Jasper lI
  • 0 in reply to Jasper Li
    Prodigy 60 points

    Thank you very much for your reply.

    So the short on the input is not a consequences of a bad layout ? is-it because my battery when low is under 3.0V ?

    And i don't understand why it's working with one coil and not with the other : any clue ?

    You're very helpfull !

  • 0 in reply to user4413718
    TI__Mastermind 46905 points

    i guess the problem is caused by the bad layout.

    do a experiment in current PCB: place a 1uF output ceramic capacitor closed to the IC and connect the capcitor to VOUT and GND pin with trace as short as possible. check if the TPS61090 is still damage, and if there is the different behavior of the two inductor.

    i meant VIN > 3V in your application, so you don't need use TPS61090. There is better solution, although TPS61090 is also OK in this application.

    thanks

    Jasper Li

  • 0 in reply to Jasper Li
    Prodigy 60 points
    Thank you very much Jasper !!

    It works like charm with the capacitor close to the IC at Vout.
    So is-it a layout issue (the component have to be closer to the IC) ?
    A last question for my knowledge : do you know (before i place the cap as you said) why with the coil TDK VLC5045T-6R8M it worked a bit and not with the WURTH ELEKTRONIK 74408943068 ? (both are 6.8UH)
  • 0 in reply to user4413718
    TI__Mastermind 46905 points
    Good to know you have solve the problem. from current information, the root cause is the bad layout, the COUT is too far with the IC.

    if the COUT is far with the IC, the voltage of SW pin could be higher than the absolute value which has the risk to damage the device. but damage does not always happen depend on the SW voltage level and device variation. i guest this is the reason that you observe some device damage very quickly, some devices last longer.
  • 0 in reply to Jasper Li
    Prodigy 60 points
    It is clearer with this reply, so how the SW voltage is related to the L1 and COUT trace length ?

    Best regards,
  • 0 in reply to user4413718
    Prodigy 205 points
    Hi Jasper ,

    we have a simmilair issue with the TPS61090. our design is also based upon the Adafruit design that was uploaded to the post. So also in our case the soource is a 3.7V (either from a Lipo - or from a supply) but whenevr we enable the TPS it immidiatly is distroid. We do not have an output cap at all at teh moment, will be in at the next release but placing it did not fix the issue. as an experiment we removed the output cap from the adafriuit design but that board remails functional. In our design we also use the Wurth inductor 7447715006 wheras Adafruit uses a differnet one. we also swapped both inductors but the results remained teh same.

    any further suggestions??

    kind regards,
    Martijn
  • 0 in reply to Martijn Broens
    TI__Mastermind 46905 points
    Martijn
    not sure about your meaning of "Adafruit design", could you upload your schematic and layout?

    the output capacitor is the most important components device for normal operation of a boost converter. the capacitor should be placed closed to IC and connected with strong and short track, otherwise TPS61090 could be damaged.

    the inductor is not the reason that the damage happened.
  • 0 in reply to Jasper Li
    TI__Guru**** 191445 points

    "  It is a popular DIY resource.

  • 0 in reply to Jasper Li
    Prodigy 205 points

    Hi Jasper,

    thanks for the reply. By Adafruit I refer to a design posted by adafruit@  nice product.

    Here's the schematics for this part of the design. Note that I have forgotten to add the output caps in the design. and the inductor from Wurth is to be replaced by a smaller footprint version. but as mentioned that may not be the cause.

    The FET's are used to control the enable signal from various sources. here's the layout section.

    I agree with you that the Output cap is vital to the design. we will add these as a patch and see what it brings. I'll report back to you asap

    regards Martijn

  • 0 in reply to Martijn Broens
    TI__Mastermind 46905 points
    hi Martijn
    could you upload the pictures again? the picture can't be open.
  • 0 in reply to Jasper Li
    Prodigy 205 points

    Hi Jasper,

    hm i see. new attempt.

    And the Layout:

  • 0 in reply to Jasper Li
    Prodigy 205 points
    Hi Jasper, for some reason the pictures do not come true. i can see them while editing the msg. but when i pres post they are gone :-(

    Any how. we yesterday evening added the output capacitors and now the design works as intended. thanks for the help. and stupid of me to forget the output caps.

    One question though. I f i remember correct TI has a nice appnote on the theory of operation for switchers. you can see the L the FETs and the capacitors. for all variants, Buck/Boost. can you point me to this appnote i can't find it anymore.

    regards Martijn
  • 0 in reply to Jasper Li
    Prodigy 80 points
    Hi Jasper,

    I saw your post while searching within the forum for a solution of our problem.
    We are also using the TPS61090 in one of our products recognizing that sometimes this product gets damaged and sometimes not. You posted that the output capacitor should be placed closed to the IC... otherwise the device could be damaged.
    So could you please indicate the maximum distance between the IC and the output capacitor?

    Best Regards,
    Patrick
  • 0 in reply to Patrick Olma
    TI__Mastermind 40080 points

    Patrick, 

    1) the decoupling capacitor should be very near the IC, especially for  the ceramic decoupling caps, if you need a distance, then I have to say as close as possible, within 2mm distance, the shorter, the better.

    2) it also depending on how you pour the grounding during the layout, as the VOUT PIN and PGND PIN is seperated by the SW PIN, so if the input PGND and output PGND is connected via vias, use as much as vias possible and use bigger size vias.

    3) add snubber between SW to PGND.

    by the way: Jasper will back to office 8 days later!

    BR, Helen

  • 0 in reply to Jasper Li
    Intellectual 480 points

    Jasper,

    Why is the output capacitor the most important component? For me, I followed the recommended design in choosing the passive components connected to the TPS61090, however I put the components far apart from the TPS61090 because I didn't think it would be an issue. I also am trying to get out 5V at a little under 1A with a battery input of 3.7V. Should all the passive components be close to the chip in order for the chip to not be damaged? Also, I didnt not attach any components to the LBO and LBI leaving them open; is this alright?

  • 0 in reply to Isaac Fenta
    TI__Mastermind 46905 points
    layout is also critical for a boost converter, please follow the application note to place components "www.ti.com/.../slva773.pdf" . it is OK to let LBO and LBI floating.
    i would suggest TPS61230 for your application.