• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 959 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.

TPS 61222 Boost-Converter wont work properly

seb
Genius 4170 points
Other Parts Discussed in Thread: TPS61222

Hello everyone,

 

i am just trying to get my TPS 61222 boost converter to work. I try postnig 2 layouts, one of them is working fine and the other one is not.

 

 

This one is my new one, but i is not working. with no load conditions, i get some kind of 5 V at the output, but as soon as i put a 10 mA load on the output it breaks down to aroung 3 V, which is about or a little higher of the input voltag.

 

I have 2 boards in front of me, and the problem is the older board is just doing fine, but the layout differences are really small, and about the ground issues i tried a lot of stuff on the new board to minimize the differences even further, it just wont work with loads.

 

 

I would be really happy if someone finds some failure that i just couldnt see myself, if not i guess i have to rebuild a board with the exact first layout or the one which is recommended by texas instruments.

 

Are those Converters really that hard to layout?? cannot belive it, as i am only using 10 mA and not the full range, so i hoped it would be rather uncritical for me.

 

Thanks for youre time and help.

 

BEst wishes.

Seb

  • 0
    TI__Guru**** 150650 points

    Which of the pictures you posted works and which does not?

    Layout of a high operating frequency power supply is always important, regardless of load.  It must be done right.  Compying the EVM layout is a sure way to avoid layout issues and we recommend doing so here.

  • 0 in reply to Chris Glaser
    Genius 4170 points

    Hello,

     

    thanks for the fast response.

     

    The first picture shows the working layout.

     

    The second one is not working, but i am right now trying to solve the issues, i soldered 2 Coils in series right now, this seems to help a lot. so i am right now working with 2x 4,7 µH.

    But is is not yet completly correct. I have to drive a max. 20 mA load with the TPS61222, this seems to work now with twize the coil inductance, but still if i am trying my layout with a resistance the voltage drops down to around 4 to 4,2 V. I will try to implement even more inductance tomorrow.

     

    Thanks for the time.

    Seb

  • 0 in reply to seb
    TI__Guru**** 150650 points

    I am not sure where your output cap is in the second picture.  The part next to the Vout pin goes to that via and doesn't seem to connect to ground.  The output cap should be right next to the IC as it was in the first picture.

  • 0 in reply to Chris Glaser
    Genius 4170 points

    Indeed, that one is the output cap. I tried different values for this as well, i am stuck with 10µF 0603 which is good with 6,3 V so it should not be a problem.

     

    It is connected to my analog Ground, that is why the path doesnt go directly back to the TPS, however due to my problems i connected every GND together also so for example i soldered the GND from the input Cap directly to the GND of the output cap. I did meassurments with the both GNDs connected and seperated, that doesnt seem to make any differences in my output voltage drop behaviour..

    Lets say i did already try a lot of the common stuff like GND issues too less input or output caps etc.

     

    But thanks for the ideas, always good to have mroe opinions.

    Seb

  • 0 in reply to seb
    TI__Guru**** 150650 points

    The output cap ground is a power ground and MUST tie directly to the GND pin of the IC.  The input cap is also a power ground.

  • 0 in reply to Chris Glaser
    Genius 4170 points

    Hello again.

    It absolutly works now.

     

    It was the missing short short short PGND trace from the output cap to the GND of the TPS.

     

    It has really to be very short, because yesterday i did already solder a approx. 2cm wirelength from the input cap to the output cap and this is already too long for the TPS.

    Today i directly connected output cap to gnd and it is working as it should.

    Now i am also switching back to the recommended 4,7 µH coil.

     

    Everything is worknig, i will build a new design now, coming back with more tourbleshooting in a few weeks i fear :)

     

    thanks for the help.

    Seb