• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 64 subscribers
  • Views 1694 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.

Vout problem on the TPS62110

mickael
Prodigy 150 points
Other Parts Discussed in Thread: TPS62110, TPS62113

hi ?

 

  My customer got a problem with eth TPS62110 . ( production STOP).  The output voltage is 0V when the input voltage is 6V.   He observe in this case that the Iout is 200 mA .

He doesn t know how it is possible to have it ( the input votage is < 17V ).

When he switch off the Vin et switch on again , the Vout is correct .

The Vin max in the system is 16 V .

The Vsync is driven by a MCU . In the application , this pin is used to force teh PWM mode ( sync =1) with a frequency of1 hz for a time of 10ms .

the main time the sync pin is 0.

could you help us for this case ?

 

thanks

mickael

 

 

  • 0
    TI__Guru**** 150930 points

    Can you post a schematic?

    Can you confirm that the device is not a TPS62113, based on the package markings?  This version has a special LBI function that can also disable the device.

    What type of load are they using?  If it is an electronic load, then I would try testing with simply resistors as electronic loads can cause issues with how they sink current to a given voltage.

    Lastly, can you post a picture of Vin, EN, SYNC, and SW?

  • 0 in reply to Chris Glaser
    Prodigy 150 points

     please find the schematics

  • 0 in reply to mickael
    TI__Guru**** 150930 points

    Thank you for posting the schematics.  Does the problem appear when the SYNC signal is not present?  Is this issue repeatable?  If so, what are the conditions required to repeat it?  What is the amplitude of the SYNC signal?  Is the signal present before Vin is?

  • 0 in reply to Chris Glaser
    Prodigy 150 points

    Can you please ask the factory or designer how the current limitation of the IC works ?

    Is it a cycle by cycle operation that reacts very quickly (1us at 1MHz) or is it working on an average on several 10s or 100S of cycles ?

    Can you also ask how the internal N and P channel power transistors are switched ?

    Is there a fixed dead time between switching both transistor or is this dead time function of Fsw ?

    In case of latchup, this transistors gate control is the key of the problem.

     

    If the inductance goes to saturation, then changing the coil is the solution of the problem.

     

  • 0 in reply to Chris Glaser
    Prodigy 150 points

    1. Question de Texas: Is this issue repeatable

    no , it appears after some days , week or month

    2 Question de Texas: Can you confirm that the device is not a TPS62113, based on the package markings? 

    please find the marking and lot

    ·        TPS62110 TI 031 A0XD

    ·        TPS62110 TI 04K CKVK

    3 - Question de Texas: What type of load are they using?  If it is an electronic load, then I would try testing with simply resistors as electronic loads can cause issues with how they sink current to a given voltage.

    load ( CPU + RF ) : between 50uA in mode PFM and 200mA in mode PWM.

    4 Question de Texas: Does the problem appear when the SYNC signal is not present? 

    no process.   They drive the Sync signal to change the mode en PWM whe a extra  load ( RF) is connected in order to reduce the noise

    generated by the Alim/power .

    5Question de Texas: If so, what are the conditions required to repeat it?

    no procedure to repeat it

    6- Question de Texas: What is the amplitude of the SYNC signal? 

    2V8_PWM=SYNC=1 → 2,50V

    2V8_PWM=SYNC=0 → 0V Pulled down by R86=10KΩ

     

    7 Question de Texas: Is the signal present before Vin is?

    no , the 2V8_PWM  is a signal from the RF chip that is alimented by the 2V8 of the TPS62110

     

     

  • 0 in reply to mickael
    TI__Guru**** 150930 points

    Thanks for finding out some answers.  This gives us a lot of information to troubleshoot with.

    So, the problem seems to be random and only happens after some long amount of time.  Is this correct?  Is this seen on one unit or several units?  Have they ruled out a manufacturing defect in either the PCB or assembly of the components?

    The IC's current limit is very simple and explained in the datasheet.  If the current in the P-channel FET is exceeded, then it turns off the FET.  No counter or anything.  Very fast.

    A waveform of Vin, Vout, SW, and the SYNC signal would help in ruling out some other possibilities.

  • 0 in reply to Chris Glaser
    Prodigy 150 points

    for the waveform , I m going to ask the customers .

    PLease find some comments of the customers =

    1 _ please find the batch concerning : TPS62110 TI 031 A0XD  TPS62110 TI 04K CKVK

    2 - do you have already seen a similar problem ?  The TPS62110 can have the kind of the function ? if yes in which case ?

    3 ) Actually they have a production STOP on the products . For some products , they can use only the PWM mode .

    in this case , the sync pin is High level.  Could you garantee that they will not have the problem ?

    4 )They calculate the value of the component with te TI simulator . This simulator uses only the PWM mode to evaluate the value .

    But they change the mode between PFM and PWM every secondes. could you confirm that the value are corect

    5) They saw in the forum a post  (Posted by JatanNaik on 5 Dec 2008 11:26 AM) concerning the TPS62110 that the simulator doesn t simulate all part in the component :

    http://e2e.ti.com/support/development_tools/analog_elab_and_tools/f/235/t/27149.aspx)

    Maybe we can do a conf call with my customer tomorrow , in order to resolve the probleme ASAP or discuss by email ...

    mickael

    « But this part doesn't really work this way it has several other confidential things that happen inside that basically make this warning not applicable for this family of devices. But the SwitcherPro tool doesn't understand that so we've basically removed the open loop gain bound in order to remove the warning »

    Could you please check with your tools that the schematics is correct ?

  • 0 in reply to mickael
    Prodigy 150 points

    to complete my answers , the EN pin is alway driven by  the Vin ( between 15V and 4 v ) .

     

  • 0 in reply to mickael
    TI__Guru**** 150930 points

    Based on your description of the problem, which I understand to be the device does not power up once in a very long while and randomly but works fine otherwise and works fine if Vin is cycled--because this is the symptom that is described, I do not think this is related to current limit, stability, proper external components, etc.  I have never heard of this type of issue before.  My guess is that this is related to a manufacturing issue of either the PCB, IC, or assembly of components onto the PCB (a bad solder joint on the EN pin for example).  A remote possibility is that their use of the SYNC pin disturbs the IC's internal circuitry somehow, but since there are no steps to reproduce the problem, this will be impossible to discover.

    My main question at this point is how many boards has this issue been seen on and out of how many?  Also, did this just come up (after making their product for years) or is this the first build with theis device?

    To directly answer your questions: The Switcher Pro design in PWM mode is ok in PFM mode also.  I do not believe this is the issue as their symptom does not point to this as an issue.

    Their schematic looks ok but I would need the output capacitor part numbers (it looks like there are 2 possible part numbers) to be sure.  Again, this is likely not the cause, if the IC works fine except for once in a while.

    I cannot guarantee that they will not see this behavior if they keep SYNC high.  But I am curious if they have ever seen this behavior without toggling the SYNC pin.