• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 577 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.

Startup problem with TPS54525

Lars Mikaelsson
Prodigy 50 points
Other Parts Discussed in Thread: TPS54525

I have experienced some startup-problem with the TPS54525.
My system consists of a webench-designed converter using TPS54525 with 6-10V input and 5V 4A output. The converter powers a ETX type computerboard consuming about 2A@ 5V. The 5 V is fed to the ETX-board via a PIC controlled Mosfet switch(IRF9130). At the first tests the switch was replaced with a jumper and this did not work. The converter could not handle the inrush current at power on and the converter died. Some startup attempts(less then one out of ten)was successful. I then increased the softstart capacitor and then it seemed to work OK.
Then I removed the jumper from the mosfet switch, and the problem was back. The converter powers up at power on but when I activate the switch to power the ETX-board, voltage drops and the converter gives up. I tried to solder a ludicrous amout of capacitors between the converter and the switch in order to handle the inrush current but the converter dies and must be restarted.
I have also tried to add a 100pF extra to the 22pF Cff but that didn't help.
According to datasheet all protective functions seems to be non-latching, but my problem is actually latching.
See attached schematics.
Any ideas how to solve this?

Best Regards

Lars Mikaelsson

www.malags.com
 

TPS54525 5V ny.pdf
  • 0
    TI__Guru**** 191445 points

    The schematic looks ok, but where does your EN pin go?  It must be actively pulled up (I would tie it to VIN).  What input voltage are you using?  Try raising it towards the high side of your range.  If those do not fix your problem I will have to see waveforms. 

  • 0 in reply to JohnTucker
    TI__Guru**** 191445 points

    did you ever resolve your issue?  Let me know if you need further assistance.

  • 0 in reply to JohnTucker
    Prodigy 50 points

    Well, yes, and no.

    I used a larger gate resistor on the MOSFET switch and that gave a softer turn-on. So the problem isn't solved but I found a shortcut around it and I would appreciate some help to solve it. What is normal behavior if the inrush current is to high(in this case I think it is all the capacitors on the ETX-board)?

    If the circuit continued stumbling and struggling for a couple of milliseconds it vould charge all capacitors and then the current vould drop to a more reasonable level. But instead it just dies. I must disconnect Vin and then restart it.

    I have also tested it with a resistive load of 4 - 5 Amp and that's ok but if I increase it it dies again, it doesn't help to remove the resistors, I must disconnect Vin before it can be restarted.

    If you want to see some scope-shots, let me know were to put the probe.

     

    /Lars

    My enable pin IS connected to Vin(without resistor).

     

  • 0
    TI__Guru**** 191445 points

    How much "extra capacitance" is there on the output?  That may affect stability, but it is not (probably) the cause of your issue.  I need to know for the current assessment though.

    I could not drive the TPS54525 into overcurrent using my bench load.  Under short circuit conditions, it does in fact latch off though.  Even if the current limit itself is not "latching" the UVP protection will latch the device off when the output voltage falls.  Can you verify the output voltage value when your supply latches off?  And how much output capacitance?  Uusually the way to decrease peak current draw during start up is to increase the slow start time.

  • 0 in reply to JohnTucker
    TI__Guru**** 191445 points

    I verified with the IC designer that in fact the over current protection is non latching and that the UVP protection is latching.