• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 63 subscribers
  • Views 270 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.

UCC28056EVM-296: Vout rises way above VoutReg under no load condition

Luca Vettorello
Prodigy 50 points
Part Number: UCC28056EVM-296
Other Parts Discussed in Thread: UCC25630-1EVM-291, UCC28056

Tool/software:

Hi, I'm currently doing some preliminary tests on two different UCC28056BEVM-296 evaluation modules that I have. 

For the test setup I'm using an AC source of 230V/50Hz and a DC Source at 12V to power the board. I'm also monitoring the output voltage using a multimeter. No load is applied to the output. 

I noticed that under this condition the output voltage rises well above the nominal value of 390V reaching values of 420V and higher. I didn't check if the OVP1 trips at 440V since I'm preventively disabling the AC source when the voltage rises above 420V to avoid any possible damage to the board.

I also noticed that by applying a very small load of 0.5W the issue is resolved. The output voltage stops increasing above 390V and is somewhat stable in the 385V-391V range.

My questions are the following:
Is this the intended behavior of the controller?
If so, do I have to add a minimum load to the output of the PFC to avoid this condition?
what is the value of this minimum load?
Alternatively is the OVP1 enough to guarantee safe operation?

If this is not the intended behavior of the controller do you have any suggestions in order to solve this issue?

Moreover do you have any insight on the mechanism behind this behavior?
is it related to the stability of the system?

Thanks in advance

  • 0
    TI__Mastermind 40580 points

    Hello Luca, 

    I would not expect the UCC28056B EVM output voltage to drift significantly higher than nominal under a no load condition. 
    This is not the intended behavior of the controller, therefore a minimum load is not normally required.  
    I think your situation is abnormal and cause of the rising voltage must be found. 

    First, since you are able to manually intercede and shut off input power before Vout rises > 420V indicates that Vout is gradually drifting higher over time, not stepping to 420V in microseconds or milliseconds.  Can you verify this conjecture, please?  If so how long does it typically take to rise from ~390V to ~420V?
    Does it do the same thing with input voltage at 115Vac?  

    Aside from the multimeter on Vout, do you have any other instrumentation attached to the EVM test points?  If so, what are they and where? 

    A previous E2E thread ( https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/720810/ucc28056evm-296-ac-current-waveform-explanation-for-burst-mode ) contains a PDF file wherein the first slide shows the AC input current of the UCC28056BEVM-296 connected to an LLC EVM UCC25630-1EVM-291 with the LLC output a 0A (no load).  Checking the LLC EVM User guide, is indicates that no load input power is 39.7mW (Table 2 in https://www.ti.com/lit/pdf/sluubq8 ) at 390V.
    This represents less than 1/10th of the load on the PFC EVM that you have. 
    I suggest to compare your EVM input burst-current pattern with that of the PFC-LLC combination no-load current in slide one. 
    Also note the PFC output voltage ripple in slide 1 (red trace, AC coupled at 5V/div) and compare to your ripple at no-load and at 0.5W load. 

    I do not believe that rising Vout is caused by a loop stability issue.  I think it is caused by some kind of leakage current issue. 
    If the average Vout is rising slowly over time (compared to typical PFC loop response time), then either the voltage loop believes that it is in regulation and there is an increasing leakage current to GND at the VO_SNS input, OR, there is an increasing leakage current into the COMP pin that defeats its attempts to extend the deadtime between bursts to keep Vout regulated.  

    Those are my thoughts based on the information available above. 
    Please investigate as recommended. 

    Regards,
    Ulrcih 

  • 0 in reply to Ulrich Goerke
    Prodigy 50 points

    Hello Ulrich, thanks for your reply.

    After a bit more testing we found out that the problem was related to the 12V power supply that we were using to power the UCC28056.
    The supply was somehow injecting noise into the circuit through the earth connection. This caused the value of Vout to slowly increase and reach 420V over several seconds of operation. By changing power supply we were able to remove the problem and now the EVM works perfectly!

    Thanks for your help