• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 4 answers
  • Subscribers 62 subscribers
  • Views 979 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.

UCC2801-EP: short circuit & restart

Hyeonseung Ju
Prodigy 40 points
Part Number: UCC2801-EP
Other Parts Discussed in Thread: UC2845, UCC2891, UCC2801

hi, I have one question 

I use UC2845 for 2switch forward converter

UC2845 doesn't have protection from short circuit,  so I test UCC2891

Test 

Insert UCC2891 instead of UC2845

Seperate self bias Vcc and connect 12V from other power supply

Operate the product

short to output

and then, Output current flow higher than 70A ( I see output current waveform to oscilloscope, the output current flows sequentially, not hiccup)

it doesn't do hiccup or restart

how i make the UCC2891 to do hiccup from short circuit?

  • 0
    TI__Mastermind 33115 points

    Hi

    your problem description is not making sense. For example UC2845 is a current mode controller with cycle by cycle peak current limiting so I’m curious why you say there is no short circuit current protection? Also, for a 2-sw forward I assume you split the PWM gate drive to drive the HS and LS simultaneously. If you replace UC2845 with UCC2891, what do you do with the 1-D output? UCC2891 is a D, 1-D active clamp PWM. If you could provide a schematic of the two solutions you are comparing as well as waveforms of the CS signals for each, I can be better equipped to help. 

    regards,

    Steve M

  • 0 in reply to Steven Mappus1
    Prodigy 40 points

    sorry, not UCC2891 but UCC2801

    UC2845 and UCC2801 are different block diagram for CS circuit

    I expect that UCC2801 protect from short circuit, CS voltage is over 1.5V, IC restart but in reality not hiccup operating 

  • 0 in reply to Hyeonseung Ju
    TI__Mastermind 33115 points

    Hyeonseung,

    Neither of these current mode PWM controllers have the internal restart timer needed to operate in OCP hiccup mode. When the UCC2801 CS > 1V the gate drive is reduced as the load current continues to increase (soft overload) and the output voltage will drop and this is a fundamental characteristic of all PWM controllers using peak current mode control. The deeper you move into overcurrent by continuing to increase the load, the more the output voltage drops and both, the UC2845 and UCC2801 behave similarly this way.

    However, when the UCC2801 CS >1.5V (short on output or transformer saturation) the controller will enter OCP and the gate drive output will immediately go to zero, FB will drop below 2V and a new soft start cycle will begin.

    The UC2845 does not have such features as internal soft start, leading edge blanking or second level OCP. 

    You mentioned you separated the self-biasing VCC and were biasing with another 12V supply. If you reconnect the the self biased VCC winding, then the self-bias winding would drop out during the OCP event and try and restart again from the line voltage but if the short were still on the output, the VCC would again drop out and repeat until the short is removed - this behavior is not true hiccup mode but can appear like hiccup mode. You would observe the VCC capacitor continually charging and discharging according to the UVLO thresholds. You are not seeing this because your 12V on VCC remains through the entire OCP event. Try reconnecting your self-bias VCC and see if you approve of the OCP behavior?

    I hope this clarifies the operation.

    Regards,

    Steve M

  • 0 in reply to Steven Mappus1
    Prodigy 40 points

    thank you for your help

    But,  the point is I can't make self biasing circuit , ex) Lot of current Flow to Transformer and Inductor, so wire getting bigger , converter will be smaller size ...

    there is many converter in the system so i make auxiliary flyback converter that gives 12V power to UC2845

     Therefore I find upgrade version form UC2845 that having OCP (or Short Current Protection) internally

    Doing expriment,  Voltage derating is fine,  but output current is very high (short current situation) , the converter is not hiccup operation

    how can i fix that problems?

  • 0 in reply to Hyeonseung Ju
    TI__Mastermind 33115 points

    Hyeonseung,

    The PWM controllers you are comparing do not include hiccup mode OCP operation. Some other TI controllers do have this feature integrated - LM5021 and UCC28950 are two examples. For peak current mode control, if the power dissipation is too high when you are operating in OCP protection mode (current foldback), you either need to over design the current sensing circuit or you will need to design discrete circuits to take whatever corrective action you desire. 

    Regards,

    Steve M

  • 0 in reply to Steven Mappus1
    Prodigy 40 points

    Hi

    You mean, most of current controller ICs don't have OCP circuit internally, right? 

    If i want to OCP circuit, is that any example circuit for OCP? 

    I think, using CS pin is most way to OCP, not using current sensors 

  • 0 in reply to Hyeonseung Ju
    TI__Mastermind 33115 points

    Hyeonseung,

    Since we have closed the discussion of UCC2901 not having hiccup OCP and we discussed how the OCP of UCC2801 and UC2845 does behave according to peak current mode control, Let's close this thread topic. I would encourage you to open a new E2E thread and please be thorough to include your schematic (relevant portions), measured waveforms and circuit description as best as possible. Thanks for posting to E2E.

    Regards,

    Steve M