• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 5 answers
  • Subscribers 62 subscribers
  • Views 140 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

UCC3895: Bipolar Output needed

Gerhard Kreuzer
Expert 2600 points
Part Number: UCC3895
Other Parts Discussed in Thread: UCC28951, UCC28950, UCD3028, TSC2000

Dear Team,

I need a voltage output which should vary from +Out to -Out. I use a phase shifted transformer driver, as you use in the App Note SLUA287 – March 2003.

Is there any chance to add something to the secondary side to get an output range which varys from +Out to -Out?

This also includes a output voltage near zero.

Thanks for helping.

With best regards

Gerhard Kreuzer

  • 0
    TI__Guru 72705 points

    Hello,

    Is this is a single output that is reference above and below ground?

    Regards,

  • 0 in reply to Mike O'
    Expert 2600 points

    Hi Mike,

    yes, a single output changing its polarity over time ..

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 72705 points

    Hello,

    I have not seen a design where the output goes from positive to negative changing polarities.  

    Could you tell me what the input and output power requirements are for your design?

    Could you tell me why you need the output voltage to change polarities?

    Regards,

  • 0 in reply to Mike O'
    Expert 2600 points

    Hi,

    I need a periodical output voltage, so its not a power supply, but works quite the same way. As I need high voltage output, a transformer is needed, but transformers for a frequency range between 100Hz and 30kHz and 2kW .. not that easy. So I put the transformer before demodulation, now it sees the switching frequency of 200kHz only, which makes things easier.

    Think about this configuration as a power supply and you periodically change the output voltage setting.

    I do this with a standard secondary stage, works fine, I get a nice sine wave with 100V pp.

    Than I found the 'current doubler' secondary stage. This will help to get copper losses down und boost the current at low frequencies (my load is an inductor).

    I used the ducument SLUA287 by TI, page 6 'An Easier Method'.

    Unfortunately I didnt get any output. The main difference is the fact, that the current through L1 & L2 changes direction, see scope. In the document the current only changes its value but not its direction. The cursor horizontal lines marking 0A. The vertical cursor lines marking the state when energy is transferred.

    Whats the difference: I didn't use single MOSFets as described, I use two MOSFets, with connected sources, forming a bilateral switch. So I draw a 'switch' symbol not a MOSFet symbol on my secondary side. The switches working as expected, both tested outside the circuit.

    Why bilateral switches? Changing the role of S1 and S2 with the change of the sign of the input singnal (digitally generated waves and than digitally converted to a PWM puls train) changes the polarity of the output voltage.

    The first test here is made with a DC signal, so only a positive outout voltage is expected to keep things simple for the first moment.

    Is there any explaination why the currents through L1 & L2 behave totally different as described (only the shape is ok)?

    Thanks for helping.

    Note: The MOSFet driver for S2 inverts the signal S2 internally, so you see the inverted signal .. sorry about that. S2 low ==> S2 is closed., S1 low ==> S1 is open.

    Current Doubler Topology - 22092022.pdf

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 72705 points

    Hello,

    It looks like you might be driving the synchronous rectifiers on the output incorrectly.   I would double check the circuitry vs the application note that you mentioned.

    When A_H and B_L are on S2 should be on.

    >This will cause the current in L1 to increase while the transformer is delivering energy.

    When A_L and B_H are on S1 should be on.  

    >This will cause the current in L2 to increase while the transformer is delivering energy.

     

    Regards,

  • 0 in reply to Mike O'
    Expert 2600 points

    .. exactly.

    I first thought there is an issue with switching the right elements, but I checked it carfully, cant find an issue. If I switch both S1 & S2 on I will get an very high current through the secondary (and primary) winding and no current though any of the inductors L1 and L2 or if S1 and S2 were swapped I guess I get a negative output, which is ok for me, see general system description.

    Don't have an idea how to switch S1 & S2 the wrong way to get correct shape of inductor currents but wrong value/direction ...

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 72705 points

    Hello,

    The more I look at your waveforms your current going negative is most likely what is causing the output not to increase.  Try replacing the SR with diodes to see if the issue goes away?

    In FSFB design using SRs it is recommended to turn the FETs off before critical conduction, so negative current does not build up in the inductors.  This will protect the FETs from over voltage at FET turnoff. 

    We have a PSFB controllers that are designed to be used with SRs.  They also have a programable DCM function that turns off the SRs before critical conduction.  They are the UCC28950 and UCC28951.  The following link will bring you to an application note that shows how to use them in a phase shifted full bridge. https://www.ti.com/lit/pdf/slua560

    Regards,

  • 0 in reply to Mike O'
    Expert 2600 points

    Hi,

    I will check the diode test.

    Sorry, I can't use standard controllers, cause I didn't need a power supply with feedback loop and all this stuff.

    I have a FPGA implementation so I can generate pretty all possible signal patterns to switch the FETS. It was very easy for me to try the topology described at SLUA287, but ...

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 72705 points

    Hello,

    Have you ever considered trying a digital  power supply controller?

    This may give you more flexibility of designing in your unique application, due to programmability. 

    the digital UCD3028 controller.

    The following will bring you to the UCD3028 programable digital power supply controller's data sheet.  You may find this device of interest for your application. 

    https://www.ti.com/product/UCD3028

    The other option would be the TSC2000 digital power supply controller as well.

    https://www.ti.com/lit/gpn/tsc2000

    If you decide to investigate using the UCD3028 or TSC2000 and have questions, please repost in the e2e with these part numbers in the titles, so the appropriate applications engineer can help you with your design or any questions that you have.

    Regards,