• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 215 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.

Original question:

UCC28056: Ton Calculation

UCC28056: UCC28056 maximum Ton time calculation

Ola Baath
Prodigy 40 points
Part Number: UCC28056

Hi,

I have a follow up question to the previous Ton time calculation question asked a few days ago. Specifically the maximum on time for a certain gff.

For example at 230 Vrms.  Gff is then the sixth selector, which equals 0.158. What I then do not understand is how the maximum on-time can be 5.26 us? To me the only way this can be is if dondch is < 1.

If I understood this correctly I then can't understand the measurement I have done below;

In the measurement snapshot a few cycles can be seen at 230Vrms at the bottom of the input voltage (~180 deg). The on time is close to the maximum specified of the datasheet (~5.28 us). But I do not understand how this is made in the calculation of the on-time.

Here ton = 5us, toff = 900 ns and Tdcm = 0. Which gives dondch = 1. How can the on-time then be more than gff(6)*Tonmax0 = 13.2*0.158 = 2 us?

And for this specific case if I also plug in the comp pin voltage in the equation the calculated on-time should be according to the equation below;

Vco = 3.36 V, Vcomax = 5.7 V, gff = 0.158, Tonmax0 = 13.2 us, dondch = 1 -> Ton = 1.23 us

  • 0
    TI__Mastermind 26770 points

    Hello Ola,

    Thank you for your interest in the UCC28056 PFC controller.

    You pose a good question.  I can't argue with your math; it checks out as presented.
    I think the reason for the long apparent on time is partly physics and partly misperception of equation (8) of the datasheet.

    First the misperception (which affected me, too, for a long time until I realized what is going on):
    Ton(theta) in equation 8 is proportional to Tonmax0 for the specific case of full load, minimum line, as stated in the first sentence on that page (page 17).

    Ton(theta) becomes proportional to TonmaxX and GffX for other higher input lines.  The text does not make this clear. 
    In the case of 230Vac input, Ton(theta) will be proportional to Gff6*Tonmax6.
    Thus, using your numbers: Ton = 3.36/5.7*0.158*5.09/1 = 0.474us. 

    Of course that hardly seems to fit the waveform, so the second part is the physics of superjunction MOSFETs. Assuming that your application is using one, they have tremendously non-linear Coss characteristics.  The drain-to-source capacitance is orders of magnitude higher at low to zero Vds, and reduces considerably at higher Vds voltages.  That means that when such a MOSFET is turned off, the peak current (quite low near the zero-crossing) must charge up the high Coss first before the drain voltage starts to rise appreciably.  It also means that near the zero crossing, the resonant energy ringing Vds down from Vout really drives Vds to GND for a long time.

    I suggest that you probe the actual DRV output signal and compare it to the Vds waveform above.  I suspect that it will appear in the middle of the low Vds, going high where the little up-tick is in the center and going low before the drain starts to rise.  I believe that the COMP voltage had to decrease to 3.36V in order to reduce Ton enough to account for the additional extra "on-time" obtained by charging up the high Coss after turn-off.   

    Please check this suggestion and let me know if it accounts for the apparent discrepancies in the original math.

    Regards,
    Ulrich

  • 0
    TI__Genius 9140 points

    Hello Ola,

     thanks for your interest in UCC28056.

    Does the waveform captured on UCC28056EVM, or do you do any change on the schematic compare to the EVM?

    I suggest you test the Vgs instead of testing VDS.

    at 230VAC input, Vco=3.36V, the controller should works at DCM with Tdcm not equal to 0. you can refer to the Figure 18 in the datasheet.

  • 0
    Prodigy 40 points

    Hello Ulrich and David,

    Thank you for your detailed replies. As you said the trick to see what is going on is as you said to measure the DRV signal and not only look at the drain voltage.

    By looking at the similar operating point It is visible that the on time should be as the formula describes. But not with Tonmax(X), x=6, are you certain this is the case? The calculated on-time at 230 Vrms (gff=6) checks out with Tonmax0. See below;

    to= 2.2 us, t1 = 0.7 us, t2 = 2us

    -> dondch = (2.2+0.7)/(2.2+0.7+2) = 0.59

    Vcomp = 3.36, gff =0.158, tonmax0 = 13.2 us

    -> Ton = 13.2/0.59*0.158*3.36/5.7 = 2.1 us

    It also matches at high line of 230Vrms

    t0 = 1.5 us, t1 = 8 us, t2 = 2.7 us

    -> Ton = 1.59 us (gff = 0.158 and Tonmax0 = 13.2 us)

    I would suppose the Tonmax1-7 is just there as a safety mechanism then and has nothing to do with the calculation of the on time?

    Kind regards,

    Ola

  • +1 in reply to Ola Baath
    TI__Mastermind 26770 points

    Hello Ola,

    Thank you for probing Vg and for your additional calculations. 
    It appears that Ton is indeed dependent only on Tonmax0 and not on Tonmax(X) regardless of Gff(X) levels.

    I was not certain of it, but came to that conclusion while trying to fit in a smaller on-time without knowing what it actually looked like, and trying to justify why the additional Tonmax(X) parameters are specified.

    If they are not useful in the Ton(theta) calculations, then I suppose they might simply be a safety mechanism as you suggest.
    Certainly, the Tonmax(0) term alone makes the calculations match the actual DRV signal timing.  

    Given this result, I think this thread can be closed.

    Regards,

    Ulrich