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.

SLUC146 and UCC25600

Other Parts Discussed in Thread: UCC25600

Hi,

 

I am working on a 2KW resonant powersupply based on the UCC25600 controller. When using the calculation spreadsheet SLUC146 for this design something weird happens.

When entering the design, the ratio Lm/Lr (  in the sheet called "m" )  is entered as a design parameter, just like expected. However when I choose Lm/Lr as 5, and look at the component values, the resulting Lm/Lr has a very different value.

I calculated the powertrain components also by using an other calculating methode, and ran some simulations in spice to check the results, this is close to what I expect.

Looking at the SLUC146 sheet, the basic values like transformer ratio, Qmin and Lr are more or less identical, however the magnetising inductance Lm resulting from the spreadsheet is roughly a factor 2 too low.

 

regards

Geert de Vries

Duran Audio BV

What is going wrong??2844.2k9.xls

  • Hi Geert de Vries,

    SLUC146 uses a different method to determine final Lm with the intention to consider the effect from secondary side transformer leakage.

    The attached is the design tool with modification which uses the formula normally seen to determine Lm by Lm = m x Lr for your reference.

    8371.Design Tool for UCC25600_Rev1.2_Ln.xls

    Regards,

    Hong Huang

     

  • Hi Hong,

     

    Thank you very much for the quick response.

    Problem solved.

     

    Regards

    Geert

  • hi, Hong  Huang,

    Is it possible to make a wide   input voltage range  LLC converter using 25600, by using 2 interleaved LLCs with a 90 deg phase shift ? ( Here because of interleaving, the

    gain required per converter shall be less, so less circulating current because of large input variation as compared to single stage llc am I right ? )

    regards ,

    Feesah

     

  • It may not be feasible to do phase-interleaved LLC resonant power converter with UCC25600 since the device does not have a frequency synchronization pin.

    Paralleling two LLC converters with the interleaved technique may not help to reduce the gain variation requirement due to the input voltage variation. The interleaved technique may help reduce the gain variation due to the load less variation.

  • Hi Hong Huang 

    Is it possible that you can comment on this, or find someone who can?

     

    Finn Br��sen posted UCC25600 in AC/DC and Isolated DC/DC Power Forum.

    Working with the UCC25600 in 800W DC/DC isolated converter.

    One thing that annoys me is the soft start (SS) function. During start-up, I have the maximum output voltage only limited by the transformation ratio, and the input voltage, until the SS period is over. This is in my case 60msec.

    Does anybody know how to deal with this except disabling the SS function.

    To my understanding all functions seems to be disabled until SS time is over.

     

    Br

     

    Finn Broesen

  • The switching frequency is controlled during the soft start by the soft start capacitance and the RT pin current. Then the output voltage is not only controlled by the transformer turns ratio but also the RT and theSS current.

    What target your design is to achieve during the soft start? 

  • Hi Hong

    My target is to start up slowly using the SS pin to switch on and off. 

    The output capacitor starts charging, but when the capacity is only 22uF, it goes quite fast. The loop is then trying to command the output voltage down but does not have any influence before the SS time is over. I can use a smaller SS time but I still have the overshoot. 

    I've tried to use OC and RT pin to switch on and off, but it gives the same result.

    The voltage output is 20V - 160V. I always start with asking for 20V without load. Transformer ratio is 1.28 and resonance is just below 100kHz. PFC voltage input.

    All in all, it works fine apart from this overshoot. 

    I can add an active load clamping the over voltage but first i'm looking for a small easy way to do it.

    Br

    Finn

  • What is the capacitor value used for soft start?

    You may add an R-C between RT pin and GND. This way can make the switching frequency close to 350kHz which will help to reduce the output voltage during start. Calculate the total current from RT and SS to make sure below 7mA at all time.

  • I've consider that solution. But i'm afraid of the impact to the good transient behavior this power supply has. When I switch from 20  - 160 V,  I need to do it as fast as possible. 

    The question is, does RT pin has any influence before SS reaches 4V. Vss is shorted. (according to block diagram)

    Standard SS capacitor is 100nF but have tried down to 10nF, corresponding to 5 - 60 msec.

    I'll try to play around with it, otherwise I need to cope with my active load.

    Br

    Finn

  • RT-pin current affects the switching frequency at all time. Use R-C in series to connect to RT and GND to allow higher switching frequency which helps to make output voltage less overshoot. When C is charged, the addtional R-C will have not effect in normal operation.

    To make less output over shoot, I think a bigger capacitor value on SS should be used.

  • Hi Hong

    Thanks, I'll try it. 

    You will here from me, so lets keep this thread open!

    Br

    Finn

  • Hi Hong

    Is there a site or an address where to I can upload some pic's and a doc?

    Br

    Finn

  • You can upload your files to this site, if you want.

  • Hi Hong

    Please have a look at this document and let me here your comments.

    Br

    Finn

    TI.docx
  • Could you please label each scope channel to allow analysis?

  • Sure!

    CH1. (yellow) is output side. (Load 130 ohm) output cap is 22uF

    CH2. Gate drive low side.

    Pic's 2 and 3 is zoom of no. 1.

    Br

    Finn

  • Look at the datasheet page 15, the switching starts when Vss reaches 1.2V. But in your test, it does not seem so. Could you please test with SS shown to find out if your design follows Figure 19 on page 15? The switching frequency at Vss = 1.2V should  be about 350kHz which helps to make output start with a small voltage and then the output voltage increases as Vss increases. I cannot see your test results with such behavior.

  • Hi Hong 

    I did some more measurements and included SS (Blue Ch3). 

    It looks to me as it follows fig.19 page 15 but unfortunately it does not enter the Burst-mode before the SS time is over.

    I would very much like to here if you have any suggestions to this!

    Br

    Finn

    UCC25600ssmeas.zip
  • The burst operation will be available only after soft start complete. During soft start there is no burst operation from UCC25600.

  • Hi Hong

    I found a solution this morning that brings the overshoot well below the uncomfortable level.

    By changing the DT time (only) during SS time and changing the SS time to a minimum (3.5 msec.) the problem is solved.

    Br

    Finn

  • Hi, Hong,

    1.  I use ucc25600 with typical application. I found that when DC-DC converter is soft-starting , OCP trigger signal will acvtive in OC Pin of UCC25600.

    how to solve this problem?

    2. if the converter run resonance frequency /switch frequency as 200KHz, the frequency range of   LLC IC UCC25600 is enough to meet the converter?

    btw ,the power supply output 1500Watts.

    thanks a lot.

    allen

  • the maximum capacitor value on SS  is recommended to 1uF in the datasheet of ucc25600.

    if i add 4.7uF SMD Cap on SS pin, ic ucc25600 work abnormal..

    it seem that it is conflict btween SS function and OCP during startup period.

  • You may increase the protection threshold by reducing Rp. The design is based on FHA which is an approximation approach. So a design needs bench tune to make it work better.