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.

600-W reference design SLOU293C Not starting above 120VAC

Other Parts Discussed in Thread: UCC28600

We are using the reference design described on the SLOU293C document for the basis of our power supply. The major differences are that we are not using the PFC, Power limiter or the clip detection circuits. The power supply seems to work well on one application so we decided to use it in another application where the schematic is identical but the PCB had to change to a small degree to accommodate the mechanical package.

Everything was fine until we noticed that 5 of our 6 prototype PCBA's will not power up above 120VAC/60Hz when loaded. The 6th PCBA does operate up to the high line condition just fine when loaded. I have spent many hours verifying components between these PCBA's and have not found why 5 units behave the way they do. Not to mention why the first version used in the other application never had this problem. We are using the magnetics specified in the reference design. The load is 2 Amps on the 50V output and 80mA on the +/- 15V outputs.

If I remove the 2A load from the 50V output, the supply starts OK. I can then load the 50V output with no problem. So the issue is when I have a load on the 50V supply that it does not want to start. The symptom is that when the power supply is loaded to 2 Amps on the 50V output and 80mA on both +/-15V outputs the power supply will start "hiccuping" and will sometimes eventually start after several tries. It will also usually start if I remove the load on the 50V output, let it start up and then apply the load. When the supply is "hiccupping", the +/-15V output is not starting which in turn prevents the 50V output from staring up.

I am thinking we are experiencing an OVP situation on the Aux supply since I can reduce the line voltage below 120VAC and it starts every time. I cannot seem to identify why the Aux supply will not start under the loaded condition. Perhaps garbage on the primary side when the 50V controller starts that interrupts the Aux controller? I would appreciate any feedback to help with solving this problem.

  • Hi Jim,

    I recommend monitor the Vcc rails of the PWM controllers to ensure that they are not dropping and hitting UVLO during the start-up sequence.

    Regards

    Peter

  • Peter,

    Thanks for your reply.

    In fact that seems to be the case. I changed the design topology to match the data sheet and SLUA399B application note. I have added the Cbias and diode "D1") mentioned in the UCC28600 design considerations doc (pg7) to isolate the bias supply on power up.This is in addition to the D2, Rbias and Cstart parts. (BTW - I removed the resistors and diode that came from Rsu and connected to the Css capacitor found in the original reference design... this circuit seemed to have no description I could find as to its true purpose). I have used the design calculator for the UCC28600 to update the values in my design for Rsu, Rovp1, Rovp2, Rcs, Rpl etc... I noticed that they vary quite a bit from the original reference design but the power supply seems to work better at this point.

    Out of the 6 power supplies I have built up and modified, 2 worked perfectly after the change suggested in the application note. The other 4 required increasing the resistance "Rbias" to 100 ohms to work up to 132 VAC. Of  those 4, one is still "hiccuping" above 120VAC. It's the problem child I suppose!

    Any comments on why the 6 power supplies have slightly different behaviors above 120VAC? Is it simply the UCC28600 variation or am I missing something else?

    Thanks,

    Jim

  • Hi Jim,

    Can you post a schematic or email it to me if you prefer at petermeaney@ti.com

    Thanks

    Peter
  • Peter,

    I will email the schematic to you this morning. Ironically, our latest rev exhibited the start up behavior again. I thought I had fixed it by modifying the circuit but apparently it still has an issue.

    regards,

    Jim