• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1050 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.

UCC39002: Load-Share Controller

Naser S.
Genius 11605 points
Part Number: UCC39002
Other Parts Discussed in Thread: UCC29002

Team:

Attached is page 25 of the UCC39002 User guide. I marked up the schematic with the latest values that customer has been testing the evaluation board. I also attached the datasheet of the power supply that Ihe is trying to get the current sharing to work at full capacity. At this time the best he  can do was to 5.1A at Module 1, 4.7A at module 2 and 4.5A at module 3. If he trys to get more current, the brick would go into hiccup mode.

 

He measured the internal pull up on the SP7753 across the +Sense pin to the +Output pin and it is 100 Ohm. He also made several scan of each module for the Band width and the lowest was 36kHz. With these values, hecalculated for R (Adj), C(CEA) and R(EAO).

 

Regards,

 

sluu166a 25 send to TI for help.pdfSP7753 Datasheet R_1.0.pdf

 

  • 0
    TI__Genius 11605 points
    Hi, He did more testing today with setting the output with 1amp load (12ohm on the power resistor).
    only have 1 module connected at a time.
    With only module 1 connected, I measure the voltage between U1 pin 4 and pin 8. This is CSO and got 0.598V.
    With only module 2 connected, I measure the voltage between U2 pin 4 and pin 8. This is CSO and got 0.43V.

    Why are the voltage not the same or at least much closer in voltage? Could this be the main reason why my setup are not able to current share at full load?

    Thank you,
  • 0 in reply to Naser S.
    TI__Genius 9140 points
    Hi Naser,
    Thanks for your interest in UCC39002. CSO is the output of the current sense amplifier. The amplitude of this pin is depended on the voltage between CS+ and CS-.
    From your information, if the output current on R10/R11 and R12/R13 is the same, the CSO voltage should be the same.
    Can you please check the R10/R11, R12/R13 value to see whether the load current is the same and to see whether the voltage between CS- and CS+ is the same?
    I think this is the reason that the three modules do not have good load share performance .
    thanks.
  • 0 in reply to David Huang33
    TI__Genius 11605 points

    Thanks David. The load current is the same on the sense resistors.

     They analyze the initial startup of the three bricks. We noticed that in the initial power up, the current sharing does not occur yet. One out of the three brick modules are always asked for max current. After the brick reaches max current, it will go in huccip mode. Then a second brick is ask for max current, then goes to hiccup mode. The last brick does the same.

     They did some more testing and made modification to the current limit of the brick. After modifying the brick output current limit from 10Amp to 20amp, they are able to get about 20 amp current sharing between two modules. One of the modules are not connected at this test. The main issue here is that the bricks we use all have current limit and huccip mode. This will not work with the UCC39002 due to the fact of the start-up logic (page 9 of the datasheet).

     

    They ordered the UCC29002/1 which does not have the start-up logic. They hope this will solve the start-up problem. Do you agree. Any inputs. Thanks..

  • 0 in reply to Naser S.
    TI__Genius 9140 points
    Hi Naser,
    Yes, I agree. Can I know what time you will get the new samples result? thanks.
  • 0 in reply to David Huang33
    TI__Genius 9140 points
    Hi Naser,
    I think we have discussed this issue in the email. let me close this post.