• State Resolved
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 380 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.

UCC28782: ucc28782 eval kit

robin gangopadhya18
Mastermind 6316 points
Part Number: UCC28782
Other Parts Discussed in Thread: TL103W, , TPS92074, TPS65987, TPS55288

Hello:

In a separate application, we need to have our UCC28782-based module behave as a "current-controlled" source. Such that its output voltage will change in accordance with the load current.

What would be your suggestion for achieving this out of this module? We will try the scheme on the TI Eval kit we have.

Thnx

-r

  • 0
    TI__Mastermind 40580 points

    Hello Robin, 

    Yes, the UCC28782 may be configured as a current-controlled source similar to the arrangement as a constant-current (CC) source using the TL103W controller. 
    However, a CC feedback uses a fixed voltage as a reference for the output current.  
    By extension, varying the reference voltage would vary the "constant" current to a target current level. Generally Vout necessarily has to vary to accomplish that. 

    Your application is a little different than that, I think, because (as I understand it) you're looking for a target Vout for a given Iout.
    I am concerned about positive feedback, since changing Vout for a given Iout may then change Iout which will then change Vout which changes Iout, etc ,etc.   
    If the reference current is independent of the output current, adjusting Vout should work okay. 
    But if the current is an indirect function of itself, this may not work.  

    Regards,
    Ulrich

  • 0 in reply to Ulrich Goerke
    Mastermind 6316 points

    Ulrich:

    Thanks for the review. I think I am trying to shoehorn one module we painstakingly developed( and you know better than any others), and fearful of starting another chase- our hunt is for a unique LED driver asap. Looking at options  in TI- and would be grateful for you suggestion of a few- it is better to use one for the purpose than shoehorning operation.
    Our power goals are 50W to 1000W, controllable, with PFC the most important feature of the circuit. All GaN as in the UCC28782 play.

    What are the options in TI?

    robin

  • 0 in reply to robin gangopadhya18
    TI__Mastermind 40580 points

    Hello Robin, 

    Conceptually, I get the typical approach for LED drivers, but experientially, I am a novice as far as design details go in this arena. 
    The best that I can do is point you to TI's starting product page for LED drivers: https://www.ti.com/power-management/led-drivers/overview.html?keyMatch=LED%20DRIVERS  which I suspect you may have already perused. 

    As far as the power range goes, the ACF can (at the limits of practicality) be extended to 120W, maybe 150W.  Beyond that other topologies are needed to keep the peak currents reasonable.  

    GaN can play a leading role in each of the sub-circuit categories (PFC, LLC, etc.) and I'm sure TI has experts for each of them, but I've reach my limit of usefulness here.  And, although you are looking to design comprehensive systems, you would get better response from these E2E forums by breaking up your queries into separate items discussing separate controllers because each thread can be directed to the appropriate support person. 

    Meanwhile, TI may already have a reference design that may be applicable to some segment of your power range. 
    Here is a link to a searchable list: https://www.ti.com/reference-designs/index.html#search?applid=120,902 
    At least it may provide an overview of the types of conversion topologies typically used at a particular lighting power level. 

    Regards,
    Ulrich

  • 0 in reply to Ulrich Goerke
    Mastermind 6316 points

    Hello Ulrich:

    Thanks a lot for the review. Let us digest that. Meanwhile, we found the TPS92074 controller- which seems perfect!

    Here is the $64 question: we are not going in this play to by $5 driver-- all of which break in much <<50,000 hrs boldly claimed but never ever seen( other than tunnel effect)...so we will have a"sensor" modifying the "constant OFF " time. 

     A freaky thought but I think  this chip is able to play it.

    Cld you pl look into that?

    Talking about "freaky" thinking, check out my post regarding using only the internal switches of TPS55288 when we have only a buck situation...! Provided of course the I2C has commands to do it....in our USB C PD module with TPS65987....

    r

  • +1 in reply to robin gangopadhya18
    TI__Mastermind 40580 points

    Hi Robin, 

    I'm glad that you found that the TPS92074 seems to fit your needs.  As I mentioned before, I'm out of my league as far as LED drivers are concerned. 
    If you need support for that part, please start a new thread with that part number in the topic and it will be routed to the appropriate group. 

    The same with the other TPS numbers.  Some other group(s) handle those parts and I have no experience with them and am unable to comment concerning their usage.  I do hope that they will work for what you have in mind. 

    I think that I have addressed your original question as to whether the UCC28782 can be can be controlled as CCVS, which I believe it can be as long as the controlling current is independent of the controlled voltage (to avoid positive feedback).  Since I can't support your TPS questions, I'll close this thread. 
    It'll be better to have a separate thread on those.

    Regards,
    Ulrich