• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 63 subscribers
  • Views 689 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.

TPS92661 LM3409 TPS92515 coil inductance

Joerg Beutel
Prodigy 125 points
Other Parts Discussed in Thread: TPS92661-Q1, LM3409, TPS92515

Refering to the description of the TPS92661-Q1 eval module (http://www.ti.com/lit/pdf/slvua51) I really wonder about the huge inductance of the coils of the two buck-converters for the LEDs (L3 and L4). The Bill of Materials is a little confusing too as it specifies 640uH but has a comment that points to 220uH. In all other current LED driver buck converter schematics I looked at recently comparable inductors had values of 22uH to 68uH ... a whole order of magnitude less.

Now the  Eval Module only uses 135mA of current, but a real design with Power-LEDs would run with about one Ampere or more. Shielded inductors with such saturation current and inductance would be huge and expensive.

I would just disregard this if there wasn't some wording in the Datasheet of the TPS92661 to the effect that the buck converter should always run in constant current mode (CCM) so that the coil never stops to supply current.

On the other hand I really don't like this as these values are in contradiction to all other schematics so that I'm not even sure that the LM3409 or the TPS92515 would work with inductors so large. Wasting board space just to find out later that it's not necessary would be bad. Not providing space for coils large enough would be worse. And testing without first making a board is not possible these days.

So assuming that the target is a high-quality per-LED dimming of 8 Power-LEDs with 1 Ampere each using above mentioned chips ... what coil specs do I really need ?

  • 0
    TI__Mastermind 25500 points

    Hello Joerg,

    The EVM is really just for evaluation and testing firmware and the reason for the high inductance is simply to keep it in CCM with such a low output current. You can easily do a design for 1A with a much lower inductor value, you just want to keep it in CCM. So you could use the datasheet to calculate for a switching frequency/inductor value you would like that will keep it in CCM and it will be a much lower value as you could theoretically have a peak-to-peak inductor current as high as 2A (although I wouldn't suggest it be that high). In any case you will likely end up on the range of 10uH to 68uH depending on switching frequency and LED ripple current requirements for you design at 1A. You just need to be a bit more careful with the thermals at the higher current (another reason why the EVM is such low current).

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 125 points
    Thank you Clint ! So in other words one either needs a high-inductance coil with low current tolerance or a low-inductance coil with high current tolerance ... or if the current can vary in large boundaries things get a little ugly :-) ... My trouble is that we still don't know how much light the users will require in the end so that we plan for 350..700mA and want to keep at least 1A as an option for a rainy day ... But now I know where that high value comes from and can make my plans :-) ... Thanks !