• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 190 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.

TPS92515HV: Practical drive current accuracy

Edward Cooper
Prodigy 30 points
Part Number: TPS92515HV
Other Parts Discussed in Thread: TPS92515, LM3406

Tool/software:

First time working with the TPS92515(HV) and I'm just working out that the delivered current is dependent on the inductor ripple current, which is then dependant on the inductance which is most cases is tolerance to +/-20%.

If I'm designing around an 82uH inductor with a tolerance of 20% then values of between 66uH to 98uH should be expected.

In my 48V -> 17V, 217KHz, 1A design I can expect the following inductor ripple., 98uH = 49%, 66uH = 73%.

When I target 82uH = 1.0A, then at 98uH I get 1.05A, and at 66uH I get 0.93A

Any tricks I'm missing to get a more tightly controlled drive current?  I suppose running the converter faster with a smaller inductor would help, but I'm trying to eek out efficiency in my design.

  • 0
    TI__Mastermind 23166 points

    Hello Edward,

    If you have lower ripple then the percent error of the average will go down due to inductor tolerance.  The other option is to use a different constant current regulator that closed loop which removes the errors due to inductance tolerance.  What is the input voltage range?  What is your load (number of LEDs, Vf of LEDs and is there a range of LEDs number)?  Are you doing PWM dimming or shunt fet dimming and if so what is dimming frequency and what is the duty cycle range?   Do you have a desired switching frequency?  Is size or efficiency a priority?  

    -Fhoude

  • 0 in reply to fhoude
    Prodigy 30 points

    Hi,  the input voltage is 46V to 50V, we have 16-bit PWM dimming at 1.4KHz with shunt dimming also implemented, our LED loads are 32V-38V on one BOM and 12V to 18V on a second BOM. Id like to be around 200KHz to 300KHz for efficiency sake, though that's not a hard rule, we are space and thermally constrained so we need to get the balance right.  Any suggestions on devices would be great, we've had really good use out of the LM3406 across a number of products, but we'd like to also be able to control current hence switching to TPS92515.  Thanks

  • 0 in reply to Edward Cooper
    TI__Expert 5730 points

    Hi Edward,

    Have you had a chance to look at the TPS92682? It has two channels which can be configured for two individual outputs. It also has closed loop, peak current-mode regulation which will give you better accuracy than the TPS92515.

    Alternatively, you can use two separate TPS92691 devices to achieve your intended application.

    Best,
    Daniel

  • 0 in reply to Daniel Louis
    Prodigy 30 points

    What I meant is we use the same design to drive a few different LEDs, so we expect two BOM/SKU/assembly variants of the driver for different current/voltage output options 0.5A at 32-38V and 1.0A at 12-18V.

    Was hoping for an integrated FET solution also.

    Will take a look at TPS92682 as we do also have a requirement for a dual channel solution.

    Thanks

  • 0 in reply to Edward Cooper
    TI__Expert 5730 points
    Edward Cooper said:
    Will take a look at TPS92682 as we do also have a requirement for a dual channel solution.

    If you're looking for a dual channel IC with integrated FETs at similar output requirements, the TPS92519 is a strong contender as it can handle your input voltage range and up to 2A continuous output current

    Edward Cooper said:
    Was hoping for an integrated FET solution also.

    Unfortunately the majority of ICs in our portfolio do not have integrated FETs. The TPS92515 and LM3406 are the only options for single channel monolithic Bucks.

    Please let us know if we can assist you further, we are happy to review your designs and help debug if needed. If confidentiality is a concern, we can arrange for an NDA and we can set up a meeting or converse over email. Feel free to reach out at any time.

    d-louis@ti.com
    Francis.Houde@ti.com

    Best,
    Daniel