• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 64 subscribers
  • Views 244 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM3463 Application Support and Alternatives

Liam Wedekind
Prodigy 20 points
Other Parts Discussed in Thread: LM3409HV, LM3463, TPS92830-Q1

Hi,

I am looking to replace the LM3409HV chip we use in a design with a multi-channel LED driver. The product has 90 separate high voltage (29-38V) LED strings that need to be controlled separately. The strings are for LED's in an individual colour and we have up to 20 strings of a singe colour. The design requires that each colour is able to be dimmed separately via an LED driver that can control from 0-500mA. It is possible for us to group these so a single chip could drive multiple strings of the same colour if that helps with the driver selection.

I have been looking for some potential replacement LED drivers and the best I have been able to find so far are:

LM3463: 

LM3463 data sheet, product information and support | TI.com

This a 6 channel (two channels are tied together which we can work around) by making the output of all 6 channels the same colour strings. The limitation with this chip is that there is a common DC bus driving all of the LED strings. The DHC circuit that is used to limit the supply voltage to keep it as close to the sum of forward voltages in the string will have limitations when it comes to the LED strings that have smaller forward voltages. 

We have noticed that with the smaller forward voltage strings the FET's used in the development board get very hot due to the excess voltage in the string. 

TPS92830

TPS92830-Q1 data sheet, product information and support | TI.com

This is a three channel LED driver that can control three channels to the same dimming control. It looks relatively simple and has the benefit of a high side switch FET so my understanding is it will fare better when the supply voltage is higher than the string voltage. 

I am looking for devices with high channel counts to reduce the number of chips required for the design. We need to be able to maintain a dimming control between 0-500mA. The supply voltage is 50-55Vdc.

Thank you.

  • 0
    TI__Mastermind 19940 points

    Hi Liam,

    Do the channels need to be turned on at the same time or sequentially? What dimming method (PWM / analog) and dimming ratio do you need?

    Best Regards,

    Steven

  • 0 in reply to Steven Li
    Prodigy 20 points

    Hi Steven,

    We have some ability to be flexible on how the strings are turned on as the strings are switched on/off in periods of hours not seconds. The current design allows for the strings to be switched at the same time and I would like to maintain this if possible. 

    I would like to be able to have a control resolution of at least 50 steps between 0mA and 500mA. 

    Thanks,

    Liam

  • 0 in reply to Liam Wedekind
    TI__Mastermind 19940 points

    Hi Liam,

    Both LM3463 and TPS92830-Q1 are linear-type LED drivers, so there won't be much difference between the power loss on the external FET if their input voltage rails are the same.

    May I understand if you are looking for an automotive-grade device or not?

    Best Regards,

    Steven

  • 0 in reply to Steven Li
    Prodigy 20 points

    Hi Steven,

    The part does not need to be an automotive device. 

    Thanks,

    Liam

  • 0 in reply to Liam Wedekind
    TI__Mastermind 19940 points

    Hi Liam,

    Then LM3463 will be a good choice for you.

    Best Regards,

    Steven

  • 0 in reply to Steven Li
    Prodigy 20 points

    Hi Steven,

    Ok no problem, I have had a good search and wasn't able to find anything similar.

    Can I ask a follow on question? 

    We have an LM3463 Eval board (AN-2255) and one thing we see is the FET's get hot quite quickly and it looks like the excess voltage on the DC rail is the culprit. You previously noted both design were linear drivers but it appears the LM3409HV driver is significantly better at managing thermal loads on the FET. Is there something we need to do to help the switching losses/conducting losses with the LM3463 low side FET?

    Thanks,

    Liam

  • +1 in reply to Liam Wedekind
    TI__Mastermind 19940 points

    Hi Liam, 

    LM3463 is a linear LED driver, the excessive voltage (Vin - LED voltage) will mainly apply to the FET in the normal working condition.

    LM3409HV is a switching-type LED driver, the FET in its system is used in a switching way.

    You can think these like a constant-voltage Buck DC/DC versus a constant-voltage LDO.

    You can refer to the "Linear Regulator" (start at around 1:46 in the video) and "Buck Topology" (start at around 3:00 in the video) Section in this video - LED Driver Topologies | Video | TI.com. Hope this helps.

    Best Regards,

    Steven