• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 473 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.

LM3406HV PWM dimming function problem

WhiteHorse
Genius 4430 points
Other Parts Discussed in Thread: LM3406HV, LM3406

Customer has following question

I am using the LM3406HV in a design, and my issues are as follows. 

 1) Ron resistor is 500K

2) MSS1278-104 is the inductor part number (100uH)

3) Cout is 10uf

4) All other values are straight out of the data sheet. 

 Everything seems to run fine and unit functions straight as the math predicts until we use the PWM dimming function. Once we begin to dim (17.7K dim freq) we run all the way down to 1% and when we return back to 100% the output switching frequency drops by 1/2, and it will overshoot its current until it bounces off the 2A hard limit then the whole IC resets, and returns back to the correct switching frequency, and correct current target. 

 Also, after the unit is dimmed, and output frequency is stuck at 1/2 the target, I can reset the part by grounding the Ron pin and the the unit will fix itself, and return back to the target frequency. 

 So long story short... I cant dim this unit. When I do, it latches into the lower switching frequency, and wont regulate to its current target until it is rest. 

Can you help.

**** end customer question

Regards,

John Wiemeyer

  • 0
    Prodigy 5 points

    Hello John,

    There is a minimum ON TIME that may come into play at very low dimming duty cycles.
    It will lower the effective frequency but not to exactly 1/2 except under one specific operating point.

    MINIMUM ON-TIME

    The minimum on-time for the LM3406/06HV is 280 ns (typical). One practical example of reaching the minimum
    on-time is when dimming the LED light output with a power FET placed in parallel to the LEDs. When the FET is
    on, the output voltage drops to 200 mV. This results in a small duty cycle and in most circuits requires an on-time
    that would be less than 280 ns. In such a case the LM3406/06HV keeps the on-time at 280 ns and increases the
    off-time as much as needed, which effectively reduces the switching frequency.

    Exactly how is dimming implemented?
    There are several ways to dim this device.

    It sounds like a possible layout issue or schematic application.
    Compare the suspect design with the LM3406 EVM users guide SNVS512C.
    Layout is very critical in any switching supply.
    Schematically how does it compare.

    I have seen noise cause a PWM to get into strange operating states.
    The data sheet talks about layout considerations starting on page 27.

    Also when you "return back" to 100%, is the duty cycle slowly increased or immediately jumps to 100%?
    With any converter it is best to slowly vary the output vs attempting a step function.
    Does it work better when the duty is slowly increased from min to 100%.

  • 0 in reply to Ed Walker
    Genius 4430 points

    Hi Ed,

    I appreciate your quick response!.

    Customer says:

    I have located the issue. When the tech was evaluating the part we were not measuring the battery voltage at the unit. The drop across the wire was causing Vin to be to close to Vout. This caused the strange behavior. The device seems to be working properly when we give it enough juice. 

    Thanks for your help Ed!

    Regards,

    John