• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 563 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.

LM3405A: problem with variation on reference circuit

Jon Elson
Prodigy 20 points
Part Number: LM3405A
Other Parts Discussed in Thread: LM3405, LM3406, TPS92515, TPS92511

I am working on an LED retrofit to be used with an AC-dimming Halogen lamp supply.  (10,000 units in the field!)

This will be running a single white LED at about 1 A max.

Since this will be running off pulsating DC, I wanted to make the boost supply power up as soon as the source comes up.

The first page "Typical application circuit" assumes a known, constant Vin.  So, my plan was to put a 5.1 V Zener between Boost and SW, and feed Boost with a resistor to Vin.

This ran FINE on a bench power supply at 12 V, but when I connected it to the dimmer supply, the LM3405 failed with a short between Boost and SW.

Probably due to a soft knee on the Zener, I'm only getting about 2.8 V from Boost to SW, so may need to feed it more current.

But, can anyone see a real problem with using a Zener between Boost and SW to limit the Boost to SW voltage?  Is there any issue with this circuit essentially being powered on and off 120 times a second?


At first I thought Figure 25 was not applicable, as the Vout would be too low with just ONE LED.  But, maybe this is a better circuit, although it would run with less Boost voltage.

Thanks for any thoughts and advice!

Jon

  • 0
    TI__Mastermind 25500 points

    Hello Jon,

    Is your input simply a rectified 12VAC? How are you attempting to dim, triac? Triacs themselves can cause a lot of issue themselves with transient if so.

    But regardless, there should not be an issue with protecting BOOT with a zener. But if it is rectified AC, chopped or not, there are two things to consider:

    1. If you are using a resistor from the input to BOOT with a zener you will also need a blocking diode in series with the resistor. Otherwise each time the input falls you will discharge BOOT through the zener and resistor. This not only puts a negative voltage on BOOT-SW which may exceed abs max, but it will require peak charging current of BOOT each cycle. Running BOOT from the output is valid in this case as well, but only if you consider #2.

    2. This device is designed for PWM dimming, but I'm not so sure about startup and shutdown at such a high rate. You could actually have really strange transient depending on how much components are charged at the point of starting back up and so on. When you completely shut down the internal compensation will discharge, probably to an unknown level, leading to even more transients. I would have to see it on the bench to be sure but I wouldn't be surprised if you have some interesting stuff going on. I only mention all of this because if I were to try and use this with rectified AC, dimming or not, I would do it a little differently. I would put a blocking diode between the input supply and VIN (with the input capacitor connected to the VIN pin). Then I would use a resistor divider from the input supply to the EN/DIM pin to disable the device at a certain input voltage. This would keep the part alive which will keep COMP and BOOT both charged for the next turn-on cycle. This is sort of the principle of how the LM3406 works which was designed to run with a square wave input voltage.

    Another option is to consider something very fast that can turn on and off at 120Hz such as the TPS92515.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 20 points
    It is a triac-style dimmer then chopped at 70 KHz, and fed through a transformer to get to 12V. But, after a full-wave bridge recitifer, it is essentially a rectified 12 VAC.

    Yes, I also had the blocking diode to prevent discharging the boost supply.

    Well, I went to the Figure 25 scheme, and it seems to work fine. So, I will do some more testing, but this is looking pretty good.

    I did try using the TPS92511, but then they told me we needed to run ONE LED at 1 A.

    Thanks very much,

    Jon