• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 403 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.

Original question:

LM3450AEV120V30: Output voltage oscillation issue

LM3450AEV120V30: LED light flashing on/off

Ali Shagerdmootaab
Prodigy 190 points
Part Number: LM3450AEV120V30
Other Parts Discussed in Thread: LM3450

Hello,

I am using the LM3450AEV120V30 development kit and made a small change to make a new prototype. The change is bypassing the second stage, which is a buck LED driver with LM3409HVMY, and making my own second stage with AL8862 on a separate PCB. I managed to connect the first stage to the new second stage with short wires. Note that, I made some adjustment in the dimming signal to make the new second stage compatible with the first stage. I tested my design with different LEDs in different current and voltages and it works fine. So, far I am happy with my changes. :)


Having said that, I tested my prototype in a different place in which the only change is the AC input voltage to the prototype and maybe the noise in the environment. Now, I am seeing the LEDs start to flash (on/off) and it only happens on some of the LEDs not all. I could not find any pattern in which the flashing happens in order to narrow down the issue.

I am not sure if this is the sensitivity of the circuit to the AC input voltage or noise in the environment affecting the wires. I read some articles that it might be due to other appliances specifically other LED drivers drawing pulsating power from the line.

Can you please give me some hint as to why this is happening?

Thank you,

Ali

  • 0
    TI__Mastermind 28655 points

    Hello Ali,

    Have you looked at the input voltage with an oscilloscope?  Have you looked at the LM3450A output with an oscilloscope?  What is the load power?  What is the AC source?  Are you dimming?

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 190 points

    Hello Irwin,

    I did not look into input voltage and output of LM3450 which I will do so, soon.

    The power range is 350mA*5*3V=5.25W and there is no dimming TRIAC/DC/PWM. I directly connect the input to the wall AC source with no dimmer.

    Thank you,

    Ali

  • 0 in reply to Irwin Nederbragt
    Prodigy 190 points

    Hello Irwin,

    I looked a the input voltage and it is a perfect sin wave when the LED turns on/off.

    The output of the LM3450A is 50 VDC and no fluctuation when the LED turns on/off.

    I attached a photo of the LED driver (AL8862) dimming signal. There is no dimming so I expect to get a DC signal on the dimming pin which I see in the photo. The DC level is about 4V when the LED is on. Only when the LED flashes off I see few little spikes around 1V on the dimming signal as you can see in the photo. I am not sure if these can cause the LED to go off. What is the reason for these spikes? I don't know!

    The LED stays on most of the time but if it flashes off the off time is about max 1 second. I don't think this is the thermal protection because it should not come back on immediately. I also checked the temperature of the new second stage LED driver and the inductor temperature goes up to 80 degree C (~176 degree F) the LED driver is about 40-50 degree C. It seems to be normal.

    The DC voltage on the dimming pin between 0.4V - 2.5V dim the LED light 0-100%. Voltage higher than 2.5V set the output current on 100%. I want to set the dim voltage to 5V in case of no dimming (higher than current level 4V) to make sure it does not fall below 2.5V due to the spikes and also add some caps to keep it steady.

    I appreciate if you have any comment.

    Thank you,

    Ali

  • 0 in reply to Ali Shagerdmootaab
    TI__Mastermind 28655 points

    Hello Ali,

    If the LM3450 output voltage is present and the dim signal is there then there is something happening with the AL8862.  Have you tried running that portion of the circuit on a lab supply and fixed/variable dim voltage?

    Noise spikes shouldn't affect the circuit, they probably are not real and being picked up by the oscilloscope probe loop since this is a switching power supply.  The dim signal should be filtered near the AL8862 input as with any LED driver to make sure it is a clean DC level.

    You inductor is getting pretty warm, have you looked at the current in the inductor with a current probe?  Have you gone through the design calculations to make sure all the parts are sized properly, such as the inductor saturation current?

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 190 points

    Hello Irwin,

    Yes, the issue was with AL8862. It has a exposed pad (EP) under the chip which has to be connected to thermal mass and also the ground which I missed. After connecting this pin to GND, now, it is working without any flashing so far.

    Thank you,

    Ali