LM3409HV

Hi Mike,

Thank you for your interest in LM3409.

Would you please let me know the something more about your LED configuration?

You want to use one LM3409 to drive 4 LED that are in parallel or you want to use 4 LM3409 to drive 4 LED but dim them using one signal? LM3409 should be able to handle both cases with proper implementation.

3.3V should be fine. You may try it on the EVM.

Best regards,

Victor

Hi Victor,

Sorry i wasn't clear.

I will be using 4 LM3409 chips to run 4 separate LEDs using the same PWM input.

Do i need to make any changes to the eval board to use 3.3V?

Thanks

Mike

Hi Mike,

By reading through the application note of the EVM board, the only impact of changing 5V to 3.3V I think is that it will lower the gate current to charge Q2 and Q3. If you really want to keep the conduction speed of the FETs same as before, you can reduce R1 and R2 by 33% which I think is not so necessary as shunt FET dimming is already very fast compare to other PWM dimming methods:)

As long as the 4 circuits are identical and controlled by the same PWM input, they should turn on at the same time and have same brightness while dimming. Be sure to make them share the same ground or to make their ground as close as possible if you will be using separate PCBs.

Best regards,

Victor

Hi victor,

I have tried to get the board working at 3.3V but it doesn't seem to be working.

Q2 doesn't appear to be switching and the LED is on contantly.

it would seem that it will not switch at any voltage below 5V.

i have also tried reducing the resistor values for R1 and R2 but this seems to make no difference.

Do you know of an alternative FET i can use which will switch at this voltage?

Thanks

Mike

Hi Mike,

I guess it was Q3 that was not switching as I found in their data sheet that Q2 has a Vgs threshold of 1.8 typical (1V min, 3V maximum) and Q3 has a Vgs threshold of  2V to 4V. .3.3V happens to be a little tricky in this situation. Sorry I didn't notice this in the first time.

May I ask the frequency of your PMW signal?

Thanks and best regards,

Victor

Hi Victor,

You are correct, it was Q3 that wasn't switching. I had my scope probe in the wrong place which confused me.

I have in fact solved the problem. I replaced Q3 with a IRLML0100TRPbF which has a lower Vgs threshold.

I can also confirm that the outputs from two Eval boards are in sync providing the circuits are set up the same.

Thanks for your help.

Mike

That's great. I am glad I can help.

Best regards,

Victor

Hi Victor,

I have now designed and built some prototype boards with the same circuit as the Eval board.

however the output as well as several other parts of the circuit which i have scoped have a very different response.

i have attached a scope trace of the output of my circuit and the eval board for comparison.

do you have any suggestions as to what might be causing the problems?

if you need any further info please ask.

thanks

Mike

the output from the Eval board:

The output from my circuit:

Hi Mike,

Have you tried to swap the components on EVM and your board?

Is there anything abnormal on the switching node?

Best regards,

Victor

Hi Victor,

It turns out I had ordered the incorrect value for Coff. 470NF instead of pico which caused steppiness due to having the wrong switching frequency.

I now have the board working well but when it is dimmed to zero it makes some audible noise.

this is also present on the eval board. i think it is because the circuit is switching as expected but the output is shorted out by the shunt fet.

is there any way to stop this noise?

thanks

Mike