Buck led driver do not stabilize current

Thread Starter

stevetto

Joined Apr 11, 2019
5
Hello,

I have been working on a LED buck driver for a while but unfortunately without any succes.
This is the design, based on a MAX 16820 chip:

I use an attiny13 to generate the pwm signal for the MAX 16820. This part seems to be ok, I can change the pwm's duty cycle.
As the duty cycle changes the LED current also changes.
But if I change the input voltage the led current also changes, so the driver do not stabilize it.

At first I tested it with an XHP70 led, than I changed it to an xm-l2 (the current sense resistor was also changed) but I still have the same issue.

I would really apprieciate any suggestions.
 

Attachments

kubeek

Joined Sep 20, 2005
5,795
What max average current through the led do you get? My calculations say it should limit at about 4.6A.
What is the supply voltage?
 

Thread Starter

stevetto

Joined Apr 11, 2019
5
Yes it is ok.
I used an XHP70 6V led. It needs about 4.8 A. So the current sense resistor is 43mOhm.
For the XM-L2 (3,3V 3A), I used a 68 mOhm resistor.

I also tried different inductors. For the XHP-70. At first I used a coilcraft XAL 7070 4,7uH than changed it to 10uH.
The same 10 uH was used for the XM-L2.

I also attached the calculation of maximintegrated's and my own excel sheet.
 

Attachments

Thread Starter

stevetto

Joined Apr 11, 2019
5
The power source is an adjustable power supply, but I also tried 2-3 cell lipos.
In the future I would like to use 3 cell lipos, but now I would like to make a driver for 2 cells only.
 

Thread Starter

stevetto

Joined Apr 11, 2019
5
At first I changed the voltage within a small range to simulate as the battery discharging.
When I changed the input voltage higher, the led burnt out.


And at higher currents (close to 5A) the driver gets smooking hot.
That is why I replaced the XHP70 by an XM-L2. To see what happens....

I have uploaded a movie here:


The movie shows an XHP70 test at 3 modes (25%, 50% and 100% pwm)
 
Top