PWM cotrolled LED supply using AL3353S-13 issues

Thread Starter

lesrhorer

Joined May 21, 2024
11
I am designing a 100 Watt LED PWM controlled power supply using a AL3353S-13 driver. The datasheet is far from comprehensive. The datasheet is here. I am using the design strait off the datasheet - almost. The only variances, really, are I eliminated D1 on the Figure 1 schematic on Page 2, because if I understand correctly, this should only be required for non-PWM operation, and I eliminated R1, since I don't think it is necessary, either. I am driving it with a GPIO from an Arduino at 5V. Someone let me know if either of these should be an issue.

Those aside, I have a couple of other issues. I think perhaps I may know what the first one is. Page 9, figure 3 shows the chip driving a P-channel MOSFET, and since I have a lot of these, I used one here. When I apply power, however, I get a near dead short of around 2 Ohms. I just noticed the other schematics show an N-Channel That would seem to explain it. Does anyone more familiar with the AL3353 have any other ideas?

The second issue is the datasheet leaves many of the component values unspecified. For example, on page 9 it says, "A best practice is to only add as much slope compensation as necessary to avoid subharmonic oscillation. Additional slope compensation minimizes the influence of the sensed current in the control loop. With very large slope compensation, the control loop characteristics are similar to a voltage mode regulator, which compares the error voltage to a saw tooth waveform rather than the inductor current"
 

Thread Starter

lesrhorer

Joined May 21, 2024
11
OK, fine. I expect avoiding subharmonic oscillations is not a bad idea, but what value of R2 should I use to accomplish this? 1K? 10K? 100K? Since I need an output current of under 4A, I am using a value of 0.1 ohms for R3 and 0.12 ohms for R6.

It doesn't give suggested values for C1, C2, or C3, either. I used a pair of 1000uF electrolytic capacitors for C3 plus a 10uF MLCC from the output of D2 to ground as I mentioned earlier. For C1 I used a single 1000uF electrolytic. For C2 I used a 0.1uF MLCC. For L1 I used a 33uH, 5A inductor.

Of course, I will be replacing the P-channel FET with an N-Channel, but I wanted to get some feedback from others, first.

(This is in 2 parts because I accidentally hit <Ctrl><Enter>, and then the time to edit expired.)
 
Top