Hello, I was trying to current control the o/p of an LM2576, by feeding in o/p of an opamp as mentioned here :
http://www.smbaker.com/lm2576-constant-voltage-constant-current-switching-power-supply
But, the limiter is not quite accurate, lets say, if I set the limit around 400mA, LED starts glowing after 400mA is crossed, and becomes brightest after more 15mA or so. That is also when the final current limit takes place(i.e. increasing voltage does not increase current).

Also, when I measure the feedback pin voltage at current limit state, it shows around 1.7V, which was supposed to stay at ~1.24V.

Considering feedback pin is only the input to an internal comparator in LM2576, I guess it should not hurt the IC much, but I would like to know why is it not behaving the ideal way (i.e. 1.24V at feedback pin). Also, if there is any way to improve performance, kindly suggest.

 

No replies ?

 

How is it set to 400mA?
It sounds like it's actually set to 415mA, the value it settles to.
The circuit is using bang-bang (oscillating) not linear current feedback through the bottom comparator.
That oscillating signal at the FB pin is likely what you are measuring, whose average value is higher than 1.2V.
Do you have an oscilloscope to look at the waveform?

What is the maximum voltage out of the comparator (supply voltage)?
You might want to add some resistance in series with the LED to the FB pin.

 

i have set the current with the pot at bottom right corner. dmm value shows 0.4V at the opamp inverting pin.

that is the original schematic by author, in my circuit, i have inserted a 220ohm between led and opamp (bottom one) o/p. also put a 0.1uf between inverting pin and o/p, hoping that will reduce high frequency oscillation.

well, i can arrange to check with an oscilloscope, will check that, may take a day or two.

 

That oscillating signal at the FB pin is likely what you are measuring, whose average value is higher than 1.2V.
Do you have an oscilloscope to look at the waveform?

I am sorry for being so late, but it took some time for me to arrange the oscilloscope pictures. Hope you understand.

But here is what I found, both the pictures are when scope was in DC mode, per Y division was set at 1V.

I set the current limit at 280mA, by setting the voltage divider at 0.28V, then connected a 10ohm resistance with o/p of the supply. Then Slowly increased the o/p voltage from 1.24V to upwards. The led started glowing when I hit 280mA, and it started limiting.

At this point, I probed the feedback pin, and it was like pic 1. The bottom of the wave was 1.24V, but the highs were high.

If I further tried to increase the voltage o/p of 2576 a lot, then it slowly slow increased the current to 290mA+. It was limiting, but there was such different between set limit and actual current. At this point the feedback pin was like pic 2.

I also probed the load resistance terminals, and found it was super spiky at this stage. But it was well regulated when below limit.

I am using op07 as the diff amp, and LM358 for driving the feedback pin of lm2576 via led.