Hello,

For a school project, I need to understand a schema of a LED driver from my teacher and then plug it on a PCB.

The last thing that I'm bloqued on, is the change of the switching frequency by an open collector comparator.
Here is the schema :

My teacher is saying that the way to change the commutation frequency is to change the resistance R2 and R3 (also the inductance but here I can not). I do understand why we need in this comparator a pull up resistor for the BJT collector but I don't understand the negative feedback...
Also, He says that D5 was useless because it as no current throught it. Anyone does have another answer ?

Here is also the LtSpice for this one.

Hope someone has a good answer to this !

 

your U4 is wired so that the more current at R4 the more negative it goes / the more positive the U3 goes / the more negative the U5 goes - thus forcing the U1 more to SGND

i donno what the LTC4441-1 does

 

There are several people working on this project.

He says that D5 was useless because it as no current throught it.

True. useless

but I don't understand the negative feedback...

It looks like positive feedback across U1
---edited---
I think this is not working the way you teacher thinks it is. So you are going to get blamed for that. Sorry.

 

It looks like positive feedback across U1

the phenomena is labelled "Hyst"eresis on the schematic ?

i don't want to dig into another d/s . . . the mystery item is the V4 ? a square wave source . . . suppose it gets modulated by current sense feedback and relatively tiny but semi-logarithmic hysteresis on LM311(/LT1011) ??? - - but it's a square wave not the sawtooth -- which unlikely gets trapetzoidal at that timing . . .

Zouglou LeMagicien : see page 9 (they say that RDVcc is regulated from 5 to 8V . . . you bypass the built in regulator in a way that is not confirmed safe)

 

Check the voltage of the AD712 operating amplifier inputs. This voltage is substantially higher than the operating amplifier's supply voltage. This is very bad!