Need Feedback on my 3A buck driver!

[Giannis_TDM]

Here it Is! I am totally open to what parts I used In order for you to give me good feedback on my design.
Specifications:
Max current 3A
Max Input voltage: 18v
Topology: buck (vout less than vin)
Protections and features:
Output short curcuit protection, Soft start, Reverse polarity protection, Thermal shutdown of the main ic, PWM dimming(pads to utilize it will be added on future revisions).
Intended for high power blue,Green and Red LDs

 

[djQUAN]

2 things:
The input polarity protection mosfet - connecting it that way would risk damaging it at 18V. most mosfets are rated at 20V max gate -source voltage. A resistor and maybe a zener across gate to source to limit voltage and protect it.

The dimming circuit. a quick look at the datasheet seems like dim input is PWM TTL. The pot might not work as planned.

Minor - the trace from the sense resistor to the feedback pin does not need to be really thick. It is just a low level signal going to a comparator.

 

[Giannis_TDM]

2 things:
The input polarity protection mosfet - connecting it that way would risk damaging it at 18V. most mosfets are rated at 20V max gate -source voltage. A resistor and maybe a zener across gate to source to limit voltage and protect it.

The dimming circuit. a quick look at the datasheet seems like dim input is PWM TTL. The pot might not work as planned.

Minor - the trace from the sense resistor to the feedback pin does not need to be really thick. It is just a low level signal going to a comparator.

> Minor - the trace from the sense resistor to the feedback pin does not need to be really thick. It is just a low level signal going to a comparator.
I did that because I wanted to minimize the resistance of the trace back to the chip and also because I had the space to do it
>The input polarity protection mosfet - connecting it that way would risk damaging it at 18V. most mosfets are rated at 20V max gate -source voltage. A resistor and maybe a Zener across the gate to the source to limit voltage and protect it.
I will add that on a future revision thanks for the suggestion!
>The dimming circuit. a quick look at the datasheet seems like dim input is PWM TTL. The pot might not work as planned.
That's what I was afraid of, I will run a lot of tests when I have assembled one. I have also ordered multiple pots to test which one works better
Thanks a lot for the feedback! Really appreciate it my goal is to design a Good solid driver with protections for the blue-green and red high power LDs If I deem the design stable with clean output ill think about doing a large run and price it low (about 15-20$). The one thing that inspired me to design the driver was that a very low amount of drivers on the market have reverse polarity protection built in and its just so easy to one day put the battery in wrong and smell the magic smoke. Again thanks for your feedback it's very much appreciated if you want the value of the components I will share the schematic and component No.

 

[gazer101]

As long as it's cheaper than SL you'll have at least one customer

 

[djQUAN]

pin 6 actually needs a thicker trace than the feedback as it is the switching input (all load power goes through it) and preferably its own tap on the input capacitor.

There is a way to artificially pull up the voltage at the feedback to lower the output current but it has a limited range when using a pot. That could be another way to make it variable.

 

[Giannis_TDM]

pin 6 actually needs a thicker trace than the feedback as it is the switching input (all load power goes through it) and preferably its own tap on the input capacitor.

There is a way to artificially pull up the voltage at the feedback to lower the output current but it has a limited range when using a pot. That could be another way to make it variable.

> pin 6 actually needs a thicker trace than the feedback as it is the switching input (all load power goes through it) and preferably its own tap on the input capacitor.
Wow, thanks for noticing! I actually didn't notice that! I increased the size of the trace going to pin 6 to 1 mm
> There is a way to artificially pull up the voltage at the feedback to lower the output current but it has a limited range when using a pot. That could be another way to make it variable.
If the potentiometer doest work Ill give that a try!

 

[Giannis_TDM]

Sorry For the double post but update:
Finalized the design based on the SO-8 Version of the ic and implemented protective components for the MOSFET. In the following weeks, I will be doing a short run of 5 that I will assemble for testing and will send 1 to paul free of charge for further testing when the low cost shipping methods become realiable again since I do not own an oscilloscope then if the design proves solid I will do a run of 20 and will see from there on.