Re: Another design

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

piglet <erichpwagner@hotmail.com> wrote:

On 10/05/2024 4:11 am, Edward Rawde wrote:

"Edward Rawde" <invalid@invalid.invalid> wrote in message
news:v1ho21$4ps$1@nnrp.usenet.blueworldhosting.com...

Now with 9V to 15V input range, direct gate drive and 500mA LED current.
 
Startup behaviour is not ideal but maybe that can be fixed if it needs
fixing.
 
What's the cheapest op amp I can use for U2 instead of OP07?
 
Is there a cheaper comparator instead of LT1719 which will work?
Also the fet is bigger than it needs to ne.
 
Is there a better device for D4? Particularly if R4 can be reduced.
 
Will it work at all? Except in simulation.
 

 
I think I've answered my own question for the current controller but I don't
think it's feasible to attempt a discrete comparator on cost grounds.
 
Version 4

(Sniiip)

 
 
The venerable LM311 is still a good value comparator, if you want to
penny pinch omit the gate driver buffers and lower the pullup resistor
but efficiency will drop a lot from 85% to ca 72%.
 
 
 
 
 

I’m not following through all the spicery, but fwiw:
The positive drive of an LM311 can be improved by 100x or more, by adding a
2-cent NPN and a 2-cent diode.  The NPN is an emitter follower, and the
diode goes from base to emitter, reverse biases when the NPN is on.
 
It pulls down to a diode drop above ground, but that can be a Schottky, and
only has to handle the base bias, so it can be below 0.2 V, but you don’t
care about 0.7 V in a gate driver anyway.
 
When the 311 goes low, the diode doesn’t start conducting till the
transistor is off, and there’s no stored charge to worry about, since the
NPN never saturates.
 

 
Oh, and take the output from the emitter.
 
Cheers
 
Phil Hobbs
 

Right , I actually did as you described in an earlier version but a PNP is
practically the same cost as a diode and provides added beef.

--
piglet