BJT power supply

[Kevin Weddle]

The regulator does not have output filtering as switching regulators must have. Is the 555 really prepared to deliver a linear result under a massive voltage swing and load on the 555 that could be anything given the the surrounding circuitry. Don't like switching regulators to begin with and I can't determine any feedback if even they use it. What is that zener doing? A constant gate voltage?

 

[Alun]

Kevin, this is not a switching regulator it's a linear regulator just like the LM7805.

The switching element just doubles the supply voltage to lift the gate of the MOSFET high enough above the output in order to turn it fully on.

The zener is not a normal zener diode, it's a programable shunt voltage regulator, it adjusts the voltage on the gate of the MOSFET to keep the output voltage constant.

 

[ante1]

Hi Kevin,

I think maybe you might have to read up on the subject! ;D

 

[Alun]

I've redesigned it, this should be a lot better.

low_dropout_reg.pdf

 

[ante1]

Hi Alun,

It might work but for stability don’t you think there need to be some kind of voltage stabilizer for the op-amp more than just the 10uF cap? The op-amp is more likely to be sensitive to fluctuations in the voltage than the original TL 431, they work in a completely different way.
Just a thought! ;D

 

[audioguru2]

Hi Guys,
If Alun uses a P-channel Mosfet in his circuit then he won't need a 555 voltage doubler. ;D
The MC33171 is a low-power fairly slow opamp. I would use a faster one.

 

[Alun]

Hi Alun,

It might work but for stability don

 

[audioguru2]

Hi Alun.
Looky, looky: ;D ;D

View attachment 37560

 

[Alun]

Sorry, I misread you last post, I'm the one who needs my eyes testing. ;D

I already knew I could've used a P-channel MOSFET, but they have a larger on resistance.

I was wrong earlier when I said I don't need a fast op-amp, I do because the MOSFET is connected as a source follower so I wouldn't getr any voltage amplification, but I would if I used the P-channel idea which would have a better transient response.

By the way couldn't you think of a cheap, faster, low power op-amp with a low saturation voltage? ;D

 

[audioguru2]

Hi Alun,
This is a linear regulator so the on resistance of its Mosfet doesn't matter unless it is drawing about 40A or more, but then the poor thingy would melt anyway.
I selected the MC34071 because it is the same as the MC33171 that you used except it is faster (10V/us), saturates with a lower loss and can output higher current. At Digikey it is cheaper than yours. Isn't it nice that the dual version of both ICs costs exactly the same as a single one of yours.
Looky, looky again. ;D

 

[Alun]

Hi Alun,
This is a linear regulator so the on resistance of its Mosfet doesn't matter unless it is drawing about 40A or more, but then the poor thingy would melt anyway.

Yes I know, but a lower on resistance means a lower dropout voltage.

 

[ante1]

Hi Guys,

Looks good, and what would you use as Vref?

 

[Alun]

Any low current constant voltage source such as a shunt referance made with a zener diode and resistor,  or a voltage referance IC such as a a band gap referance.

 

[ante1]

Hi Alun,

So, why did you omit the TL 431 then?

 

[Alun]

To make people think. ;D

No seriously, often things like voltage referances and current sources aren't shown because they can be any of many things. Here it could've been a zener diode and pullup, band gap reference such as the LM113 or the programable shunt regulator TL431A. The person using this circuit can decide and choose which type of volatge referance is best for thier application.

 

[ante1]

Hi Alun,

Well you did it, made me think that is! I had to since you went from a complete circuit to an improved one and left some blanks in there! ;D
I know what you mean. 8)