Linear Regulated Power Supply with LM317T & MOSFET 50N06

[MMDuino]

Dear All

I've built a (linear power supply 20V DC 6A) the output driven by MOSFET 50N06 but i got heat problem, i dont know why there is overheating on the MOSFET even with low loads, e.g. 0.42 mA.

(I know i have to use PWM signal to drive the MOSFET but here i just used simple circuit to drive the MOSFET).

could anyone advice me on this issue...

[Seekonk]

Why the inductor, with some wiring that might oscillate.  Might I suggest a TL431 at the gate instead with a simple pullup and give it some real regulation.   Get you down to 2.5V and maybe a second 431 for a negative offset.

[BradC]

Please excuse me if I have the wrong end of the stick, but in who's world does a fixed voltage on the gate of a mosfet cause it to "regulate" anything? Are you trying to use a MOSFET like a bipolar transistor?!?

[PSR B1257]

I've built a (linear power supply 20V DC 6A) the output driven by MOSFET 50N06 but i got heat problem

That's not your biggest problem.
You are driving the MOSFET wihtout feedback, that's not the best way to build a power supply, in fact it is completely wrong.

And what should the LM317 be good for?

By the way, there is a even more inherent mistake in the circuit, the 3V Zener. You cant get the MOSFET to "output" 20V, when the supply is only 24V and a 3V Zener at the gate. In this way, the gate-source Voltage is limited to 1V, which is insufficient to drive the MOSFET.
Stupid me 
The Zener in this case is obviously doing nothing, since the is basicly no current (i don't take the leakage current of the MOSFET into a count) through it.

[Alex Eisenhut]

Where does this "24V/6A" come from, and why do you write 6A?

If that's really the maximum load, then that 1 ohm resistor will drop 6V, therefore dissipating 36W, yet you have a 1W unit there?

Let's start there. Why did you put that resistor there? Besides the fact that even below full load it will catch fire, it will only leave 18V for the rest of the circuit at full load...

That is, until it catches fire.

[PSR B1257]

There is no loadcurrent through this resistor 

[Alex Eisenhut]

There is no loadcurrent through this resistor 

Oh, he's using the LM317 as a reference? Interesting.

[BradC]

That is, until it catches fire.

Isn't that resistor just in the regulation line for the gate on the MOSFET? So it shouldn't catch fire (although that would be really funny).

[Alex Eisenhut]

That is, until it catches fire.

Isn't that resistor just in the regulation line for the gate on the MOSFET? So it shouldn't catch fire (although that would be really funny).

But why write 24V/6A there?

[BradC]

But why write 24V/6A there?

?? No idea, but it's no less puzzling than the rest of the circuit.

[Alex Eisenhut]

But why write 24V/6A there?

?? No idea, but it's no less puzzling than the rest of the circuit.

I didn't get that far I guess, just saw the 317... It's not rated for 6A in any case. In that case the 1R resistor doesn't do anything either, so it's still puzzling.

I just wanted to take things step by step.

[BradC]

I just wanted to take things step by step.

I understand your desire (I think). I just saw the MOSFET gate connected to a voltage divider via an oddly positioned zener diode and switched into "take nothing seriously" mode.

Having said that, I picked this thread up from the "show unread posts" link and missed it was in Beginners. So I probably should not be so unkind.

[wraper]

Really strange schematic. Actually it is pretty useless as regulated power supply. And completely wrong way how to increase current capability of LM317.

[Alex Eisenhut]

I just wanted to take things step by step.

I understand your desire (I think). I just saw the MOSFET gate connected to a voltage divider via an oddly positioned zener diode and switched into "take nothing seriously" mode.

Having said that, I picked this thread up from the "show unread posts" link and missed it was in Beginners. So I probably should not be so unkind.

Well we should help people before they hurt themselves. If he really has a power supply that can deliver 6A at 24V, he's in for serious trouble at some point with this kind of circuitry. I mean almost 150W into whatever can cause problems, 6A through small wires or breadboards can also be a problem, etc.

By starting on the left, I wanted to get all the assumptions out in the open first, as well as a lesson on Ohm's Law...

I guess my lesson is to go all the way to the right and look at what's there. I blame the half bottle of Glengoolie Black I had for supper last night.

[Zero999]

This isn't the correct way to do this.

The MOSFET is configured as a source follower so the gate voltage needs to be 3V higher than the source but you've got a zener diode in there too which increases it further.

The LM317 has good voltage regulation but connecting a source follower to its output is not the correct way to go about increasing its output capability. The usual way to achieve this is to use a bypass transistor to direct current round the LM317 when the input current exceeds a certain level.

[BradC]

The MOSFET is configured as a source follower so the gate voltage needs to be 3V higher than the source

A "source follower"? WTF? Got a reference?
<edit> let me rephrase that. Where do you get the 3V reference from? Isn't it completely dependent on the FET, its gain, temperature and phase of the moon?

[PSR B1257]

http://en.wikipedia.org/wiki/Common_drain

[Zero999]

The MOSFET is configured as a source follower so the gate voltage needs to be 3V higher than the source

A "source follower"? WTF? Got a reference?
<edit> let me rephrase that. Where do you get the 3V reference from? Isn't it completely dependent on the FET, its gain, temperature and phase of the moon?

It's dependant on the gate threshold of the MOSFET and the drain current.

[MMDuino]

Dear All

i've changed some items in my circuit, and now it's working correctly without heat, i got DC-DC adjustable regulator from 3.4vdc to 18vdc with 6 Amp output.

please if i have wrong things in my circuit, i'll be highly appreciated for your correction replies.

[PSR B1257]

Well, everything in front of the 20k potentiometer, which drives the MOSFET is essentially useless. You achieve the same result if you connect the 20k pot directly* to the input voltage. The diode at the output too.

Of course you can vary the output voltage, but there is no way to actually control it. That's where opamps come in really handy 

*) But thats not a good idea either. In this case you pass the maximum gate-source voltage of the MOSFET, if you got a low resistane load at the output.

[MMDuino]

the OpAmps (Lm324 & LM339 more over everything here it's made by China  , so no comments) it's always confuses me, for that i used LM317, but why that is useless to use it, practically it's works well.

[MMDuino]

I know very well the MOSFET behavior almost same with (thyristors & AC voltage), but i don't know practically its works well here...

Please excuse me if I have the wrong end of the stick, but in who's world does a fixed voltage on the gate of a mosfet cause it to "regulate" anything? Are you trying to use a MOSFET like a bipolar transistor?!?

[PSR B1257]

practically it's works well.

Yes, as I sayed. As long as the MOSFET has a voltage at its gate, it will do something. But it is far of controlled in any way.
Here is a very simple circuit for voltage regulation.

Edit: You can also use your N-cannel MOSFET, it just requires to flip things around here and there.

The LT1009 is a voltagereference. You don't essentially need it, instead you can use your LM317, even if it is not the best reference.

[Zero999]

As I said before, there's no point in adding buffer after LM317.

Here's the correct way to increase the current capacity of the LM317.

When the current through the LM317 exceeds about 60mA, Tr1 starts to conduct, diverting the current round the LM317.

This circuit has its shortcomings: mainly being the lack of short circuit protection, but this can be added fairly easily if needs be.

[wraper]

I know very well the MOSFET behavior almost same with (thyristors & AC voltage), but i don't know practically its works well here...

Works well??? Then try to measure load regulation of this junk.