0-40V out voltage regulator circuit

[panfilero]

Hello,

I'm trying to make an adjustable voltage output circuit that can swing from 0-40V... I'm planning on using a LM317HV and was going to use one of the circuits from their datasheet:



So, I tried to see what's going on here, and I figure it's

Vout = 1.25V + 1.25(R2/R1) - 1.2

which (if my equation is right) would give me approx 0V when R2=0, but when R2 = 5k.... I get.... Vout=41.7V...

so my questions are... is my equation wrong? is it possible to modify this circuit so it can give 0-40V out?

much thanks!

 

[ronsimpson]

By pulling the bottom of R2 negative 1.2 volts will give you the ability to reach 0 volts.

 

[Willbe]

1.2v/150Ω= 8 mA.
8mA x 5kΩ=40v.

 

[Hero999]

I hope you're aware of the negative supply and that it won't be ab;e to supply 1A over the entire voltage range.

Here's another way of doing it which is more accurate.

0 to 13.8V LM317 Power Supply

 

[ronsimpson]

Linear Technology - LT3080 - Adjustable 1.1A Single Resistor Low Dropout Regulator
This regulator is very simple. Does not require a negative supply. Operates to 0 volts! Can parallel 2 to get 2 amps. Will not do 40 volts!
LT3080

 

[crutschow]

The LM317 has an Absolute Maximum input-output differential of 40V so, for good reliability, you should have a maximum input voltage of a few volts less than 40V. You especially need to watch the voltage out of a diode capacitor rectifier DC source whose voltage will reach near the peak AC voltage under no load.

Also the minimum in-out differential is about 3V, so the maximum regulated output from the L317 is the input voltage minus 3V.

Thus realistically you shouldn't expect to get more than about 35V out of the LM317.

 

[crutschow]

I hope you're aware of the negative supply and that it won't be ab;e to supply 1A over the entire voltage range.

You misunderstand the circuit. The negative supply only has to provide bias current, not load current.

 

[MrAl]

Hi again,

There is also a 'cheapie' version of this circuit where you use two regular si
diodes in series to get the 1.2 (or so) voltage drop needed to be able to adjust
the output to zero volts. Dont waste money on the LM113 in any case, use
a cheaper reference diode if you go with that circuit.

There is also another two si diode circuit which does not need a negative
reference. With this circuit the power supply needs to be able to handle the
full load current (which is would anyway). The two diodes also need to be
able to handle the full rated current. If you need a schematic let me know.

The diode versions have a slight temperature drift.

 

[Hero999]

You misunderstand the circuit. The negative supply only has to provide bias current, not load current.

I know that but it still requires a negative supply, even if it's only 10mA. This means that if he wants to power it from a battery he'll need a charge pump to get -V or a voltage doubler on AC.

 

[crutschow]

I know that but it still requires a negative supply, even if it's only 10mA. This means that if he wants to power it from a battery he'll need a charge pump to get -V or a voltage doubler on AC.

Sorry. I realize I misunderstood your comment.

If you don't want to use a negative supply then ronsimpson's suggestion to use an LT3080 may be best, unless you really need 40V.