Low Voltage Adjustable Power Supply

Thread Starter

Jblack

Joined Feb 24, 2016
39
I need some help with a project I'm working on. I'm trying to test various Joule Thief circuits, namely what voltages they cut out at. I need some sort of power supply that I can adjust down to very low voltages. Something in the range of 0v-2v. Normally I would use a step down converter or a LM317, but they have minimum output voltages of 1v-2v (1.25 for the 317). I need something that can go lower. I've tried various circuits, both copied and of my own design, but I haven't been happy with any of them so far. One big problem I've been having is the voltage drop on the supply. For example, I'll set the supply to say 1.50v, but when I connect the Joule Thief, it'll go down to 1.20v. Some voltage drop is ok, but not that much. Anyone have any ideas? And no, I'm not going to buy a variable power supply, I don't have the money.
Thanks
~Jblack
 

jellytot

Joined May 20, 2014
72
If your output is 1v-2v, you can stick a resistor or diode to force a voltage drop before the load. But it will be unregulated, and you need to be sure your resistor or diode can handle the heat.
If you need regulated output, I'm pretty sure they make regulators that output less than 1 Volt (for example MCP1826S-0802E/AB-ND). You can also get adjustable output regulators that go below 1 Volt (I personally haven't used these though).
 

#12

Joined Nov 30, 2010
18,224
Using a diode or two to get rid of about 0.7 volts each would be a very easy way to do this.
You can also make a regulator out of a few transistors
 

ronv

Joined Nov 12, 2008
3,770
I need some help with a project I'm working on. I'm trying to test various Joule Thief circuits, namely what voltages they cut out at. I need some sort of power supply that I can adjust down to very low voltages. Something in the range of 0v-2v. Normally I would use a step down converter or a LM317, but they have minimum output voltages of 1v-2v (1.25 for the 317). I need something that can go lower. I've tried various circuits, both copied and of my own design, but I haven't been happy with any of them so far. One big problem I've been having is the voltage drop on the supply. For example, I'll set the supply to say 1.50v, but when I connect the Joule Thief, it'll go down to 1.20v. Some voltage drop is ok, but not that much. Anyone have any ideas? And no, I'm not going to buy a variable power supply, I don't have the money.
Thanks
~Jblack
What power supply are you starting with (voltage) and how much current do you need for the joule thief?
 

Thread Starter

Jblack

Joined Feb 24, 2016
39
If your output is 1v-2v, you can stick a resistor or diode to force a voltage drop before the load. But it will be unregulated, and you need to be sure your resistor or diode can handle the heat.
If you need regulated output, I'm pretty sure they make regulators that output less than 1 Volt (for example MCP1826S-0802E/AB-ND). You can also get adjustable output regulators that go below 1 Volt (I personally haven't used these though).
Thanks for the suggestions. I hadn't considered using a diode to drop the voltage, but I'm gonna play around with that idea now. I took a quick look at the regulator you suggested and I'll keep it in mind for future projects, but it only goes down to 0.8v so it wouldn't work for this one.
 

Thread Starter

Jblack

Joined Feb 24, 2016
39
Try the LT3080 voltage regulator chip; its output can be adjusted down to 0 volts. Digi-Key has them, price in single quantities is $4.40.
That looks just about perfect, thanks. I'm going to try to make a supply without using hard to get expensive chips, but I may end up using that if I can't make a good one.
 

Thread Starter

Jblack

Joined Feb 24, 2016
39
What power supply are you starting with (voltage) and how much current do you need for the joule thief?
Well, I have several I could use, I could start at 12v, 6v, 5v, or 3.3v, I could also use a basic LM317 adjustable supply. For the Joule Thief, I need at least 200mA of continuous, but preferably more than that for surges, like short circuits.
 

Thread Starter

Jblack

Joined Feb 24, 2016
39
Using a diode or two to get rid of about 0.7 volts each would be a very easy way to do this.
You can also make a regulator out of a few transistors
I will experiment with using diodes, hopefully that will work. I have made a few different regulators with transistors, but the problem has been the voltage drop when a load is across them. Some will drop 0.1-0.2v with only a 5-6mA load, others dropped as much as .6v with the same load. If you know of any good regulator circuits with transistor, could you post a schematic or link to one?
 

AnalogKid

Joined Aug 1, 2013
10,987
With a 6 V or 12 V source, I'd use an LM4040-2.5 for the reference, a pot across it for adjustment, and an LM358 plus a power darlington as a high-current voltage follower for the output. The LM358 has an input common mode range that extends below its negative rail (GND in your case), so the circuit should regulate down to 0.1 V or below. The power darlington has built-in resistors across the two base-emitter junctions that will act as a minimum load to maintain regulation.

EDIT: I whipped this out on the fly while trying to pay attention to something else. The power darlington internal resistors do *not* pre-load the regulator. oops. See post #14.

ak
 
Last edited:

Thread Starter

Jblack

Joined Feb 24, 2016
39
With a 6 V or 12 V source, I'd use an LM4040-2.5 for the reference, a pot across it for adjustment, and an LM358 plus a power darlington as a high-current voltage follower for the output. The LM358 has an input common mode range that extends below its negative rail (GND in your case), so the circuit should regulate down to 0.1 V or below. The power darlington has built-in resistors across the two base-emitter junctions that will act as a minimum load to maintain regulation.

ak
Thanks for the suggestion, it sounds like it would work, but I don't know exactly what that would look like. Would you happen to have a schematic of some sort?
 

ronv

Joined Nov 12, 2008
3,770
Thanks for the suggestion, it sounds like it would work, but I don't know exactly what that would look like. Would you happen to have a schematic of some sort?
Or since you have a lot of power supplies you could use one and a couple of diodes to get down to zero with your 317.
upload_2016-6-13_19-22-43.png
 

crutschow

Joined Mar 14, 2008
34,285
Below is the LTspice simulation of a power supply circuit similar to that suggested by AK.
The output is adjustable from 0V to 3V for the values shown as pot U3 is varied from 0% to 100% (0 to 1 for wiper position w).
The TL431a is a commonly available adjustable 2.5V reference (shown adjusted to 3V by R4 and R5).
The current is limited to about 300mA maximum by by Q3 to make the circuit tolerant of short duration shorts.
The test load Rload is shown being varied from 0.1 to 10Ω.
The power dissipated in Q2 is a maximum 1.2W at 200mA output so no heatsink is needed for the 2N3055 as long as the steady-state current is no higher than 200mA or so.


upload_2016-6-13_23-24-19.png
 

Attachments

AnalogKid

Joined Aug 1, 2013
10,987
Here you go. A variation of this has worked for me in the past. Depending on the characteristics of the load circuits, you might need the increased stability provided by Wally's circuit.
Note - U1 current is too high. Increase R1 to something from 4.7K to 10K.

ak
Low-Volt-Regulator-1-c.gif
 

Attachments

Thread Starter

Jblack

Joined Feb 24, 2016
39
Below is the LTspice simulation of a power supply circuit similar to that suggested by AK.
The output is adjustable from 0V to 3V for the values shown as pot U3 is varied from 0% to 100% (0 to 1 for wiper position w).
The TL431a is a commonly available adjustable 2.5V reference (shown adjusted to 3V by R4 and R5).
The current is limited to about 300mA maximum by by Q3 to make the circuit tolerant of short duration shorts.
The test load Rload is shown being varied from 0.1 to 10Ω.
The power dissipated in Q2 is a maximum 1.2W at 200mA output so no heatsink is needed for the 2N3055 as long as the steady-state current is no higher than 200mA or so.


View attachment 107682
Thank you very much! I believe I have all the parts laying around somewhere, so once I get them I'll put that circuit together and test it out. I had completely forgotten about using an op-amp and a transistor as voltage regulator, I think it'll work.
 

Thread Starter

Jblack

Joined Feb 24, 2016
39
Here you go. A variation of this has worked for me in the past. Depending on the characteristics of the load circuits, you might need the increased stability provided by Wally's circuit.
Note - U1 current is too high. Increase R1 to something from 4.7K to 10K.

ak
View attachment 107701
Thank you for the schematic! I think I'll go with the circuit that crutschow provided, simply because I don't have a darlington transistor at the moment, and it seems like it may be a bit more stable at higher currents.
 

crutschow

Joined Mar 14, 2008
34,285
..................Depending on the characteristics of the load circuits, you might need the increased stability provided by Wally's circuit.............
Yes, compensation cap C2 appears to be needed if there's any significant filtering capacitance on the output.

Edit: I also found that, in the current-limit mode, the simulation shows the limiter going into a kind of "hiccup" oscillation, with short current spikes occurring at about a 7KHz rate.
This has the advantage of reducing the average Q2's power consumption to a few hundred milliwatts under short-circuit conditions, much less than the 1.8W a steady 300mA current limit would generate.
 
Last edited:
Top