Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Yet another basic LM317 PSU

Status
Not open for further replies.

Blueteeth

Well-Known Member
Hi,

Since I've been doing lots of work with DIYing switch mode power supplies and portable applications, I needed to measure current quite often, in ranges from up to 5A, down to <1mA (standby current for uC). Getting my multimeter in series with the power supplies is a hassle, plus, at low currents, the voltage drop across the multimeter becomes unacceptable...so I need another one to measure the 'actual' power supply voltage my device is seeing.

So, I just spent 20 minutes with eagle coming up with a very simple and cheap little 1A power suppliy with selectable voltages. There are LOTS of these schems on the internet I know, but I haven't seen one with in built current measurement - which converts the current draw to a 0v referenced voltage, with selectable ranges, handy to plug a multimeter in in its DC volts range.

It measures current on the high side with a fairly large current sense resistor (a whopping 0.1ohms as oppoosed to the mili-ohm range) and a very cheap LM358. Low side current measurement is easier, but makes voltage regulation a bit harder. Of course, one could add a microcontroller, some PWM, precision reference etc.. but then it starts to get messy, and the LM317's regulation is pretty good on its own.

Designed to take DC input from your average wall-wart power supply, or anything you have to hand...I have many laptop PSU's, ATX psu's...etc.. so no point in reinventing the wheel by including a transformer and rectification.

Please note: not all the parts have values, as I couldn't be bothered to calculate them for the voltages. Thats the easy part and comes later.

Criticisms welcome!
 

Attachments

  • LM317test.png
    LM317test.png
    15.7 KB · Views: 1,291
Couldn't you do this a lot more simply by just using a precision shunt resistor (0.1 ohm or whatever) and putting 2 additonal external connections to the shunt resistor so you could connect an external multimeter (voltmeter)?
 
Thats exactly what I was originally going to do :) I added the opamp because I was planning on adding a microcontroller to it, and the ADC requires a 0v referenced voltage (thus the opamp voltage follower as a buffer). Although of course I omitted it, it does form a uilding block, allowing any ADC, or a comparator to be used.

Also wanted to add an opamp in there to add some gain...even though my multimeter is autoranging, unless I use a 1 ohm resistor (quite possible) I would have to do multiply the voltage by 10 all the time to get the current. Although its simple, its a great way to screw up readings at the end of a day at work, when you're not as sharp as you were.

Granted, it pretty much doubles the components but does make life easier in the long run..at least I think so :D
 
Hi Blueteeth,

Eagle has a very nice function called "smash". :)

If you smash the parts, use a finer grid size and move part names and values to be readable the schematic gets more distinctive. Looks cleaner, too. :D

Regards

Boncuk
 
I think you'll be disappointed at the voltage stability, especially at higher output voltage settings. R1's loss won't be compensated by the LM317, rather it will be amplified by the ratio of (R2+R5)/R2.

R1 is best placed before U$1, and subtract out about ~5mA which is consumed by R2 and U$1.(You would need a different op amp.) Second best is to place R1 after the connection of U$1 output and R2.

You'll also be very disappointed by the errors in current monitoring. The sense resistor of 0.1Ω is 100µV/1mA. LM358's typical offset voltage is 2mV, meaning that with no load the current monitor can show +/-20mA. The 1V/1mA and 1V/10mA settings will probably be off the scale at all times.
 
You must never use a set of plugs to change the voltage on the adj pin of the 3-terminal regulator, as per your circuit diagram.
If the supply is connected and the plug is removed, the output voltage will rise to nearly the input voltage and damage the device being powered.
 
You must never use a set of plugs to change the voltage on the adj pin of the 3-terminal regulator, as per your circuit diagram.
If the supply is connected and the plug is removed, the output voltage will rise to nearly the input voltage and damage the device being powered.

This is easily corrected. Just place the smallest resistor value in series with the plug/switch block such that when all resistors are removed the voltage associated with the smallest resistor will be selected. You will now have 5 positions instead of 6.

The other 5 resistor values will need to be adjusted so the actual value is equal to the desired value - the fixed series resistor
 
Last edited:
This is easily corrected. Just place the smallest resistor value in series with the plug/switch block such that when all resistors are removed the voltage associated with the smallest resistor will be selected.

All the values have to be re-calculated.
 
Last edited:
You must never use a set of plugs to change the voltage on the adj pin of the 3-terminal regulator, as per your circuit diagram.
If the supply is connected and the plug is removed, the output voltage will rise to nearly the input voltage and damage the device being powered.

Yeah, got that covered, I'm using a 'make before break' rotary switch :D That, along with a 10uF cap on the adj pin to 0v should stop the output voltage shooting up when selecting voltages. I actually changed the schem slightly, having resistors on the rotary switch in series. Good call though! if I hadn't realised that myself you would have saved a few PIC's from meeting their maker.

As for recalculating resistor values, thats the easy part. I have an excel spread sheet which picks out E12/E24 resistor values, with a maximum of 3 in series (or two in parallel) for setting the output voltage.
 
I think you changed the schematic after you were warned, as is seems improbable that you twigged to the fact that the output rises when the adjust pin floats – as your circuit diagram gave no indication of this.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top