In this instructable I am going to show you how to make a bench power supply. The bench power supply would consist of 6 fixed voltages and a variable voltage from 7v to 1v

Step 1: What Is a Bench Power Supply and Why Do I Need It ?

A bench power supply is a very important and useful tool in electronics. You may not understand the importance of it until you make one or use one. The same was my case as I found it useless until I made one and millions of new possibilities opened up.

Basically a bench power supply is a tool through which you can get multiple voltages. So the bench power supply that we will make today will have 12v,10v,9v,8v,6v,5v and a variable supply ranging from 7v to 1v.


A wide range of materials would be required for making this project and you should spread the work in one day to get better results.
So without further ado the materials are
•Enclosure for the projects
• Power supply(as I didn't want to do AC to DC conversation I simply used a 12v wall vort)

•LM 317
• 5k ohm pot
• 220 ohm resistor
• 0.1uf cap
• 1 uf cap
• banana clips male
• banana clips female
• switch

•voltmeter( buy voltmeter from ebay )

The tools required for the build are
• Soldering iron
•Dremel or cutting tool
• A day or 2
you can get all the materials in about 5$ to 10$ depending on the place you live in.


You can follow the schematics given above and make the circuit . If you are not well versed with reading schematics check this out reading schematics

After you have the circuit on the purf board you can test it with a multimeter. DO NOT TEST IT WITH THE VOLTMETER as the max voltage on the voltmeter is 9 v and u will have tho test it through the 7808 the 8v regulator.

We will be doing the same in the power supply as to get 7 volts max.


For making the fixed voltage outputs, solder all the voltage regulators directly on the board without putting them through as the thickness of the pins is very large. then give the 12v to the input pin and ground to the ground to all the voltage regulators. I put the 7808 or the 8v regulator directly into the LM317 circuit.

The outputs of these regulator would be used later to connect them to the banana clips.


For preparing the enclosure first put white tape all over the enclosure. then make the required markings and cut them using the dremel or the cutting tool. You may have to use the sanding tip also to clean the edges and give a good fit.

the enclosure is the major part and should be done with the most carefulness.

The holes for the banana clips can be drilled with the appropriate drill bit and can be screwed on well. Don't apply force on the plastic as it may break as it is very brittle.


After the banana clips are fixed in the enclosure put in the 2 pcbs and wire the connections. The first clip is the variable clip and then followed by 12v,10,9v,8v,6v and 5v.

The voltmeter has three wires - one white,one red and one black. The black one is for the ground the,red is for the vcc or the positive terminal which can directly be connected to the 12v from the adapter. The white line goes to the output of the LM317 circuit.

A black banana clip is connected to the ground.


The last step is to close the enclosure. If you have access to a multimeter then test all the connections before switching on and also check the output from each clip.

Isolate all the electronics and put the screws in the holes and close the box.


<p>Several people have commented regarding possible over-current problems. These 3-terminal regulators have internal current limiting that protects them from burnout, but they do need a heat sink if operated near full power.</p><p>Using a wall-wart to power the unit usually will limit currents to safe values because Class-2 devices have inherent current limiting in the transformer...but not always.</p><p>A string of solar cells and battery to power this might be interesting for portability,</p>
<p>Thanks a lot</p><p>it ends my problem to explain to all as you have done it so well</p>
<p>Except that's complete nonsense. It is extremely easy to overheat one without exceeding the current limit - a limit that depends upon it staying under the temperatures specified in the data sheet.</p><p>Understand this clearly! At only 500mA current you aren't anywhere near full current rating but can't drop voltage by much at all because V * A = watts heat. They have a tab on the back for a reason - to put a heatsink on it.</p>
<p>Please please do more research before building something like this. The way this is built is incredibly prone to overheating. And even if it doesn't overheat, a 12 V wall wart doesn't give a constant 12 V out. If it is rated for 1 Amp then it will give about 12 V at 1 Amp, but more at lower current draw and less at higher current draw. Your 12 V linear regulator requires an input at least 1.25 V higher, so 13.25 V. Which means you should make sure you are using a wall wart rated at 14 V at or above whatever max current you plan to draw (which without heatsinks had better not be very much). Also remember that these are not isolated power rails, so you need to have one rated at 14 V at or above the TOTAL current draw you might use if all rails are connected to a load.</p>
Dude. It's really good.
thanks really INVITING right?<br>plan to make it
lm317 gets very hot so you should connect a heatsink to it
had it already just didn't show it
<p>You'll get a cheap heatsink at alpha man</p>
<p>ive built one close to this one using a pc power supply but instead of an LM317 i used a buck down converter and swapped the small trim pot to a bigger one of the same value</p>
you do not heat the regulators? sink occupy no? I once did a regulator and warmed me
<p>Awesome build! It looks like I have a new item on my to-do list. One question: did I miss the part where you put in a fuse, or is that a &quot;Version 2&quot; thing?</p>
<p>You dont need a fuse as you are using a wall adapter and not doing the AC to DC conversion but if you want to try that it is all up to you.</p><p>also i have tried connecting the positive and negative together many times sometimes deliberately also and the result was that the adapter switched off.no damage. fuse is not really required if you are using a adapter.</p>
<p>You probably have a switching AC-DC adapter then, since it shuts off but then works again. Keep in mind that it puts stress on it every time you do that. Having it save itself from damage a few times is not proof it will keep doing that without the stress of it frying a component.</p><p>However I do agree that you don't need a fuse for this, and in fact there may be a thermal fuse inside the ac-dc adapter which has never tripped.</p>
<p>Good beginning, but might not allow for much current draw without burning out your transistors. Have you considered making a PWM based unit with a 555 timer and larger transistors to handle more current?</p>
Surely Michael I agree to you as the max current output is only 1.5 amps on the lm317. we can make a 555 pwm but I think there would be many other solutions also that I would like to check out. Thanks for your input and to tell you I plan to make another power supply which would be more efficient maybe a boost power supply. <br>But anyways thanks a lot
<p>No! Your max current is significantly lower. Because the regulators do not have heatsinks they are limited to about 1W maximum loss (as heat) for long life. This is the inherent drawback of linear regulators, that all voltage is lost as heat.</p><p>If your AC-DC adapter is regulated so it is actually 12V instead of a higher float voltage around 15V (which would make matters even worse) then the math is as follows:</p><p>7805: 12V - 5V = 7V drop, 1W / 7V = 143mA max<br>7806: 12V - 6V = 6V drop, 1W / 6V = 167mA max<br>7808: 12V - 4V = 4V drop, 1W / 4V = 250mA max<br>7809: 12V - 9V = 3V drop, 1W / 3V = 333mA max<br>7810: 12.V - 10V = 2V drop, 1W / 2V = 500mA max<br>LM 317: Depends on its setting, 12V - (n)V = ?, 1W / ? = </p><p>Using heatsinks would significantly increase these values but then at some point for continuous operation you would also want to put vents, or perhaps even active cooling in the project box.</p><p>Frankly if you do not need multiple outputs simultaneously I would get rid of all the regulators except the LM317 and put a larger heatsink on it, or use two LM317 for dual variable outputs, then either a 2nd LED voltage display or a toggle switch to change which output the display is showing for adjustment purposes.</p>
this is a bench power supply built with only the basic materials. <br>the idea of two lm317 would not work as firstly because it would be very difficult to tune to the right voltage and secondly because having preset voltages help on the go
<p>Two LM317 would work fine. You simply substitute a potentiometer for one of the resistors in the feedback look on each, then your output voltage meter could switch between them to set the voltage for each.</p><p>If you do not wish to do it that way, certainly it will still work the way you prefer, but to keep them cool enough it means more regulators mounted to heatsinks, a higher part count when you probably won't need to use more than two different voltages simultaneously, or at least not often.</p><p>In fact I do think that at a minimum you should keep an LM7805 because so very many things in electronics were set for 5V input, although these days everyone is pushing for lower voltages to suit the mobile crowd that likes to use 3.7V nominal Li-Ion battery packs, and yet, LM317 set to 5.0V has cleaner output than LM7805 though this is largely due to the capacitor in the feedback loop and you could accomplish the same adding equivalent to LM7805 instead of sticking with the intentionally low part count datasheet example(s), &quot;IF&quot; you really need the cleanest output possible which is the usual reason people build linear regulated supplies these days.</p>
<p>Excellent summary - totally proves this project cannot really be used as a bench supply. Sure you could add chunky heat sinks to get over the 1W but overall this is more expensive than using a switch mode power supply and why computer power <br>supplies and many supplies are SMPS.</p><p>You would also want some <br>input\output caps on the other regulators too as per their <br>data-sheets.</p>
<p>Well heatsinks aren't all that expensive if you consider that you can cannibalize them out of old CRT TV sets purposefully made for TO-220 sized (as shown in the pics) components, and for certain circuits, particularly amplification or audio buffers it's a very good idea to stay with linear regulators to reduce the signal deviation or error rate, but in this case his project could relatively easily be upgraded to work safely to at least 3W per regulator which doesn't seem like much but would make it a lot more useful, then of course the things that require high current generally don't need it as clean (except in pro audio) so a switching supply would do fine for that.</p>
<p>With an output current of 1.5A with 28V input and 5V output gives 34.5 Watts dissipated in the LM317. That's not possible, at least not for more than a few seconds.</p><p>That is why linear regulators are a bad idea for wide range supplies.</p><p>Also:</p><p>I don't know why you would want to make a pwm supply based on a 555. There are lots of good pwm controller chips out there. All you have to do is read the data sheet and solve the equations for each of the design parameters. It isn't hard to do and you will have a far superior power supply.</p><p>If I were doing this, I would use the small pwm supplies available on ebay and replace the trim pot with a panel mounted pot. You get the advantage of high efficiency, compact design, and stable operation. If you get one with constant Voltage and constant current capability, you can also set the current limit to whatever you feel safe with.</p>
<p>Very nice build, Paurush. I love instructables like this. Very useful, practical, inexpensive, and instructional. </p>
<p>thanks a lot it helps me promote myself to do more</p>
<p>This is very nice project. You might take advantage of break throught limited current like LM317 by check it out on Digikey.com. They're expensive but they're the better. You are look at this <a href="http://www.digikey.com/product-search/en/integrated-circuits-ics/pmic-voltage-regulators-linear-ldo/2556290?k=Voltage" rel="nofollow">PMIC - Voltage Regulators - Linear (LDO)</a> that I'm talk about. <br><br>Or you could building DC to DC with switching mode power supply. </p>
thanks for the advice <br>but the problem here in India is that you are not able to exploit the services like digikey. the take a lot of time to deliver thus I resort to make my own<br>
<p>I really like the simplicity of this bench power supply. I can't wait to try it out. I'm watching out for version 2 ;).</p><p>I like the suggestions down below for version 2... Having the capacity for more current is a big one for me.</p>
Thanks for posting this. I need one in my shop.
<p>surely do share it with me </p><p>also ask me any queries</p>
<p>Hmmm directions are pretty sketchy. You show lots of detail on the soldering of the various voltage regulators, drilling the enclosure but little on how the electronics are connect to the banana clips, the display, the switch, pot, resistor, etc etc... So I can understand why some of your viewers can't follow you directions. Could you provide a little more detail?</p>
<p>Surely I will work on it </p><p>But if you need something specific i will send it to you</p>
<p>oh man that was cool i tried to make it ; but mine did not work</p>
why that's sad<br>tell me the problem I may be able to find a solution post the problem here or mail it to me
...and yours looks a whole lot nicer than mine!
another nifty trick <br>I wasn't able to fit the adapter through the hole in my desk so I cut it and put 3 female headers and 3 male headers in each end and put ground in the middle and vcc on the opposite ends<br>makes it very easy to put in and to carry
thanks it makes it very easy to carry around
Excellent build and write-up! I made a small one and stuck it to the top of my breadboard. One thing I included that you might find useful is a 555 square-wave generator. I used an 6-position rotary switch with various capacitors to be able to set a clock signal from 1 Hz to 100 kHz in powers of 10. Very useful when working with ICs that need to be clocked.
thanks man cool will surely try it in version 2. You people are all the more promoting me to make version 2.<br>keep an eye on my page for it and thanks a lot again
One can find a LM2596 adjustable PWM dc - to - dc converter module with 3A output on eBay for as little as $0.99.
thanks Peabody for the information<br>will help a lot in bench power supply version 2

About This Instructable




Add instructable to: