very impressed with your write up even included all hyper links to supplys much appreciated the time you put into this project. top notch work my friend
While creating a corresponding blog article (and/or Github repo when there is code involved) takes a ton of additional time, I like to provide what I feel is everything needed for someone to duplicate or adapt one of my projects. I always got frustrated when I was first starting out (and didn't know a lot) and would watch a "how-to" video that omitted a big portion of the actual "how-to".... or required me to try to freeze frame the video to see something like a wiring diagram. Thanks so much for the kind words. It is comments like yours that encourage me to continue to invest time in things like related blog articles and wikis.
Thanks. There are a lot of different versions and designs on RUclips. I build this one for my particular needs... I wanted the fixed 3.3V, 5V and 12V... but also wanted an option for a variable voltage, as sometimes I want 9V or 7.4V, etc. This does make the design a bit bigger but does give both options... and option for multiple simultaneous outputs. Thanks for watching and taking the time to leave a comment!
This is great. I was actually watching a couple videos last week about making something similar but you did the best job explaining all the parts and how to do it step by step. I always enjoy your LED videos as well. You’re a very good teacher. Also, now I don’t feel so bad for having a bunch of old computers lying around. I’m not alone lol. I also always wondered why the power supplies had so many different plugs coming out of them. I just never paid attention to that part of the PC build. As long as the plugs fit I was happy lol.
Thanks so much for the kind words. While it does make my videos longer than most, I do try to always explain the "why" as well as the "what" so that newer folks might be able to take the concepts and adapt them to their own projects. I'm always trying to come up with ideas on how to recycle and extend the life of my old electronics. I have a few other ideas and hope to implement them if I can ever find the time. Thanks again for the comment!
Great work. I have been planing to do one of these but most of the things I could find didn't give as thorough an explination, thanks. I now have a better understanding of what is going on in the power supply that i should be able to figure it out.
You are welcome! I'm not sure that a 30 minutes video would be considered "concise" by RUclips standards, but I do try to include everything someone would need to know to create their own version of a project. Thanks for watching and taking a moment to post a comment. I appreciate it!
hi, you create very useful and detailed content, I really like your videos, they teach a practical approach, not just bragging about what I have. Thanks!
Thanks! While it does make my videos a bit longer, I always try to explain 'why' I'm doing something as opposed to just "how". I do this so that hopefully viewers can take my ideas or projects and adapt them to their own projects or needs. Thanks for watching... and taking the time to post your kind words!
Excellent video, thank you for the time taken to share it with us. I built one very similar 'cept I used an old enclosure from a worn out VCR made in the '90s. Just had to make a new front panel.
That's great! And you found new life for even another old piece of electronics in the process. I probably hang on to too many old devices, but I do really like when I can turn one of these old devices into something that I can use today. Thanks for watching and taking a few minutes to post a comment!
Thats awesome. I just need a variable supply, but now that I've seen this it makes me feel like I'd be wasting the other rails for fixed voltages if I don't use them. I also have a 275w Dell psu with the two 12v rails. Hmm.
Honestly, I use the fixed rails much more often than the variable side. But most of my projects are ESP based so pretty much everything runs off of 5V or 3.3V. Occasionally, I'll have something that needs 12V. About the only time I use variable is if I need something that represents a battery... like 9V or 7.4V. But the majority of my projects use one of the three fixed voltages. But the nice thing I like is that I have the flexibility with this type of supply... including the ability to provide up to 3 different voltages simultaneously on the very rare case it is needed. But it could be made smaller if desired by just creating a variable output.
@@ResinChemTech I'm looking at using one of these "LTC3780" buck/boost converters I seen in another video. I really only need current limiting, so I can track down short circuits on other boards with voltage injection.
Don't those fuse holders only have a rating of 10A ? 🤔 I made similar bench power supply but i just used inline spade fuse holders they are used in Automotive industry for bigger Amps than i will ever use but just using 20A on mine for 2 separate 12v feeds on variable outputs then 3.3v 5v separately all on individual voltmeters. Great video 👍
Brilliant video. I wanted to add 50 amp 75 mV shunts to the meters. Would you happen to have a shunt hook up schematic that relates to this particular project. Thank You.
Hello. This instruments need shunt for more then 10 amps. I build a similar project with boost and buck on the variable side. The voltage works great but the amps is not so good. Could be the shunts?
Fantastic video, thank you for the time taken to create a detailed explanation. I have been wanting to buy one of these, but being a fan of DIY, I think I will rather build my own as well.
Thanks! This is one of those projects that is a good DIY project that doesn't require a lot of soldering or 'advanced' electronic skills. However, unlike some of my others, it isn't necessarily a cost-savings one when compared to retail versions. But there's the DIY satisfaction plus the recycling of old components into something new. Thanks for watching and taking time to comment! Good luck if you decide to build your own... and let me know if you have any questions. Definitely peek at the blog article as there are some things in there that I didn't cover in the video just due to time/length of the video.
I am a new subscriber. I have watched a majority of your LED videos and enjoyed them all, however, this video is what made me finally click the subscribe button. The longer format is excellent for me because it allows enough time to enjoy and understand your work.
Thanks! I know that my videos are simply too long for a lot of people. But I know that when I was first learning, I'd get frustrated at times when watching other project videos because so much of the detail was skipped that I couldn't really understand what was being shown... nor could I duplicate it without doing a lot of additional research. I try to show not only the 'how'... but also include the 'why', so that viewers that are a bit newer in their journey can understand the concepts and maybe adapt them to their own projects. Lately, I've been trying to do a quick overview of the final project first (for those that don't care about the details), then follow that up with the longer step-by-step process for those that want to hang around a bit longer. Thanks for watching... and for subscribing! Welcome to the channel!
Super appreciate the in-depth explanation of your video(s). I was curious to know the gauge of wire you used to wire in the 2 different 100k pots? Will the gauge thickness alter the potential of the potentiometer?
I believed that I used 18 gauge, but I don't recall absolutely for sure. These are very short runs, and if any voltage drop is truly present, it really doesn't matter since the actual voltage being output to the terminals is shown on the meter, which is measuring voltage after the potentiometer and any other wiring. The only place that this might matter would be at the very top end (e.g. 12V). Say there really was a 0.1V drop due to wire resistance. The max output of the meter (and the true voltage) would only be 11.9V. I don't think this is really the case... and even it if was, I'd just use the fixed 12V side anyway. The net result is that the potentiometer is used to adjust the desired voltage, which is measured AFTER any resistance that would be introduced by any wiring. I hope I explained so that it makes some sense! The key thing is to just select a potentiometer with a resistance that isn't too sensitive and yet allows you to cover the range from around 1V up to 12V. If wire gauge impacts that, then a different potentiometer can be used.
Hello. Me again. I got my power supply put together and upon testing it, the fixed side works perfectly :) Not so much on the variable side :( I get 12V out of the jacks and the meter and my DVOM show that, but the variable part is not working. I tested my potentiometers and they check out good. But when I turn the knobs when they are hooked up in circuit...nothing. I had some trouble getting the little blue potentiometers off of the buck converter and also in attaching the red, white, and black wires. My guess is that I screwed up something...too much heat and I cooked some of those SMD capacitors and/or resistors on the bottom of the PCB? When I test the set of three connections on the PCB that feed the CV pot and the three from the CC side I get nothing. My uneducated guess that I cooked something and that I need a new buck converter. Any advice would be greatly appreciated. Thanks for your time
Removing those tiny pots and soldering wire on... while assuring none of those wires touched each other... was probably one of the trickiest parts of the entire build. After getting the wires soldered on for the first connection, I used liberal amounts of liquid tape to assure that I didn't inadvertently create a connection when soldering on the second set. I can only recommend what I did. First, I tested the buck converter fully with the existing pots... feeding in 12V and using a multimeter on the output side to measure voltage as I turned the tiny set screw. I then de-soldered the existing pots and connected just the first set of wires (voltage adjustment I believe) and repeated my tests. This is also where I tried different rotary pots to select a value with a decent range vs. precision. When all that was done, then I threw on the liquid tape, soldered the second set and tested again. I don't know if any of that is helpful... but I do know this particular step can be a bit difficult.
@@ResinChemTech Thanks for getting back to me. So, based on my description above, do you think I messed something up in the buck converter and I should get a new one?
It's hard to say for sure whether the buck converter has been damaged or if it is a wiring/soldering issue. But if you aren't getting any measurable voltage out, it's obviously not going to work. Any possibility that you can try de-soldering the wires you added and then try soldering them back again? If you don't have a good connection where the original pots used to be, I'm guessing you'd have an open circuit and not be getting any output. I guess I'd try to troubleshoot and assure the current buck converter is damaged or not salvable before buying a new one. But it's up to you to decide whether the effort to try to fix the first one is worth the cost of just buying a new one.
@22:43 I tried something similar with exact same model of CC CV regulator but the external 100k pot seem to have damaged my unit ! After discussing this matter on several electronic forums it was concluded that due to their sensitivity inherited from manufacturing process, you cannot simply replace an onboard Trim Pot with external potentiometer as the added capacitance / inductance can destroy the DC DC regulator.. Did you experience something like this at all ?
I haven't experienced any issues like you describe. I've been using the power supply for well over a year... often using the variable voltage... and the system has been holding up fine. I will admit that I fried a number of pots initially as I tried finding the correct ones to use for the best range vs. precision, but that was due to a miscalculation on my part with the estimated current. I don't doubt what others are saying... I am by no means an electrical engineer or even anything close to an expert when it comes to things like capacitance and inductance. But it has been working just fine for me... at least on the voltage side. I rarely limit the current mostly because the types of projects that I work on don't need to work within a limited current range. I probably wouldn't even add the current limiter if I were to do it over again. Maybe I've just been lucky... but so far I haven't experienced any issues that damaged any of the components. For my variable range, I'm generally using it for voltages between the 5 and 12V range that I can't get out of the fixed voltage outputs.
Have you thought about adding stackable banana plugs for your meter so you don't have to unscrew it all the way You can just unplug it and stack it on to another connection?
Since I made this video, I have created a bunch of different jumpers and connectors, including banana plug jumpers similar to what you mentioned that lets me quickly switch the voltage... although most of the time it either gets used for 5V on the fixed side or the variable side gets used for "off" voltages for things like 7.4V or 9V. Thanks for the suggestion!
Hi great presentation of this technical project. I thought I’d give it a try. Well at this point I get the stand by light to go on, then turn the switch on and the power on light (gray wire) will not turn on. I have green to switch then to black. The purple is connected to the stand by led and lights up so led works. I’m using 120 ohm ¼ w resistors soldered to cathode side then solder to black (-) buss. I tested each led circuit and they were ok. When I turn the switch on the power on does not light up (gray) fan runs for like 2 secs. Where do I go from here? All other wiring is as it should be. Again great presentation, I love a challenge. thanks
First off, have you verified the wiring color for your particular power supply? I was using a Dell power supply and they were notorious for using non-standard/proprietary components. So it is possible that the wiring on yours is slightly different, especially for the secondary wiring. But more importantly, did you provide a load? You mentioned a resistor for the LEDs, but most power supplies will not power on (or immediately power off) if a load is not detected. I had to add a 10W 10 ohm resistor to the 5V line. If you haven't looked at my blog article, take a peek at that and the section under "providing a load": resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html Those are my initial thoughts. But if it runs for like 2 seconds and shuts down, it is likely due to a safety feature... one of which is not detecting a load (a short/over-voltage, etc. are other reasons). But be sure you are supplying a load... that sounds like the most likely cause here.
hi thanks for the reply. i went back and watched the video and reread the step0by-step you put together and i realized i did not use the brown wire (sensing wire). i saw it in one of the diagrams in the writeup but it was not labeled. you mention this wire 3 times and it's usage is explained but you have to dig for it, i like a challange. so i connected the brown to 3.3v (orange) wire and it worked! i then swred it up and no go, something happened. i had another psu so i broke it down, wired it up and it didn't work either. i think both psu's have other issues, so i will get another psu and try is again. i also had the load suing 5v (red) w/10ohm10w resistor. so back to the challenge for me...
Hello. Please excuse my ignorance but if the max input is 12vdc, is the max output 12dc as well? Can I use a different converter to get say 36vdc at like 10 amps? Thanks for your time.
You could use something like a step-up boost converter to shift up the voltage. But you'd also need to be sure you don't exceed the power ratings for the power supply itself (36V x 10A = 360 watts) and that the boost converter (and all wiring/connections) were rated to handle the 10 A current.
Hello, Im from Brazil and I´m using this same Dell Power Supply in my project, but I´m trying to get 24 voltage using the Blue wire (-12v) and +12voltage yellow or white .... But I don´t know what is happening, when I connect this some times I got a burn in the Diode of -12V wire... I can use one Module XL4005 and only regulate the C. voltage and not CC ?
I can't say for sure and I'm certainly not an electrical engineer, but my gut tells me that even though there is a +12V and -12V available, the power supply likely was not designed to truly produce 24V. I don't know of any components on the original server that would have required 24V.... although there may have been both some +12V and -12V components. If you truly need 24V, the only reliable option may be to integrate a 24V boost converter in your build, something similar to this: amzn.to/4etwxso. You could even replace the buck converter with this, but you would need to feed in a lower voltage (3.3 or 5V) if you want the lower variable voltages, as this is only a boost and not a buck converter. It also has a lower max amp rating and would not offer the constant current option. Or you could add this in addition to the buck converter and create a fixed 24V output as well. There are numerous options, but it sounds like using +12V and -12V to get 24V directly may not work with only the power supply itself.
@@ResinChemTechYes, I agree whit you... I believe the buck send elevated CC and its damage The diode or other components... probably I will use only 12v regulated tensions like your project... but this trhouble is very strange for me, cause the most atx are prpjected to do 24v, using - 12v negative for this... wellz thx for help and congratulations for this channel
very helpful video and really well explained, I'm building one of these by myself too but I can't find any information about the resistive load. All I know is that it is needed if the voltages are not stable at the max ampere rail. In my case is the 5v 30A, this one is stable but the 12v and 3,3v are not, do I need to put a load one both of those rails? or being the higher amp rail stable is enough? sorry for my bad english.
You probably need to find and review the documentation for your particular power supply as it may be different than mine. For my Dell power supply, it required a load on the 5V line. In my case, a 10W 10 Ω resistor placed on a 5V lead provided enough resistance. Also, assure if you have a sense wire, it is connected to the proper voltage as it will sense the power output from the supply and make adjustments to keep the voltage stable. In my case, it was a 3V3 sense wire. If this isn't connected, it could result in the voltage being constantly adjusted, leading to what you are seeing. But you need to check the documentation for your particular power supply as it may be different from mine.
Thank you so much for posting this. I'm building this now. One question. When I do the calculations for the resistor size to use on the led indicator lights on a 12V power supply, I come up with I get around 400-800 ohms (using the ave. value for and led). In your plans it shows resistance values of 120 and 140 ohms. Am I missing something? Thanks
The LEDs are being powered by the PS5AUX and PS_PWRGOOD lines, which are 5V and not 12V. So the resistors selected are based on 5V. But they are really bright! If I had it to do over again, I might select larger resistor values to dim those LEDs down just a bit.
Not as designed. The only voltages natively available from the PC power supply are 12V, 5V and 3.3V. For the variable side, I use a step-down buck converter that can take the maximum 12V feed and step it down anywhere from the source 12V down to 1.2V. To get higher voltages, you'd need to supply a step-up or boost converter. In doing so, you'd have to be careful not to exceed the maximum wattage for the PC supply and also assure you do not exceed expected current ratings.
There are special plugs people use to test ATX PSUs without load. I am quite certain they do not have any heating issues. So perhaps there's a better way to keep the PSU on - the heat inside the enclosure is hardly critical, but it kinda sets off the perfectionism alarm
Hi, congratulations for the video. I would like to try to imitate your project. I would like to ask you if it is possible to have the schematic of the project. Thanks and good job.🛩
You can find the complete parts list, wiring diagrams and step-by-step directions in the related blog article to this video: resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html
The fuse size depends upon both the voltage and the specifications/wattage of your particular power supply. You fuse it so that you do not exceed the max power for a given voltage. I cover the particular fuses that I used for my power supply in the written version of this project: resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html
Not sure what you mean by the feeding wire, but you can use a metal front plate instead of a 3D printed faceplate. Just be sure that everything is properly insulated and the metal front panel does not become energized by any of the circuits.
Nice one! Other fixed voltages are also possible... The -12V (also -5V on even older PSUs) can be used in combination with the other voltage rails to produce a variety of fixed voltages. For example, if the -12 is made available with its own socket, one could use it as the negative terminal, and the +12V for the positive, and end up with a fixed voltage of 24V. But because the weaker rail is only 1A, however, that rating will become the maximum power handling for this combo voltage. Pretty much any combination is possible, even with the positive voltages you already have. The resulting voltage of a combo is based on the difference between the rails used. All the while, remembering that the weakest of the two rails becomes the maximum amperage for that connection. Important note: What you have labelled as GND on the front panel of this project is not really a "ground"... it is the 0V terminal, which is different. This is relatively unimportant on your positive-only build, but it would become the positive terminal if negative voltages were to be used. Also, if I wanted to power an analog op-amp, for example, with a pos and neg voltage, the 0V terminal becomes the "return" for both voltage rails. So, for any build with negative voltages, I would recommend renaming that terminal to "0V". Here are more examples of other fixed voltages possible by combining rails: Resulting Voltage : Pos. connection / Neg. connection (Max. Amperage) 1.7V : +5V/+3.3V (Max. 17A) 7V : +12V / +5V (Max. 18A) 8.7V : +12V / +3.3V (Max. 17A) Obviously, the addition of negative rails extends this even more. Some old Apple PSUs even have 28V rails, that would be interesting to play with! How useful these other voltages are actually gonna be is up to you, I'm just showing they are there if you need 'em. ;-]
Thanks for the info. For my use case and projects, the standard 12V, 5V and 3.3V are what I almost exclusively use. Occasionally, I need something like 9V or 7.4V so I opted to add the variable side so I can get all these other "in between" voltages...anything between 3.3V and 12V. Often I'll need two different voltages (like 3.3V and 5V) at the same time so I created this particular design to be able to provide up to three different voltages (two fixed + 1 variable) simultaneously. Most of my projects are generally very low current (microcontrollers, etc.), so I don't have to be concerned too much with hitting max wattage unless I'm powering something like a large LED display. But it's good to know that other fixed combinations could be possible by using the negative voltages. I don't particularly have a use for other 'fixed' voltages since I can use the variable side for those... and don't have a need for anything over 12V, it but others might, so thanks for sharing the details. It is appreciated!
The master power switch on the left? No, it actually powers off the feed from the power supply to all jacks. There is a low power standby 5V line that is always hot and this keeps the power on to the LEDs and the small displays any time the power supply is plugged in, but the power switch turns on/off everything else from the power supply. Note that the jacks will retain some power for a bit after switching off due to retained power in the capacitors on the PC power supply. But this is shown on the power meters if connected.
@@ResinChemTech Was talking about the mystery switch in your design prototype that you pulled out of the final version. Your channel rocks, thanks for all your hard work
Ah! I actually had to go back and watch the video to remember what I had shown. And actually, I built the power supply almost 3 years ago, well before I made the video. I even looked at my related blog article, where it says "I eliminated the second toggle in the final design"... but I actually don't recall what I might have originally planned for a second switch! There's always the possibility that I just didn't know which style of switch I wanted to use, so I had them both sitting there on the initial layout plans. If I did have some sort of plan for the second toggle, I can't for the life of me recall what I was going to use it for! Yeah... the ol' memory ain't what it used to be! But thanks for watching and thanks for the kind words. If I ever do recall what I planned for the second toggle (if it was anything at all), I'll be sure to follow up!
Thanks for the vid and for the time you took to explain everything into detail. I just have one question. Is there any 3 position switch/selector that could bring the voltage to your above positive connections? (Without the need for the jumper). I hope I explain the idea more or less ok. 🙂
You could certainly wire in a three way switch that would select which 'feed' from the bottom jacks connect to the one on the top. You'd really just need to connect the + voltage line... as the grounds are all common. Just be sure that any switch is properly rated for both the voltage and current. As an aside, you can connect directly to the + voltage post at the bottom and either one of the grounds, skipping the 'jumper'. This would just bypass the meter... so if you connected a lead to the +5 post (blue) on the bottom and ground post and get 5V.. or the 3.3V post (green) and ground to get 3.3V. The jumper is really just running the voltage up to the meter for displaying the voltage. But yes, you could use three leads with a selector switch to accomplish the same thing and move the switch position instead of a physical jumper.
@@ResinChemTech Many thanks for the inputs. Its really a balance between the cost of the project. But I really liked this one over the hundred of examples all over the internet. !
@@ResinChemTech many thanks.. we always need to balance the cost and it seems the switch to handle this current is not cheap. But i will consider. I really liked your project. Congrats
It was hard to show in the video, but where higher current may be used, multiple wires are bundled together... at least three. This is exactly for the reasons you describe. I also didn't mention that from the terminal block to the front outputs, I used higher gauge wire. Most of my projects are pretty low current but as I cover in the related blog, if you plan in using higher current, you may want to consider 12 or even 10 gauge wire. Thanks for the comment!
@@ResinChemTech Thanks for clearing that up. I was a little unclear on what you meant by bundling up the wires. I'm going to assume that the power supply has some overcurrent protection. Also, you did mention appropriate sizing of the fuses. All in all, an excellent build video. Thanks for the time you took to put it together.
very impressed with your write up even included all hyper links to supplys much appreciated the time you put into this project. top notch work my friend
While creating a corresponding blog article (and/or Github repo when there is code involved) takes a ton of additional time, I like to provide what I feel is everything needed for someone to duplicate or adapt one of my projects. I always got frustrated when I was first starting out (and didn't know a lot) and would watch a "how-to" video that omitted a big portion of the actual "how-to".... or required me to try to freeze frame the video to see something like a wiring diagram.
Thanks so much for the kind words. It is comments like yours that encourage me to continue to invest time in things like related blog articles and wikis.
best Benchtop Power Supply from an old PC on YT ,it looks nice.
Thanks. There are a lot of different versions and designs on RUclips. I build this one for my particular needs... I wanted the fixed 3.3V, 5V and 12V... but also wanted an option for a variable voltage, as sometimes I want 9V or 7.4V, etc. This does make the design a bit bigger but does give both options... and option for multiple simultaneous outputs.
Thanks for watching and taking the time to leave a comment!
This is great. I was actually watching a couple videos last week about making something similar but you did the best job explaining all the parts and how to do it step by step. I always enjoy your LED videos as well. You’re a very good teacher.
Also, now I don’t feel so bad for having a bunch of old computers lying around. I’m not alone lol. I also always wondered why the power supplies had so many different plugs coming out of them. I just never paid attention to that part of the PC build. As long as the plugs fit I was happy lol.
Thanks so much for the kind words. While it does make my videos longer than most, I do try to always explain the "why" as well as the "what" so that newer folks might be able to take the concepts and adapt them to their own projects.
I'm always trying to come up with ideas on how to recycle and extend the life of my old electronics. I have a few other ideas and hope to implement them if I can ever find the time. Thanks again for the comment!
I like the design - fun project. I may build this or something similar in the future.
Thanks! It's a handy thing to have on hand for working on other projects.
Great work. I have been planing to do one of these but most of the things I could find didn't give as thorough an explination, thanks. I now have a better understanding of what is going on in the power supply that i should be able to figure it out.
Thanks. This was one of my favorite projects and also resulted in something that I use on a regular basis. Glad you found the video helpful.
This is by far the most concise and easy to follow instruction on RUclips. thank you.
You are welcome! I'm not sure that a 30 minutes video would be considered "concise" by RUclips standards, but I do try to include everything someone would need to know to create their own version of a project.
Thanks for watching and taking a moment to post a comment. I appreciate it!
Great video, well-explained and useful. Thanks for sharing
You are very welcome. Glad you found it useful. Thanks for the comment.
Excellent video.
Thanks! I greatly appreciate it!
Awesome video, thank you!!!
You are welcome! And thanks for watching.
hi, you create very useful and detailed content, I really like your videos, they teach a practical approach, not just bragging about what I have. Thanks!
Thanks! While it does make my videos a bit longer, I always try to explain 'why' I'm doing something as opposed to just "how". I do this so that hopefully viewers can take my ideas or projects and adapt them to their own projects or needs.
Thanks for watching... and taking the time to post your kind words!
Excellent video, thank you for the time taken to share it with us. I built one very similar 'cept I used an old enclosure from a worn out VCR made in the '90s. Just had to make a new front panel.
That's great! And you found new life for even another old piece of electronics in the process. I probably hang on to too many old devices, but I do really like when I can turn one of these old devices into something that I can use today.
Thanks for watching and taking a few minutes to post a comment!
very cool
excellent
About to power a larger fan from a PSU, you answered my question, can I connect the periphial wires in parallel to get to a higher wattage. Thanks!
Thats awesome. I just need a variable supply, but now that I've seen this it makes me feel like I'd be wasting the other rails for fixed voltages if I don't use them. I also have a 275w Dell psu with the two 12v rails. Hmm.
Honestly, I use the fixed rails much more often than the variable side. But most of my projects are ESP based so pretty much everything runs off of 5V or 3.3V. Occasionally, I'll have something that needs 12V. About the only time I use variable is if I need something that represents a battery... like 9V or 7.4V. But the majority of my projects use one of the three fixed voltages.
But the nice thing I like is that I have the flexibility with this type of supply... including the ability to provide up to 3 different voltages simultaneously on the very rare case it is needed. But it could be made smaller if desired by just creating a variable output.
@@ResinChemTech I'm looking at using one of these "LTC3780" buck/boost converters I seen in another video. I really only need current limiting, so I can track down short circuits on other boards with voltage injection.
Don't those fuse holders only have a rating of 10A ? 🤔 I made similar bench power supply but i just used inline spade fuse holders they are used in Automotive industry for bigger Amps than i will ever use but just using 20A on mine for 2 separate 12v feeds on variable outputs then 3.3v 5v separately all on individual voltmeters. Great video 👍
Brilliant video. I wanted to add 50 amp 75 mV shunts to the meters. Would you happen to have a shunt hook up schematic that relates to this particular project. Thank You.
No, sorry. I don't have any diagrams or schematics that show use of a shunt for this project.
@@ResinChemTech ok, thank you.
Hello. This instruments need shunt for more then 10 amps. I build a similar project with boost and buck on the variable side. The voltage works great but the amps is not so good. Could be the shunts?
You’re great
Fantastic video, thank you for the time taken to create a detailed explanation. I have been wanting to buy one of these, but being a fan of DIY, I think I will rather build my own as well.
Thanks! This is one of those projects that is a good DIY project that doesn't require a lot of soldering or 'advanced' electronic skills. However, unlike some of my others, it isn't necessarily a cost-savings one when compared to retail versions. But there's the DIY satisfaction plus the recycling of old components into something new.
Thanks for watching and taking time to comment! Good luck if you decide to build your own... and let me know if you have any questions. Definitely peek at the blog article as there are some things in there that I didn't cover in the video just due to time/length of the video.
Ficou muito bom, parabéns! É uma pena eu não ter como fazer um gabinete tão bonito assim.
I am a new subscriber. I have watched a majority of your LED videos and enjoyed them all, however, this video is what made me finally click the subscribe button. The longer format is excellent for me because it allows enough time to enjoy and understand your work.
Thanks! I know that my videos are simply too long for a lot of people. But I know that when I was first learning, I'd get frustrated at times when watching other project videos because so much of the detail was skipped that I couldn't really understand what was being shown... nor could I duplicate it without doing a lot of additional research.
I try to show not only the 'how'... but also include the 'why', so that viewers that are a bit newer in their journey can understand the concepts and maybe adapt them to their own projects. Lately, I've been trying to do a quick overview of the final project first (for those that don't care about the details), then follow that up with the longer step-by-step process for those that want to hang around a bit longer.
Thanks for watching... and for subscribing! Welcome to the channel!
Super appreciate the in-depth explanation of your video(s). I was curious to know the gauge of wire you used to wire in the 2 different 100k pots? Will the gauge thickness alter the potential of the potentiometer?
I believed that I used 18 gauge, but I don't recall absolutely for sure. These are very short runs, and if any voltage drop is truly present, it really doesn't matter since the actual voltage being output to the terminals is shown on the meter, which is measuring voltage after the potentiometer and any other wiring.
The only place that this might matter would be at the very top end (e.g. 12V). Say there really was a 0.1V drop due to wire resistance. The max output of the meter (and the true voltage) would only be 11.9V. I don't think this is really the case... and even it if was, I'd just use the fixed 12V side anyway. The net result is that the potentiometer is used to adjust the desired voltage, which is measured AFTER any resistance that would be introduced by any wiring. I hope I explained so that it makes some sense! The key thing is to just select a potentiometer with a resistance that isn't too sensitive and yet allows you to cover the range from around 1V up to 12V. If wire gauge impacts that, then a different potentiometer can be used.
Hello. Me again. I got my power supply put together and upon testing it, the fixed side works perfectly :) Not so much on the variable side :( I get 12V out of the jacks and the meter and my DVOM show that, but the variable part is not working. I tested my potentiometers and they check out good. But when I turn the knobs when they are hooked up in circuit...nothing. I had some trouble getting the little blue potentiometers off of the buck converter and also in attaching the red, white, and black wires. My guess is that I screwed up something...too much heat and I cooked some of those SMD capacitors and/or resistors on the bottom of the PCB? When I test the set of three connections on the PCB that feed the CV pot and the three from the CC side I get nothing. My uneducated guess that I cooked something and that I need a new buck converter. Any advice would be greatly appreciated. Thanks for your time
Removing those tiny pots and soldering wire on... while assuring none of those wires touched each other... was probably one of the trickiest parts of the entire build. After getting the wires soldered on for the first connection, I used liberal amounts of liquid tape to assure that I didn't inadvertently create a connection when soldering on the second set.
I can only recommend what I did. First, I tested the buck converter fully with the existing pots... feeding in 12V and using a multimeter on the output side to measure voltage as I turned the tiny set screw. I then de-soldered the existing pots and connected just the first set of wires (voltage adjustment I believe) and repeated my tests. This is also where I tried different rotary pots to select a value with a decent range vs. precision. When all that was done, then I threw on the liquid tape, soldered the second set and tested again. I don't know if any of that is helpful... but I do know this particular step can be a bit difficult.
@@ResinChemTech Thanks for getting back to me. So, based on my description above, do you think I messed something up in the buck converter and I should get a new one?
It's hard to say for sure whether the buck converter has been damaged or if it is a wiring/soldering issue. But if you aren't getting any measurable voltage out, it's obviously not going to work. Any possibility that you can try de-soldering the wires you added and then try soldering them back again? If you don't have a good connection where the original pots used to be, I'm guessing you'd have an open circuit and not be getting any output. I guess I'd try to troubleshoot and assure the current buck converter is damaged or not salvable before buying a new one. But it's up to you to decide whether the effort to try to fix the first one is worth the cost of just buying a new one.
@22:43 I tried something similar with exact same model of CC CV regulator but the external 100k pot seem to have damaged my unit ! After discussing this matter on several electronic forums it was concluded that due to their sensitivity inherited from manufacturing process, you cannot simply replace an onboard Trim Pot with external potentiometer as the added capacitance / inductance can destroy the DC DC regulator.. Did you experience something like this at all ?
I haven't experienced any issues like you describe. I've been using the power supply for well over a year... often using the variable voltage... and the system has been holding up fine. I will admit that I fried a number of pots initially as I tried finding the correct ones to use for the best range vs. precision, but that was due to a miscalculation on my part with the estimated current.
I don't doubt what others are saying... I am by no means an electrical engineer or even anything close to an expert when it comes to things like capacitance and inductance. But it has been working just fine for me... at least on the voltage side. I rarely limit the current mostly because the types of projects that I work on don't need to work within a limited current range. I probably wouldn't even add the current limiter if I were to do it over again.
Maybe I've just been lucky... but so far I haven't experienced any issues that damaged any of the components. For my variable range, I'm generally using it for voltages between the 5 and 12V range that I can't get out of the fixed voltage outputs.
@@ResinChemTech Thanks for the feedback. I still have a spare regulator module so let see if I get lucky the 2nd time :).
Have you thought about adding stackable banana plugs for your meter so you don't have to unscrew it all the way You can just unplug it and stack it on to another connection?
Since I made this video, I have created a bunch of different jumpers and connectors, including banana plug jumpers similar to what you mentioned that lets me quickly switch the voltage... although most of the time it either gets used for 5V on the fixed side or the variable side gets used for "off" voltages for things like 7.4V or 9V.
Thanks for the suggestion!
Hi great presentation of this technical project. I thought I’d give it a try. Well at this point I get the stand by light to go on, then turn the switch on and the power on light (gray wire) will not turn on. I have green to switch then to black. The purple is connected to the stand by led and lights up so led works. I’m using 120 ohm ¼ w resistors soldered to cathode side then solder to black (-) buss. I tested each led circuit and they were ok. When I turn the switch on the power on does not light up (gray) fan runs for like 2 secs. Where do I go from here? All other wiring is as it should be. Again great presentation, I love a challenge. thanks
First off, have you verified the wiring color for your particular power supply? I was using a Dell power supply and they were notorious for using non-standard/proprietary components. So it is possible that the wiring on yours is slightly different, especially for the secondary wiring.
But more importantly, did you provide a load? You mentioned a resistor for the LEDs, but most power supplies will not power on (or immediately power off) if a load is not detected. I had to add a 10W 10 ohm resistor to the 5V line. If you haven't looked at my blog article, take a peek at that and the section under "providing a load": resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html
Those are my initial thoughts. But if it runs for like 2 seconds and shuts down, it is likely due to a safety feature... one of which is not detecting a load (a short/over-voltage, etc. are other reasons). But be sure you are supplying a load... that sounds like the most likely cause here.
hi thanks for the reply. i went back and watched the video and reread the step0by-step you put together and i realized i did not use the brown wire (sensing wire). i saw it in one of the diagrams in the writeup but it was not labeled. you mention this wire 3 times and it's usage is explained but you have to dig for it, i like a challange. so i connected the brown to 3.3v (orange) wire and it worked! i then swred it up and no go, something happened. i had another psu so i broke it down, wired it up and it didn't work either. i think both psu's have other issues, so i will get another psu and try is again. i also had the load suing 5v (red) w/10ohm10w resistor.
so back to the challenge for me...
Hello. Please excuse my ignorance but if the max input is 12vdc, is the max output 12dc as well? Can I use a different converter to get say 36vdc at like 10 amps? Thanks for your time.
You could use something like a step-up boost converter to shift up the voltage. But you'd also need to be sure you don't exceed the power ratings for the power supply itself (36V x 10A = 360 watts) and that the boost converter (and all wiring/connections) were rated to handle the 10 A current.
Hello, Im from Brazil and I´m using this same Dell Power Supply in my project, but I´m trying to get 24 voltage using the Blue wire (-12v) and +12voltage yellow or white .... But I don´t know what is happening, when I connect this some times I got a burn in the Diode of -12V wire... I can use one Module XL4005 and only regulate the C. voltage and not CC ?
I can't say for sure and I'm certainly not an electrical engineer, but my gut tells me that even though there is a +12V and -12V available, the power supply likely was not designed to truly produce 24V. I don't know of any components on the original server that would have required 24V.... although there may have been both some +12V and -12V components.
If you truly need 24V, the only reliable option may be to integrate a 24V boost converter in your build, something similar to this: amzn.to/4etwxso. You could even replace the buck converter with this, but you would need to feed in a lower voltage (3.3 or 5V) if you want the lower variable voltages, as this is only a boost and not a buck converter. It also has a lower max amp rating and would not offer the constant current option. Or you could add this in addition to the buck converter and create a fixed 24V output as well. There are numerous options, but it sounds like using +12V and -12V to get 24V directly may not work with only the power supply itself.
@@ResinChemTechYes, I agree whit you... I believe the buck send elevated CC and its damage The diode or other components... probably I will use only 12v regulated tensions like your project...
but this trhouble is very strange for me, cause the most atx are prpjected to do 24v, using - 12v negative for this... wellz thx for help and congratulations for this channel
very helpful video and really well explained, I'm building one of these by myself too but I can't find any information about the resistive load. All I know is that it is needed if the voltages are not stable at the max ampere rail. In my case is the 5v 30A, this one is stable but the 12v and 3,3v are not, do I need to put a load one both of those rails? or being the higher amp rail stable is enough? sorry for my bad english.
You probably need to find and review the documentation for your particular power supply as it may be different than mine. For my Dell power supply, it required a load on the 5V line. In my case, a 10W 10 Ω resistor placed on a 5V lead provided enough resistance. Also, assure if you have a sense wire, it is connected to the proper voltage as it will sense the power output from the supply and make adjustments to keep the voltage stable. In my case, it was a 3V3 sense wire. If this isn't connected, it could result in the voltage being constantly adjusted, leading to what you are seeing.
But you need to check the documentation for your particular power supply as it may be different from mine.
@@ResinChemTech oh ok that is good information. Thanks for the help!
Thank you so much for posting this. I'm building this now. One question. When I do the calculations for the resistor size to use on the led indicator lights on a 12V power supply, I come up with I get around 400-800 ohms (using the ave. value for and led). In your plans it shows resistance values of 120 and 140 ohms. Am I missing something? Thanks
The LEDs are being powered by the PS5AUX and PS_PWRGOOD lines, which are 5V and not 12V. So the resistors selected are based on 5V. But they are really bright! If I had it to do over again, I might select larger resistor values to dim those LEDs down just a bit.
@@ResinChemTech Thank you for taking the time to reply. I knew I had to be missing something!
I would like to know variable side can you used 18v or 36v ?
Not as designed. The only voltages natively available from the PC power supply are 12V, 5V and 3.3V. For the variable side, I use a step-down buck converter that can take the maximum 12V feed and step it down anywhere from the source 12V down to 1.2V. To get higher voltages, you'd need to supply a step-up or boost converter.
In doing so, you'd have to be careful not to exceed the maximum wattage for the PC supply and also assure you do not exceed expected current ratings.
There are special plugs people use to test ATX PSUs without load. I am quite certain they do not have any heating issues. So perhaps there's a better way to keep the PSU on - the heat inside the enclosure is hardly critical, but it kinda sets off the perfectionism alarm
They only test whether the PSU comes on (by shorting the green wire to ground). To actually work correctly does require a load on the 5v rail.
Hi, congratulations for the video. I would like to try to imitate your project. I would like to ask you if it is possible to have the schematic of the project. Thanks and good job.🛩
You can find the complete parts list, wiring diagrams and step-by-step directions in the related blog article to this video: resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html
@@ResinChemTech Thanks.
how many amp fuses did you use?
The fuse size depends upon both the voltage and the specifications/wattage of your particular power supply. You fuse it so that you do not exceed the max power for a given voltage. I cover the particular fuses that I used for my power supply in the written version of this project: resinchemtech.blogspot.com/2022/07/bench-top-power-supply.html
sorry if it is a stupid question. Can I still use the feeding wire if the front panel is in metal and not 3D printed?
Not sure what you mean by the feeding wire, but you can use a metal front plate instead of a 3D printed faceplate. Just be sure that everything is properly insulated and the metal front panel does not become energized by any of the circuits.
Nice one!
Other fixed voltages are also possible...
The -12V (also -5V on even older PSUs) can be used in combination with the other voltage rails to produce a variety of fixed voltages. For example, if the -12 is made available with its own socket, one could use it as the negative terminal, and the +12V for the positive, and end up with a fixed voltage of 24V. But because the weaker rail is only 1A, however, that rating will become the maximum power handling for this combo voltage.
Pretty much any combination is possible, even with the positive voltages you already have. The resulting voltage of a combo is based on the difference between the rails used. All the while, remembering that the weakest of the two rails becomes the maximum amperage for that connection.
Important note: What you have labelled as GND on the front panel of this project is not really a "ground"... it is the 0V terminal, which is different. This is relatively unimportant on your positive-only build, but it would become the positive terminal if negative voltages were to be used. Also, if I wanted to power an analog op-amp, for example, with a pos and neg voltage, the 0V terminal becomes the "return" for both voltage rails. So, for any build with negative voltages, I would recommend renaming that terminal to "0V".
Here are more examples of other fixed voltages possible by combining rails:
Resulting Voltage : Pos. connection / Neg. connection (Max. Amperage)
1.7V : +5V/+3.3V (Max. 17A)
7V : +12V / +5V (Max. 18A)
8.7V : +12V / +3.3V (Max. 17A)
Obviously, the addition of negative rails extends this even more. Some old Apple PSUs even have 28V rails, that would be interesting to play with! How useful these other voltages are actually gonna be is up to you, I'm just showing they are there if you need 'em. ;-]
Thanks for the info. For my use case and projects, the standard 12V, 5V and 3.3V are what I almost exclusively use. Occasionally, I need something like 9V or 7.4V so I opted to add the variable side so I can get all these other "in between" voltages...anything between 3.3V and 12V. Often I'll need two different voltages (like 3.3V and 5V) at the same time so I created this particular design to be able to provide up to three different voltages (two fixed + 1 variable) simultaneously. Most of my projects are generally very low current (microcontrollers, etc.), so I don't have to be concerned too much with hitting max wattage unless I'm powering something like a large LED display.
But it's good to know that other fixed combinations could be possible by using the negative voltages. I don't particularly have a use for other 'fixed' voltages since I can use the variable side for those... and don't have a need for anything over 12V, it but others might, so thanks for sharing the details. It is appreciated!
Presumably the switch was for turning on the variable power supply on / off ?
The master power switch on the left? No, it actually powers off the feed from the power supply to all jacks. There is a low power standby 5V line that is always hot and this keeps the power on to the LEDs and the small displays any time the power supply is plugged in, but the power switch turns on/off everything else from the power supply. Note that the jacks will retain some power for a bit after switching off due to retained power in the capacitors on the PC power supply. But this is shown on the power meters if connected.
@@ResinChemTech Was talking about the mystery switch in your design prototype that you pulled out of the final version.
Your channel rocks, thanks for all your hard work
Ah! I actually had to go back and watch the video to remember what I had shown. And actually, I built the power supply almost 3 years ago, well before I made the video. I even looked at my related blog article, where it says "I eliminated the second toggle in the final design"... but I actually don't recall what I might have originally planned for a second switch! There's always the possibility that I just didn't know which style of switch I wanted to use, so I had them both sitting there on the initial layout plans. If I did have some sort of plan for the second toggle, I can't for the life of me recall what I was going to use it for! Yeah... the ol' memory ain't what it used to be!
But thanks for watching and thanks for the kind words. If I ever do recall what I planned for the second toggle (if it was anything at all), I'll be sure to follow up!
Thanks for the vid and for the time you took to explain everything into detail. I just have one question. Is there any 3 position switch/selector that could bring the voltage to your above positive connections? (Without the need for the jumper). I hope I explain the idea more or less ok. 🙂
You could certainly wire in a three way switch that would select which 'feed' from the bottom jacks connect to the one on the top. You'd really just need to connect the + voltage line... as the grounds are all common. Just be sure that any switch is properly rated for both the voltage and current.
As an aside, you can connect directly to the + voltage post at the bottom and either one of the grounds, skipping the 'jumper'. This would just bypass the meter... so if you connected a lead to the +5 post (blue) on the bottom and ground post and get 5V.. or the 3.3V post (green) and ground to get 3.3V. The jumper is really just running the voltage up to the meter for displaying the voltage. But yes, you could use three leads with a selector switch to accomplish the same thing and move the switch position instead of a physical jumper.
@@ResinChemTech Many thanks for the inputs. Its really a balance between the cost of the project. But I really liked this one over the hundred of examples all over the internet. !
@@ResinChemTech many thanks.. we always need to balance the cost and it seems the switch to handle this current is not cheap. But i will consider. I really liked your project. Congrats
Is it really ok to combine two thin wires with the intention of running more than the rated current?
It was hard to show in the video, but where higher current may be used, multiple wires are bundled together... at least three. This is exactly for the reasons you describe. I also didn't mention that from the terminal block to the front outputs, I used higher gauge wire. Most of my projects are pretty low current but as I cover in the related blog, if you plan in using higher current, you may want to consider 12 or even 10 gauge wire.
Thanks for the comment!
@@ResinChemTech Thanks for clearing that up. I was a little unclear on what you meant by bundling up the wires.
I'm going to assume that the power supply has some overcurrent protection. Also, you did mention appropriate sizing of the fuses.
All in all, an excellent build video. Thanks for the time you took to put it together.
❤❤