tysm for this video, i've been trying to get this to work for the longest time and couldn't figure out why my supply kept triggering overcurrent, but you explained it perfectly!
This is awesome. Thanks for sharing. I do some retro computing and sometimes I want to replace the original power supply for a new one. But new ones lack negative 5V, mostly. I will try this with my IBM 386, which needs a few mA on -5V...
This is great, really appreciate you sharing this. DC-DC converters with a negative voltage output are virtually non-existent. I was able to finish the re-design of a vintage computer power supply with easy to get parts. Thanks!
What are you talking about? The ICs are easily available. You mean that the full circuit as a cheap bare component is harder to get? Get a isolated one instead!
Hi, I found this site a few years back looking for a regulated negative supply (6v, ~500mA) for my tape recorders from some 18v power tool batteries. It works very well. I added the ripple filter described in the datasheet, giving me about 5-10mV of ripple. I wanted to try two buck converters to give +12/-6V but I haven't had the guts lol. I might get some more boards and study your video more. Thanks for making it!
Those switchmode converters are a very useful tool. I'll tell you one useful application, in analogue circuits where you need an adjustable bias voltage below zero but have only a positive single rail supply
Outstanding. I haven't actually thought to verify this or check that it makes sense, but I never would have guessed the boards could be used in this way. Very clever!
Man I just want to thank you! This video is excellent, and it very helpful. It is the only which I've found about the topic. Really nice content, thank you!
Thanks for that interesting video. I will try to use the negative voltage to get a better control of the p-MOSFET I am using. Lets see if I got everything correct :)
Also remember: Maybe you do not need one! Only the electrodes of the circuit need to be within the rail, that does not mean that the input resistors need to be.
simplemente excelente, aunque solo entiendo ingles y no puedo hablarlo, pude entender todo y fue muy claro. Muchas gracias por tu tiempo y conocimiento.
great video. I hope to use the simplest wiring scheme with two of these boards to get +/-12VDC from the positive 24V rail in my home made amplifier. May i borrow some of your expertise before potentially melting something? Would it be possible to use one of these buck converters on the negative rail in my? My build has two DC power supplies wired with a center ground to get +24,0,-24 VDC the amplifier chips. I already have one of these buck converters on the positive rail to get +12V to some accessories. So, it seems that by having one buck converter on the -24V rail to get -12V, instead of a second on the +24V to get -12V, then i could avoid running LM2596 at 36V, so close to its limit, even though the 350w supplies can probably handle the boot strapping current.
Thank you! I'm sorry, but I'm afraid you can't use this module to regulate - 24V down to - 12V. If you connect ground to IN- and -24V to IN+ it will, sadly, explode. But I'm guessing this is not what you meant. If you connect - 24V to IN - and ground to IN+ the module will see +24V and you'll be able to adjust it for +12V on the output (with respect to -24V), so if you then measure it with respect to your ground it will, indeed, show up as 12V. The problem is you won't be able to power anything from it, as a buck converter can't sink current, it can only source it; this means that current has to flow out of OUT+ and back into OUT-. If you connected any load between this 'fake' -12V and ground the current would have to flow into IN+ which it can't. It's actually the same with linear regulators but it's easier to see there, because you don't have external inductors or diodes, just a series pass element, like a bjt, between the input and the output. Even then it's not that easy to fully explain why it can't work, let's just leave it at transistors are not reversible. Anyway, to properly sink current (in a linear regulator) you need a transistor with a different carrier type, so a pnp instead of an npn, or a P-channel mosfet instead of an N-channel one. Ok, but I digress, my point is the module can't sink current. Now, for the sake of an argument, what if we put a big load on the output of the module, between OUT- and OUT+, like a power resistor that would pull, for example, two amps. For one, we'd lose most of that famous switching regulator efficiency wasting all that power (24W), but I think it would work (up to 2A of load) - the module would only source current out of its OUT+ terminal, with no load between OUT+ and your ground it would source the 2A for the resistor, when some load shows up it would simply source less. But (aside from the aforementioned abysmal efficiency) there's another reason why this idea is not that great - ripple. The ripple on your -24V and the module's ripple would add up. I can't really test what I've just written about, because I've just moved and all my equipment is still in boxes, so take this comment with a grain of salt. Anyway, I hope I answered your question and good luck!
@@FPiorski thanks for the reply, ill just tack it all on to the +24 rail and hope the current draw doesn't cause a big thumb in the speakers. thanks again for your expertise and especially for condensing it into an illustrated video guide for everyone.
@@FPiorski if i have a 12V dc positive and negative Based on what you are saying ishould be able to get +5 -5 +3.3 -3.3 and +12 -12 I just have to mix and match correct?
It's obviously a bit late in replying, but surely you can just use an L7812 linear regulator on the +24V rail and an L7912 on the -24V rail to generate +12V and -12V? You don't say what you want the ±12V for, or what current you want to draw, but that is almost always going to be the right solution for audio work as the noise from SMPS makes them unusable for that.
Will this board supply enough current to replace a 9V battery pack made of AA batteries? I have a metal detector that runs on +12v and -9v with a shared GND. I was wanting to replace it with a higher capacity 12V LiPo pack. Then use a circuit like in this video...to get the -9v.. Would this circuit work for what I am trying to do?
Hi Filip. Thanks for this video. There are also small modules (1 or 2€) which give directly outputs v+, gnd and v-. The output current is not very high (100mA to 800mA, with sometimes a difference between the max current on + and - output). Have you ever been test these modules and what do you think about them?
You can put any two voltage sources in series, as long as they're isolated. So you can chain two batteries, define the point where they meet as ground and say that the other terminal of one of the batteries is V+ and the other terminal of the other battery is V-. You *cannot* do that with most mains SMPS's, because their output grounds are tied to earth ground! In other words, they are not isolated with respect to each other. Don't try to disconnect the earthing from one of them, as then if something fails inside and one of the mains wires touches the chassis, instead of a breaker or an RCD breaking, the chassis will become live and you'll get shocked. So no, you can't *safely* take two mains-powered switchmode power supplies and put them in series to get a negative voltage, unfortunately.
@@FPiorski but its safe to use an old power brick for that right? Connecting positive to ground and using the "gnd" wire as a negative voltage. An alternative is to use a battery for that.
i wished i watched this video sooner it is for over a year now i was trying everything to get a negative voltage from a positive voltage without moving the ground so i can have positive ground and negative and everything i tried ended up to not working or moving the ground i tried the lm2596 too with the schematic in the datasheet but it never worked god knows how much i thought about that and why it doesnt work and finally it seems i know why i cant bellieve the answer was just more amp and i was thinking there must be a short in the circuit i built so i teared everything apart and started again and again and again from scratch i was almost ready to give up on it when i found this video
I have one query other than this: I am using TC7662, its working on breadboard giving me -10V from +10V but when i use it on zero PCB its not. Tried everything still not working. Is it also some kind of startup problem. I am posting this problem to every electronic expert on youtube. Its freaking me out :(
It's a simple charge pump, so either you made a mistake while creating the layout (maybe mirrored it or mistaken the position of pin 1), I have no idea what could be wrong. Maybe you've shorted the output, remember that charge pumps can only source a small amount of current and an unnoticed direct short could be mistaken for a non-working converter. If everything is correct then try swapping the ic, maybe you got a broken one, however unlikely that is.
@@FPiorski have checked it some 1000 times. That's what bothering me. It's simple and not working. When I soldered caps directly on DIP pins two are working, the other 2 don't. It has become an issue of research thn negligence. Do you think inductance and capacitance of PCB have some play or ESR of electrolyte caps I am using.
Oh yes, most certainly. I've spent the last couple of months finishing my degree and working part-time, but now I hope I'll have a bit more time on my hands to make new videos. I have a google doc with video ideas that's more than two pages long, hopefully I'll get to cross some things off the list soon.
Damn :( Yeah, sometimes it has trouble starting up, even with all the mods I mentioned in the video. For me it's still easier than the alternatives though.
@@FPiorski If you are having trouble with it starting up, try adding a rectifying diode between OUT- and OUT+. From the datasheet, "Additional diodes are required in this regulator configuration, ... Without diode D3, when the input voltage is first applied, the charging current of CIN can pull the output positive by several volts for a short period of time. Adding D3 prevents the output from going positive by more than a diode voltage." I had the problem that without the diode the voltage would be indeed positive, about 2.3V, and then it would just simply continue to draw power without ever turning on.
Hello All, I am facing a strange situation. Circuit gives -0.22 output irrespective of potentiometer setting. But whenever on/off pin is touched with dmm circuit starts to work fine. I have also desoldered the components from the pcb and resoldered them on vero board to reproduce this situation. Result is the same. Any help is appreciated.
"Corporate needs you to find the difference between these two schematics"
tysm for this video, i've been trying to get this to work for the longest time and couldn't figure out why my supply kept triggering overcurrent, but you explained it perfectly!
This just blew my mind you can just "Hook it up differently and get a negative voltage". Thank you my man
This is awesome. Thanks for sharing. I do some retro computing and sometimes I want to replace the original power supply for a new one. But new ones lack negative 5V, mostly. I will try this with my IBM 386, which needs a few mA on -5V...
How can I get both positive and negative supply to drive 741cn operational amplifier
This is great, really appreciate you sharing this. DC-DC converters with a negative voltage output are virtually non-existent. I was able to finish the re-design of a vintage computer power supply with easy to get parts. Thanks!
What are you talking about? The ICs are easily available. You mean that the full circuit as a cheap bare component is harder to get? Get a isolated one instead!
@@0MoTheG Hmm, I'm dumb. can you give me an example of one on Amazon? I can't find any.
Hi,
I found this site a few years back looking for a regulated negative supply (6v, ~500mA) for my tape recorders from some 18v power tool batteries. It works very well. I added the ripple filter described in the datasheet, giving me about 5-10mV of ripple. I wanted to try two buck converters to give +12/-6V but I haven't had the guts lol. I might get some more boards and study your video more. Thanks for making it!
Those switchmode converters are a very useful tool. I'll tell you one useful application, in analogue circuits where you need an adjustable bias voltage below zero but have only a positive single rail supply
Outstanding. I haven't actually thought to verify this or check that it makes sense, but I never would have guessed the boards could be used in this way. Very clever!
This is exactly what I needed, thanks from Colorado.
Works great. I just implemented it for the analog section of my DIY oscilloscope.
Man I just want to thank you! This video is excellent, and it very helpful. It is the only which I've found about the topic. Really nice content, thank you!
Thanks for that interesting video. I will try to use the negative voltage to get a better control of the p-MOSFET I am using. Lets see if I got everything correct :)
I used the xl7015 buck converter for negative voltage in a similar way, it is excellent
Thank you for sharing. I have the same Buck converter. While I do get a negative voltage, it starts getting really hot.
Hi, I try it, but mine does not work at all.
I don't do any mod still.
Any idea?
Excellent video. Informative an entertaining. You earned a new sub. also, good narration voice. Please keep making more!
I probably wouldn't thought of this. Nice!
Also remember: Maybe you do not need one!
Only the electrodes of the circuit need to be within the rail, that does not mean that the input resistors need to be.
Oh yes, most definitely.
On that note, bootstrapping op amps is an interesting technique (like in AN-1593)
simplemente excelente, aunque solo entiendo ingles y no puedo hablarlo, pude entender todo y fue muy claro. Muchas gracias por tu tiempo y conocimiento.
Thanks! Any photos or diagrams of the specifics of that mod? Thanks in advance👍🏻
I trying the same but not works.
Power supply show a short circuit.
great video. I hope to use the simplest wiring scheme with two of these boards to get +/-12VDC from the positive 24V rail in my home made amplifier. May i borrow some of your expertise before potentially melting something? Would it be possible to use one of these buck converters on the negative rail in my? My build has two DC power supplies wired with a center ground to get +24,0,-24 VDC the amplifier chips. I already have one of these buck converters on the positive rail to get +12V to some accessories. So, it seems that by having one buck converter on the -24V rail to get -12V, instead of a second on the +24V to get -12V, then i could avoid running LM2596 at 36V, so close to its limit, even though the 350w supplies can probably handle the boot strapping current.
Thank you!
I'm sorry, but I'm afraid you can't use this module to regulate - 24V down to - 12V. If you connect ground to IN- and -24V to IN+ it will, sadly, explode. But I'm guessing this is not what you meant. If you connect - 24V to IN - and ground to IN+ the module will see +24V and you'll be able to adjust it for +12V on the output (with respect to -24V), so if you then measure it with respect to your ground it will, indeed, show up as 12V. The problem is you won't be able to power anything from it, as a buck converter can't sink current, it can only source it; this means that current has to flow out of OUT+ and back into OUT-. If you connected any load between this 'fake' -12V and ground the current would have to flow into IN+ which it can't. It's actually the same with linear regulators but it's easier to see there, because you don't have external inductors or diodes, just a series pass element, like a bjt, between the input and the output. Even then it's not that easy to fully explain why it can't work, let's just leave it at transistors are not reversible. Anyway, to properly sink current (in a linear regulator) you need a transistor with a different carrier type, so a pnp instead of an npn, or a P-channel mosfet instead of an N-channel one.
Ok, but I digress, my point is the module can't sink current. Now, for the sake of an argument, what if we put a big load on the output of the module, between OUT- and OUT+, like a power resistor that would pull, for example, two amps. For one, we'd lose most of that famous switching regulator efficiency wasting all that power (24W), but I think it would work (up to 2A of load) - the module would only source current out of its OUT+ terminal, with no load between OUT+ and your ground it would source the 2A for the resistor, when some load shows up it would simply source less. But (aside from the aforementioned abysmal efficiency) there's another reason why this idea is not that great - ripple. The ripple on your -24V and the module's ripple would add up.
I can't really test what I've just written about, because I've just moved and all my equipment is still in boxes, so take this comment with a grain of salt. Anyway, I hope I answered your question and good luck!
@@FPiorski thanks for the reply, ill just tack it all on to the +24 rail and hope the current draw doesn't cause a big thumb in the speakers. thanks again for your expertise and especially for condensing it into an illustrated video guide for everyone.
@@FPiorski if i have a 12V dc positive and negative
Based on what you are saying ishould be able to get +5 -5 +3.3 -3.3 and +12 -12
I just have to mix and match correct?
It's obviously a bit late in replying, but surely you can just use an L7812 linear regulator on the +24V rail and an L7912 on the -24V rail to generate +12V and -12V? You don't say what you want the ±12V for, or what current you want to draw, but that is almost always going to be the right solution for audio work as the noise from SMPS makes them unusable for that.
Thank you very much. Just so happen to have one of those on hand.
I did the exact same and my buck converter just gets warm am i doing something wrong ? im using a LM2596S the board looks the same as the video
Hello from EiTI 👌🏽
Will this board supply enough current to replace a 9V battery pack made of AA batteries?
I have a metal detector that runs on +12v and -9v with a shared GND.
I was wanting to replace it with a higher capacity 12V LiPo pack.
Then use a circuit like in this video...to get the -9v..
Would this circuit work for what I am trying to do?
This video is fantastic. Please please please make more videos!
Great presentation, thank you man!
cool vid my guy! i like your style and your builds are sick
I used to use buck-boost converter for negative voltages
wait, is the output capacitor ok with this being polarized?
what data sheet did you get the info from,i can't find the lm2596 used as an inverter anywhere.
Hello sir ,can i know what is the value of the tiny yellow capacitor that you put? thanks
Thank you very much for sharing. Very helpful
Is there any way to do the startup input current trick on the XL4016? It doesn't have an enable pin
Please I need variable analog converter circuit diagram that can convert 0-10v positive 0 to -10v output.
Hi Filip. Thanks for this video. There are also small modules (1 or 2€) which give directly outputs v+, gnd and v-. The output current is not very high (100mA to 800mA, with sometimes a difference between the max current on + and - output). Have you ever been test these modules and what do you think about them?
Bought one of those recently. Waiting for it to arrive. Those are low cost and do seem much easier👍🏻
can we use two smps dc psu unit in serial and we have negative voltage? For op-amp ics ?
You can put any two voltage sources in series, as long as they're isolated. So you can chain two batteries, define the point where they meet as ground and say that the other terminal of one of the batteries is V+ and the other terminal of the other battery is V-.
You *cannot* do that with most mains SMPS's, because their output grounds are tied to earth ground! In other words, they are not isolated with respect to each other. Don't try to disconnect the earthing from one of them, as then if something fails inside and one of the mains wires touches the chassis, instead of a breaker or an RCD breaking, the chassis will become live and you'll get shocked.
So no, you can't *safely* take two mains-powered switchmode power supplies and put them in series to get a negative voltage, unfortunately.
@@FPiorski but its safe to use an old power brick for that right? Connecting positive to ground and using the "gnd" wire as a negative voltage.
An alternative is to use a battery for that.
Very handy discovery. Cheers!
i wished i watched this video sooner it is for over a year now i was trying everything to get a negative voltage from a positive voltage without moving the ground so i can have positive ground and negative and everything i tried ended up to not working or moving the ground i tried the lm2596 too with the schematic in the datasheet but it never worked god knows how much i thought about that and why it doesnt work and finally it seems i know why i cant bellieve the answer was just more amp and i was thinking there must be a short in the circuit i built so i teared everything apart and started again and again and again from scratch i was almost ready to give up on it when i found this video
That's painful to read without full stops.
I buy the ti chip and made the inverting circuit, damn I got plenty of that board🤦
I have one query other than this:
I am using TC7662, its working on breadboard giving me -10V from +10V but when i use it on zero PCB its not. Tried everything still not working. Is it also some kind of startup problem. I am posting this problem to every electronic expert on youtube. Its freaking me out :(
It's a simple charge pump, so either you made a mistake while creating the layout (maybe mirrored it or mistaken the position of pin 1), I have no idea what could be wrong. Maybe you've shorted the output, remember that charge pumps can only source a small amount of current and an unnoticed direct short could be mistaken for a non-working converter. If everything is correct then try swapping the ic, maybe you got a broken one, however unlikely that is.
@@FPiorski have checked it some 1000 times. That's what bothering me. It's simple and not working. When I soldered caps directly on DIP pins two are working, the other 2 don't.
It has become an issue of research thn negligence.
Do you think inductance and capacitance of PCB have some play or ESR of electrolyte caps I am using.
Such boards from China normally have a fake chip, it is not from TI. But of course the board will work.
Great video!
Really great idea.....thanks!
Just turn your meter wires the other way round!!
YMMD 🤣
Also while the chip being fake it doesnt make it that different. Sometimes All that really changes is the long term durability
It is a bad regulator because in the event of a short, the unregulated voltage passes to the load.
Yikes!!!
They have dc dc step down modules. Can't we just take one of these and one buck?
Very interesting and useful 🔥
hey, do you plan to make any new videos?
Oh yes, most certainly. I've spent the last couple of months finishing my degree and working part-time, but now I hope I'll have a bit more time on my hands to make new videos. I have a google doc with video ideas that's more than two pages long, hopefully I'll get to cross some things off the list soon.
Thank you so much!
This situation will also happen in step you boost converter
Excellent Ty very much
Thank you
If anyone wants to talk about electronics or something you can now find me on discord
discord.gg/2g8sNeYb84
Neat !
no funciona genera corto y se calienta el integrado
Greate
nobody else spots the issue?
What issue 😂 please tell 😂😂
I took inspiration from your video and made
two sources. I didn't include regulated output.
ua-cam.com/video/IOWBzwuNXaw/v-deo.html
Didn't work
:(
Damn :(
Yeah, sometimes it has trouble starting up, even with all the mods I mentioned in the video. For me it's still easier than the alternatives though.
@@FPiorski :) but I learnt some new things from your video, Thank you
@@FPiorski If you are having trouble with it starting up, try adding a rectifying diode between OUT- and OUT+. From the datasheet, "Additional diodes are required in this regulator configuration, ... Without diode D3, when the input voltage is first applied, the charging current of CIN can pull the output positive
by several volts for a short period of time. Adding D3 prevents the output from going positive by more than a
diode voltage." I had the problem that without the diode the voltage would be indeed positive, about 2.3V, and then it would just simply continue to draw power without ever turning on.
Hello All, I am facing a strange situation. Circuit gives -0.22 output irrespective of potentiometer setting. But whenever on/off pin is touched with dmm circuit starts to work fine. I have also desoldered the components from the pcb and resoldered them on vero board to reproduce this situation. Result is the same. Any help is appreciated.
MvP
.m @Filip Piorski