I'm designing an adjustable power supply (using LM338 instead of LM317 to get 5A out of it) and I do have a second output in my transformer (connected to a separate rectifier and regulator) which I use to power the voltmeter I'm using. If I put an LM337 there, wouldn't it be a better solution to get my -1.25V?
If the input voltage is high enough, this would be able to go up to about 36 volts (max 40 for the lm317). Remember that the higher the voltage drop over the regulator, the more heat will need to be disapated.
In many my devices where i use ac voltage, i create negative voltage by simply adding diode backwards and after that capacitor. So i think this schematic may be simplified by removing c3 and d4. D3 connected then to source. 2 diodes as reference + resistor, i would replace by trimmer, say 10kohm. But then weve got classic power supply. Anyway its simple when you have ac. Using dc its little bit more complicated. Previously i said that 2 diodes in series makes output closer to 0v thinking about dc at input. Hmm interresting thing, that these 2 diodes here are just reference. Which also explains how voltage references work. Cheers G P. S. I forgot to reply you about these diodes and about diode characteristics, but i think it may work where you have some margin.
Thanks for the comment. Yes, with a DC supply it gets messy. There are better regulators for this but if you realty want to get precise you use a transistor, voltage reference and op-amp to make your regulator (see HP power supply schematics for the 1970's). I find there is always a trade-off between simplicity and accuracy. The great thing about having simulators is that we can experiment and play, without cost! I forgot to mention, the reason for the decoupling capacitor is so this will work with a full-bridge rectifier as well. Scott
@@ScottCard when i was young i had idea to create regulated power supply with digital memory. There was lm350, because 5a instead of i think 3a in lm317. But controling elements were transistors connected simply to bit lines in static ram. Another things were simple. I do not remember how voltages were entered, i mean by buttons or what. I think i used buttons there, because also then i made diode matrix keyboard. Also then i discovered how to build one cell memory with few 4013 (i think), to keep a digit. 8 bit gives 25,5v suplly voltage with step 0,1v - not so bad, of course output voltages may be another than 25,5 with another resolutions. For example 12,5v will have 0,05v step. Hehe. I had crazy ideas. Today we may buy everything. Cheers Greg
@C_04 Yea. With the chip shortage these days, I have had to revert to first principals and build discrete circuits for many 'common' parts. Have even considered some 74xx series IC to get the job done😬. You project sounds interesting, defiantly a low cost option (could even use a decade counter to toggle through the steps). I have used a DAC in a micro before, that works well but the output stage is where the fun is! Thanks, Scott
@@ScottCard hmm it is possible that i considered decade couters connected to carry digits. Dac's because were expensive i made with resistors, in fact used that way in my ps idea. Logic circuits world is somehow wonderful world. Of course when you lack in programming. I hope i will learn soon enough how to programm, before i will be unable do something. I had some cheap electronic store near my electronic school. There were mainly some circuits and logic ics. I purchased them in tens hehe. Even today they really cheap, but when calculate cost of pcb and overal size of circuit, energy which is needed to power this up... Sounds is rather for educational purpose. Still useable. Cheers Greg
Well, it is not an easy from a single source. What you can do however is to add a 1.5V battery to the system. You would connect the + to ground and the - to the V-adjust. Hope this helps, and thanks for the comment.
Here is a link to the file. www.northshore-innovation.ca/product-page/lm317-0-12v-regulator-file😃
I'm designing an adjustable power supply (using LM338 instead of LM317 to get 5A out of it) and I do have a second output in my transformer (connected to a separate rectifier and regulator) which I use to power the voltmeter I'm using. If I put an LM337 there, wouldn't it be a better solution to get my -1.25V?
I imagine so. Just need to tie the reference to the ground and let the negative ride lower. Let us know how you make out.
Nice, is it practically work , I m going to make lab bench power supply
Great! Let us know how you make out. I keep this small one on my desk for testing, especially during firmware programming.
can we make it 0v until 25v or more using potentiometer
If the input voltage is high enough, this would be able to go up to about 36 volts (max 40 for the lm317). Remember that the higher the voltage drop over the regulator, the more heat will need to be disapated.
In many my devices where i use ac voltage, i create negative voltage by simply adding diode backwards and after that capacitor. So i think this schematic may be simplified by removing c3 and d4. D3 connected then to source. 2 diodes as reference + resistor, i would replace by trimmer, say 10kohm. But then weve got classic power supply. Anyway its simple when you have ac. Using dc its little bit more complicated. Previously i said that 2 diodes in series makes output closer to 0v thinking about dc at input. Hmm interresting thing, that these 2 diodes here are just reference. Which also explains how voltage references work.
Cheers
G
P. S. I forgot to reply you about these diodes and about diode characteristics, but i think it may work where you have some margin.
Thanks for the comment. Yes, with a DC supply it gets messy. There are better regulators for this but if you realty want to get precise you use a transistor, voltage reference and op-amp to make your regulator (see HP power supply schematics for the 1970's). I find there is always a trade-off between simplicity and accuracy. The great thing about having simulators is that we can experiment and play, without cost!
I forgot to mention, the reason for the decoupling capacitor is so this will work with a full-bridge rectifier as well.
Scott
@@ScottCard when i was young i had idea to create regulated power supply with digital memory. There was lm350, because 5a instead of i think 3a in lm317. But controling elements were transistors connected simply to bit lines in static ram. Another things were simple. I do not remember how voltages were entered, i mean by buttons or what. I think i used buttons there, because also then i made diode matrix keyboard. Also then i discovered how to build one cell memory with few 4013 (i think), to keep a digit. 8 bit gives 25,5v suplly voltage with step 0,1v - not so bad, of course output voltages may be another than 25,5 with another resolutions. For example 12,5v will have 0,05v step. Hehe.
I had crazy ideas. Today we may buy everything.
Cheers
Greg
@C_04 Yea. With the chip shortage these days, I have had to revert to first principals and build discrete circuits for many 'common' parts. Have even considered some 74xx series IC to get the job done😬. You project sounds interesting, defiantly a low cost option (could even use a decade counter to toggle through the steps). I have used a DAC in a micro before, that works well but the output stage is where the fun is!
Thanks,
Scott
@@ScottCard hmm it is possible that i considered decade couters connected to carry digits. Dac's because were expensive i made with resistors, in fact used that way in my ps idea. Logic circuits world is somehow wonderful world. Of course when you lack in programming. I hope i will learn soon enough how to programm, before i will be unable do something. I had some cheap electronic store near my electronic school. There were mainly some circuits and logic ics. I purchased them in tens hehe. Even today they really cheap, but when calculate cost of pcb and overal size of circuit, energy which is needed to power this up... Sounds is rather for educational purpose. Still useable.
Cheers
Greg
hi. ı have a LM317 and I have power supply of 9V DC. How could I get -1.25 V from DC in real life. I do not have AC source.
Well, it is not an easy from a single source. What you can do however is to add a 1.5V battery to the system. You would connect the + to ground and the - to the V-adjust. Hope this helps, and thanks for the comment.
How do you implement the same idea on a variable power supply on multisim?
Hi, sorry I do not use Multisim, but the design should still work.
Hi dear ,nice concept to be discussed. can you please add the proteus file link
Here is a link to the file. www.northshore-innovation.ca/product-page/lm317-0-12v-regulator-file😃
@@ScottCard thanks
You are welcome!
Awesome video! Thanks for sharing.
You are welcome, I am glad you enjoyed it! Than you for leaving a comment. 😎