My name is Andrey.I found the schematic of 12 VOLT 30 AMP POWER SUPPLY on the Internet at the following link : Circuit Exchange International (CXI)
I am not an expert - have only basic knowledge in electronic. I want to build such a power supply and I have few questions about it. May be some one can help with the answers.
One of the author's concerns is the six power transistors TIP2955. In the case of a short on the "OUT" the six or at least some of them will be burnt. Does anyone has an idea about a simple schematic solution - to add to the schematic simple short circuit protection? And another question: If the LM7812 has built in "short" and "thermo" protection, why the 1 AMP fuse to protect the OUT of the IC LM7812 needed? Thanks
I am not an expert - have only basic knowledge in electronic. I want to build such a power supply and I have few questions about it. May be some one can help with the answers.
One of the author's concerns is the six power transistors TIP2955. In the case of a short on the "OUT" the six or at least some of them will be burnt. Does anyone has an idea about a simple schematic solution - to add to the schematic simple short circuit protection? And another question: If the LM7812 has built in "short" and "thermo" protection, why the 1 AMP fuse to protect the OUT of the IC LM7812 needed? Thanks
It would pay (time and money) to examine your end use/application in more detail rather than jumping in feet 1st.
for example wouldn't an ATX power supply fit the bill for little time and money? How to Convert a Computer ATX Power Supply to a Lab Power Supply
but in any case this would be a better starting point for a linear solution with some better guidance High Current 13.8V Power Supply
for example wouldn't an ATX power supply fit the bill for little time and money? How to Convert a Computer ATX Power Supply to a Lab Power Supply
but in any case this would be a better starting point for a linear solution with some better guidance High Current 13.8V Power Supply
12 VOLT 30 AMP POWER SUPPLY
Thanks to those who replied to my Q.
In reply to DUG's statement: I understand the purpose of the both fuses. But why the LM7812 needs a fuse if it has a built in protection?
I agree with you tomchr, that a fuse protection is not reliable. That's why I asked my original Q. how to protect electronically the power transistors.
Do I actually *need* a 12 V, 30 A supply? May be not 30 A, But I need 10 or 15 amps of 12 V. I want to feed kitchen under cabinet 5630 led strips. I want to change the tiny led lights of the kitchen range hood to more powerful 12 V led lights, under the cabinet lights in the laundry and bathrooms. All these will add up to quite an amperage - I would say 10 or 15 amps. But to be on the safe side, I am thinking to build 25 or 30 amps power supply.
Thanks to infinia. I got your point. If I was a professional, I would pay money and time too. But the think is - I am not. In other words: I do not have enough knowledge to tell if it is going to work or not; and if it works, how reliable it is. What I liked in this 12 VOLT 30 AMP POWER SUPPLY - it is very simple, and if something goes out, it's easy to find and fix. And it is relatively not expensive - the power transistors on Ebay - 10pc for $8, the regulator - 3 for $2. The most expensive part would the 47,000u capacitor. If some body built it, I can assume it works. The idea of ATX computer power supply could be better, but it is a very complicated device. If it stops working I would not be able to fix it. It already happened to one of my computers and I had to change the whole power supply. Tell me your opinion about this: why you think the 12 VOLT 30 AMP POWER SUPPLY may not work and what needs to be changed. To me the idea of High Current 13.8V Power Supply is better then the Computer ATX Power Supply, again - because of simplicity.
Thanks to those who replied to my Q.
In reply to DUG's statement: I understand the purpose of the both fuses. But why the LM7812 needs a fuse if it has a built in protection?
I agree with you tomchr, that a fuse protection is not reliable. That's why I asked my original Q. how to protect electronically the power transistors.
Do I actually *need* a 12 V, 30 A supply? May be not 30 A, But I need 10 or 15 amps of 12 V. I want to feed kitchen under cabinet 5630 led strips. I want to change the tiny led lights of the kitchen range hood to more powerful 12 V led lights, under the cabinet lights in the laundry and bathrooms. All these will add up to quite an amperage - I would say 10 or 15 amps. But to be on the safe side, I am thinking to build 25 or 30 amps power supply.
Thanks to infinia. I got your point. If I was a professional, I would pay money and time too. But the think is - I am not. In other words: I do not have enough knowledge to tell if it is going to work or not; and if it works, how reliable it is. What I liked in this 12 VOLT 30 AMP POWER SUPPLY - it is very simple, and if something goes out, it's easy to find and fix. And it is relatively not expensive - the power transistors on Ebay - 10pc for $8, the regulator - 3 for $2. The most expensive part would the 47,000u capacitor. If some body built it, I can assume it works. The idea of ATX computer power supply could be better, but it is a very complicated device. If it stops working I would not be able to fix it. It already happened to one of my computers and I had to change the whole power supply. Tell me your opinion about this: why you think the 12 VOLT 30 AMP POWER SUPPLY may not work and what needs to be changed. To me the idea of High Current 13.8V Power Supply is better then the Computer ATX Power Supply, again - because of simplicity.
This cant be true, ATX PS can be had anywhere any time for not much money. free in most cases. should be reliable up to 15A. For 30A I would still buy a more modern >500 W name brand model. most simple thing > it breaks replace it'
I think you underestimate your time and cost build DIY stuff from scratch unless you have chassis / heat sinks and big power transformer in hand. I reckon the scrap value $/Kg of linear power supply costs more than a brand new 500W ATX PS.
PS abandon your 1st schematic! I doubt anyone has built that thing as drawn.
Last edited:
Look at this data sheet on page 20 at the bottom:
https://www.fairchildsemi.com/datasheets/LM/LM7805.pdf
When the short circuit transistor starts to turn on it reduces drive to the pass transistors and shifts current into the 78xx. The 78xx then limits the current by reducing the output.
Same thing on page 15 here:
http://www.ti.com/lit/ds/symlink/lm340-n.pdf
That is probably what you are looking for.
https://www.fairchildsemi.com/datasheets/LM/LM7805.pdf
When the short circuit transistor starts to turn on it reduces drive to the pass transistors and shifts current into the 78xx. The 78xx then limits the current by reducing the output.
Same thing on page 15 here:
http://www.ti.com/lit/ds/symlink/lm340-n.pdf
That is probably what you are looking for.
cuz it's a crude current dump circuit, depending on output loads (read reactive) and dynamic events like start up turn off and output shorts the helper PNP pass devices could drive high current back into the 3 terminal regulator. These current dumper schemes only really work reliably under steady state conditions.
If you go linear, there are better, cleverer options than the circuit you found, see here for example:
http://www.diyaudio.com/forums/powe...d-psu-using-pass-transistors.html#post4071001
Better regulation, lower dropout, predictable current limit, etc
Mainly, I will use the 12 Volt 30 Amp Power Supply to feed my led projects around the house. But who knows - in the future I may use it for something else.
There was an opinion that my original schematic was not going to work. Could be. Here is the link
12 Volt 30 Amp Power Supply
Found another one
1.5 to 37 Volt 30 Amp Power Supply
Similar to my original PS. Somebody modified it to be adjustable.It would make a cheap and convenient lab power supply (if it works)
And thanks for the explanation of the purpose of the fuse. It makes sense.
There was an opinion that my original schematic was not going to work. Could be. Here is the link
12 Volt 30 Amp Power Supply
Found another one
1.5 to 37 Volt 30 Amp Power Supply
Similar to my original PS. Somebody modified it to be adjustable.It would make a cheap and convenient lab power supply (if it works)
And thanks for the explanation of the purpose of the fuse. It makes sense.
I have just recently visited these two exact circuit configuration in this thread,
http://www.diyaudio.com/forums/chip-amps/261125-lm3886-problem.html#post4037783
As you read through you will find how I have discovered that both of these configuration suffer greatly in load regulation as more current is drawn form the regulator.
Since my circuit here was intended for up to 60v of input voltage I ultimately used the LM317HV basically as just a transistor and added a precision opamp and a volt reference to keep the output voltage in check.
The result was a very stable and very quiet regulator.
The final circuit as built is found here minus D1 it is not need and was only there to see what it would do and I forgot to delete it when I made the post,
http://www.diyaudio.com/forums/chip-amps/261125-lm3886-problem-2.html#post4073963
In the process of this design thread I had discovered a non-linear current limiting action of the Regulator chip itself.
You would think that if the voltage drop across the regulator itself was low that it should be able allow more current to flow through the device, But, this is not the case!!
At a higher voltage output settings, it limits the current output more so by dropping the voltage earlier than it does when it is set at a lower output voltage and drawing more current.
It is the opposite of what you would expect!!
By Using LTspice I found it to be necessary to use another driver transistor to drive the stack of paralleled output transistors in a darlington configuration in order to not exceed the current limiting factor of the regulator chip itself.
I have not built the bigger version yet but it is on my list of thing to do.
At the top of the range of the regulator circuit the amount of current available was about 400ma to 500ma according the LT spice and nearly 1 amp available at output voltages of about 20v to 25v or less.
What doesn't make since for my circuit is that with a 50v input and a 20v output I could get 1 amp to drive the transistors this is asking the LM317HV to dissipated 30watts.
And, For 50v in and 40v out I could only get .4 amps and this only comes out to 4 watts that the regulator has to dissipated.
Strange indeed?!!!!
Also it did not seem that the spice model is exact and under certain conditions the simulations generated under a heavy load.
I got some very false results that would never happen in the circuit otherwise no matter what the operating conditions were.
I am not sue if this was because of the LM317 model or the LT2057HV model that I was using at the time.
I have already provided the models for all to mess with in the thread.
Anyhow using the configuration with just one output transistor and no more than 26V input I have found this strange non-linearity to be true as well as I have mentioned in the thread.
Many years ago I have made another high current regulator circuit that was able to produce 50 amps using only 4 output device before and I had a lot of fun with it.
I don't have the exact schematic from back in 2001, But what I did was I used a LM317 for my variable stable voltage reference and fed it into an opamp feeding a IRF510 and this supplied the necessary base current to feed the four paralleled NPN output devices.
It worked great and the load regulation was good until I started drawing enough current to cause a voltage drop do to the supply's (and transformer's) internal resistance.
I used a transformer out of a 450watt UPS that I had that was burned out and didn't work as it was supposed too, but the transformer was still good.
It was a very cool and fun project, I could get nearly 50 amps out of it when the voltage drooped to about 9V.
But it would still maintain about 30 to 34 amps at 12v just fine.
FWIW
jer
http://www.diyaudio.com/forums/chip-amps/261125-lm3886-problem.html#post4037783
As you read through you will find how I have discovered that both of these configuration suffer greatly in load regulation as more current is drawn form the regulator.
Since my circuit here was intended for up to 60v of input voltage I ultimately used the LM317HV basically as just a transistor and added a precision opamp and a volt reference to keep the output voltage in check.
The result was a very stable and very quiet regulator.
The final circuit as built is found here minus D1 it is not need and was only there to see what it would do and I forgot to delete it when I made the post,
http://www.diyaudio.com/forums/chip-amps/261125-lm3886-problem-2.html#post4073963
In the process of this design thread I had discovered a non-linear current limiting action of the Regulator chip itself.
You would think that if the voltage drop across the regulator itself was low that it should be able allow more current to flow through the device, But, this is not the case!!
At a higher voltage output settings, it limits the current output more so by dropping the voltage earlier than it does when it is set at a lower output voltage and drawing more current.
It is the opposite of what you would expect!!
By Using LTspice I found it to be necessary to use another driver transistor to drive the stack of paralleled output transistors in a darlington configuration in order to not exceed the current limiting factor of the regulator chip itself.
I have not built the bigger version yet but it is on my list of thing to do.
At the top of the range of the regulator circuit the amount of current available was about 400ma to 500ma according the LT spice and nearly 1 amp available at output voltages of about 20v to 25v or less.
What doesn't make since for my circuit is that with a 50v input and a 20v output I could get 1 amp to drive the transistors this is asking the LM317HV to dissipated 30watts.
And, For 50v in and 40v out I could only get .4 amps and this only comes out to 4 watts that the regulator has to dissipated.
Strange indeed?!!!!
Also it did not seem that the spice model is exact and under certain conditions the simulations generated under a heavy load.
I got some very false results that would never happen in the circuit otherwise no matter what the operating conditions were.
I am not sue if this was because of the LM317 model or the LT2057HV model that I was using at the time.
I have already provided the models for all to mess with in the thread.
Anyhow using the configuration with just one output transistor and no more than 26V input I have found this strange non-linearity to be true as well as I have mentioned in the thread.
Many years ago I have made another high current regulator circuit that was able to produce 50 amps using only 4 output device before and I had a lot of fun with it.
I don't have the exact schematic from back in 2001, But what I did was I used a LM317 for my variable stable voltage reference and fed it into an opamp feeding a IRF510 and this supplied the necessary base current to feed the four paralleled NPN output devices.
It worked great and the load regulation was good until I started drawing enough current to cause a voltage drop do to the supply's (and transformer's) internal resistance.
I used a transformer out of a 450watt UPS that I had that was burned out and didn't work as it was supposed too, but the transformer was still good.
It was a very cool and fun project, I could get nearly 50 amps out of it when the voltage drooped to about 9V.
But it would still maintain about 30 to 34 amps at 12v just fine.
FWIW
jer
Last edited:
The other thing that I mean't to mention in the other thread but never did was that the gain curve of your typical NPN output device is very non-linear as well.
In this case using a 2N3055 or 2N3772/3 the initial current gain (hfe) can be as high 30 to 50.
But, as you start pulling more current through the device the gain can drop as low as 10 or so by the time you get above 3 to 6 amps of current draw through the collector.
This can very easily exceed the amount of current that the Three Terminal regulator's can produce, as I had stated in my last post.
And as found, This can cause the regulator to prematurely start dropping the output voltage.
Therefore it is the main reason to use another buffer transistor in a darlington configuration as suggested.
In my sim's the power dissipation for just the buffer transistor can be as high as 30 to 50 watts or more depending on the actual current load the circuit is under, and, its operating voltage range.
jer
In this case using a 2N3055 or 2N3772/3 the initial current gain (hfe) can be as high 30 to 50.
But, as you start pulling more current through the device the gain can drop as low as 10 or so by the time you get above 3 to 6 amps of current draw through the collector.
This can very easily exceed the amount of current that the Three Terminal regulator's can produce, as I had stated in my last post.
And as found, This can cause the regulator to prematurely start dropping the output voltage.
Therefore it is the main reason to use another buffer transistor in a darlington configuration as suggested.
In my sim's the power dissipation for just the buffer transistor can be as high as 30 to 50 watts or more depending on the actual current load the circuit is under, and, its operating voltage range.
jer
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.