Linear Power Supply

[Umma]

Hello ronv
how are you
please send that my mail received or not
i think that my mail not received you
thanks

 

[ronv]

15 volt 2 amp

Hi Umma,

I get your e-mails but they have no words. I don't know what is going on.
Here is a schematic for a better power supply. It is 1.2 to 15 volts at 2 amps. Much cheaper and cooler than the 30 volt 5 amp one. Lets get some comments and see what happens. If you must have the 30 volt one we can do a switcher out of transistors and op amps that would be much better than the big expensive linear one.

 

[Umma]

Hello ronv
how are you
i sent about 4 E-mail but not know that why have no words.
i tell first please send only voltage regulator
very very thanks ronv

 

[ronv]

15 volt 2 amp

How is the circuit above?

 

[Umma]

Hello ronv
how are you

only voltage regulator discuss
then current limiter circuit add
then final whole power supply discuss

please send first only voltage regulator

 

[EN0]

At 0 volts (or close to it) the regulator is going to have to dissipate 150 watts of power... A simple buck switch mode supply before the linear regulator can help this dramatically. But if you can't tell by the odd English that the user said they designed a 0-30V 5amp linear supply but they want us to supply the diagrams that something might be wrong with the original posters intentions...

That being the case, I think he should use resistors to dissipate the power before hand.

 

[ronv]

voltage regulator

Here is voltage regulator without current limit.

 

[Sceadwian]

EN0, a power resistor works fine for that scenario, it also wastes 150 watts of heat still! That's no small amount of power to have to do something with. Easy Bake ovens cook on only a 100 watt light bulb, and my soldering iron is only 35 watts.. It's not just wasteful of energy it's poor design practice given what's easily achievable using a switch mode pre-regulator.

 

[indulis]

You have no "series pass element"... or am I blind?

 

[Umma]

Hello ronv
how are you
in this circuit the +ve lead is common
1. what is the advantage of this if -ve lead is common then what advantage ad disadvantage

2. Out put volatge formula

3. if i use comparator lm339 instead of lm324 op amp then please tell me that any problem in this circuit

4. feed back resistor that connect for gain control not necessory for lm324

5. the output voltage 1.25-15v if i want output voltage 0-15V then change in circuit

6. please tell me that how these transistors works

thanks a lot ronv

 

[ronv]

Linear supply

Hi Umma,
I have attached another set of schematics with a few part numbers so we can talk about the same part. The 2N3055 pass transistor was unlabled in the last schematic. There are a couple of advantages to regulating the negitive rail. To me the most important is that it is easy to measure current in the low side so you don't need a special circuit to measure current in the high side. This was more important when we wanted current adjust. It also uses NPN transistors which are easier to find and cheaper for the same performance. It also makes the case of the 2N3055 ground so you can use the metal box (if you use a metal box) as the heat sink.
The voltage at the output is the same as the voltage at pin 3 of the 324 plus vref (1.25v). This is the same reason 1.25 volts is the lowest it can go. Let me think if there is an easy way to get the output to go to zero. Maybe someone on the forum has some ideas.
I'm not sure which resistor you are talking about in the feedback of the 324 but I will guess you are talking about R1. This is just to keep the adjustment from going so high that the ripple will show up at the output.
I don't think it is a good idea to change the 324 with a 339 because the circuit may become unstable with the higher gain/bandwidth.
The TIP 122 is a darlington transistor, so by the way it is hooked up it has very high gain (about 1000) so it amplifies the output of the 324 so there is enough power to turn on the 2N3055. The 2n3055 is just a big power transistor that is capable of driving 2 amps into the load without getting to hot. The 2N2222 is there to prevent a short circuit from burning things up. If the current goes above about 3 amps it will shut off the output.

 

[indulis]

There are a couple of advantages to regulating the negitive rail. To me the most important is that it is easy to measure current in the low side so you don't need a special circuit to measure current in the high side.

To monitor the current in the high side is just as easy.

Without a negative supply, how are you planning on dealing with the Vce sat voltage of the 3055 and the voltage developed across the current sence resistor?

 

[ronv]

supply

Opps! 2 different ground symbols. The ground at the botton of R2 should be the same as the - output. This puts the drops in the loop. If you have an easy high side solution please post it. Most good ones I have seen are specific ic's not readily available in Pakistan. Any ideas for getting the output to 0 volts?

 

[bountyhunter]

Hi Umma,

I don't think it is a good idea to change the 324 with a 339 because the circuit may become unstable with the higher gain/bandwidth.

I think it's a REALLY bad idea t replace a 324 with a 339 because the former is a unity-gain stable op amp and the latter is a COMPARATOR.


FYI: you waste a lot of output voltage swing driving a three NPN darlington from an op-amp. You may want to use a medium power PNP to drive the base of the 2N3055 and then drive the PNP pulling down.

 

[indulis]

How about something along these lines...

also, Q4 in the upper drawing need a base to gnd resistor...

 

[ronv]

power supply

The short circuit protection in the top drawing is good, but how do you implement the adjustable current limit like the bottom one on the high side.

I think you may be right about needing the darlington. We started out with a lot more current where we needed the gain.

 

[indulis]

...but how do you implement the adjustable current limit like the bottom one on the high side.

With a diff amp.

 

[bountyhunter]

With a diff amp.

It's true you can use a diff amp on the top rail for sensing but I have always found that to be less practical for simple reasons:

1) You usually develop very small voltage drops across the current sense resistor (like 50 - 100 mV) to reduce power dissipation there and save wasted voltage drop.

2) This means the diff amp has to be a pretty expensive and precise one to avoid adding serious gain errors to that voltage. The cheap op amps (like I use) have several mV of input offset voltage).

If you use negative rail sensing you don't need a diff amp to level shift and that eliminates the added error of that gain stage.

 

[indulis]

...but I have always found that to be less practical for simple reasons:

1) You usually develop very small voltage drops across the current sense resistor (like 50 - 100 mV) to reduce power dissipation there and save wasted voltage drop.

The I²R and voltage loss would be the same be it in the "top" or "bottom".

2) This means the diff amp has to be a pretty expensive and precise one to avoid adding serious gain errors to that voltage. The cheap op amps (like I use) have several mV of input offset voltage).

$$$$$ goes along with performance. To do it "properly" in the low side, you would need a diff amp a well.

 

[bountyhunter]

The I²R and voltage loss would be the same be it in the "top" or "bottom". .

You completely missed the point. I was saying that power considerations make the source voltage low which requires the gain blocks after it to have high precision. You want to use as few amp stages as possible after the sense resistor.

$$$$$ goes along with performance. To do it "properly" in the low side, you would need a diff amp a well.

No, you certainly don't. You use the inside voltage point (negative side of the resistor in the negative return line) as a negative going signal to feed directly to an error amplifier for current control or current limiting. The level shifting diff amp is not used. believe me, I have deigned a few dozen of these.

Here is an example of the design concept. U1A is the current error amp. It senses the voltage developed across the 0.25 Ohm resistor in the ground return lead. If you had a high side resistor, you would need an additional diff amp to feed the error amp and the gain errors of the diff amp would be additive.