More current from the LM317T Voltage Regulator

[MrAl]

Mr. Al, why are you counteracting the op amp? Sharing the output with ground should almost certainly burn it out. Using a capacitor is better than the trimpot.

Hi,

I am not completely sure what you mean, but from Tony's post too i can see that maybe you are talking about the 2k pot not the 20k pot?

This 2k secondary pot is to allow the user to set the 'top' voltage for the output. This setting is exactly like the original intention for the pot from the designers of the LM317 and similar. The purpose of that pot is so that in case the 'new' control circuit (op amp and related parts) do not perform as well as we would like with various loads and load and line transients, the top voltage setting will take over and limit the output just as in any more basic LM317 design. So it is a protective feature to have that second pot. It protects the load from unintended surges.

The adjustment procedure would be to adjust that 2k second pot first before connecting any load, then adjust the final voltage with the 20k pot.
For example, we might set the 2k secondary pot so we get 5.5 volts output, then set the 20k pot for 5.00 volts output.
Of course you may have to change the values of the pots for your own use, and add any series resistors you want to use to limit the output adjustment range too. For stability a small capacitor may also have to be added for a little compensation.

 

[iLoveJeeps]

I will definatly build your schematic MrAl... Thanks everyone for what you've had to say.. Im really learning alot with you guess just chatting amongst yourselves !!

 

[atferrari]

Hola MrAl

What is the function of R1 at the input of the regulator? ??

Edit to add:

I would like to control this with a small micro but I am at lost where to insert the control signal. Initially I considered replacing the signal coming from the coarse pot but not sure really.

/Edit

 

[MrAl]

I will definatly build your schematic MrAl... Thanks everyone for what you've had to say.. Im really learning alot with you guess just chatting amongst yourselves !!

Hi and welcome to the forum,

I am happy you found it interesting and hope you have fun and learn chatting about these circuits here on ETO.

Good luck with your projects.

 

[MrAl]

Hola MrAl

What is the function of R1 at the input of the regulator? ??

Edit to add:

I would like to control this with a small micro but I am at lost where to insert the control signal. Initially I considered replacing the signal coming from the coarse pot but not sure really.

/Edit

Hi,

R1 is a rudimentary current limit. Depending on the input and output voltages, that resistor limits current to a certain level. Better though would be a transistor in the output circuit that feeds back to limit current. That's easy to add too.

To control with a micro controller one way would be to use a digital pot. You would have to check on the response time of the pot however to make sure it can respond fast enough. The 'top' of the pot cant go over 5v though, so the upper scaling resistor has to be chosen so that it does not allow more than 5v to get to the pot.
Alternately, maybe a D to A converter that drives a transistor, and the transistor acts as the pot. If the D to A converter has a reference input and that input can have a wide range of input voltages, then the output of the power supply can be scaled down with a resistive divider and the output of the D to A converter can drive the inverting input.

Alternately, you may be able to take over the function of the voltage reference diode with a D to A converter or perhaps a PWM modulated filter circuit. I would prefer the D to A converter though as it naturally functions as a true voltage memory as well as being adjustable. I would say some min voltage like 0.5v, and max maybe 5v. Together with the output voltage divider, you should be able to get a wide range of output voltages.

One thing you have to remember though, is that a micro controller will slow down the speed of response, and it could be by a lot if the micro is a low end device. Many of them will only read the analog input with a frequency of around 50kHz, while a purely linear circuit will be much faster. This will lead to a little output ripple similar to a switching power supply, and the speed of response to a change in input voltage wont be as fast either.

The transistor current limit circuit is quite simple if you want to use that. The current set for that is not very simple however, so it is best for a single set point current limit.

 

[atferrari]

To control with a micro controller one way would be to use a digital pot. You would have to check on the response time of the pot however to make sure it can respond fast enough. The 'top' of the pot cant go over 5v though, so the upper scaling resistor has to be chosen so that it does not allow more than 5v to get to the pot.
Alternately, maybe a D to A converter that drives a transistor, and the transistor acts as the pot. If the D to A converter has a reference input and that input can have a wide range of input voltages, then the output of the power supply can be scaled down with a resistive divider and the output of the D to A converter can drive the inverting input.

Alternately, you may be able to take over the function of the voltage reference diode with a D to A converter or perhaps a PWM modulated filter circuit. I would prefer the D to A converter though as it naturally functions as a true voltage memory as well as being adjustable. I would say some min voltage like 0.5v, and max maybe 5v. Together with the output voltage divider, you should be able to get a wide range of output voltages.

One thing you have to remember though, is that a micro controller will slow down the speed of response, and it could be by a lot if the micro is a low end device. Many of them will only read the analog input with a frequency of around 50kHz, while a purely linear circuit will be much faster. This will lead to a little output ripple similar to a switching power supply, and the speed of response to a change in input voltage wont be as fast either.

Thanks for replying.

What I understand from your explanation:

The first option (painted red) incorporates (micro+DAC) into the feedback loop.

The second option (in bold) keeps the voltage reference (micro+DAC) out of the feedback loop and the third paragraph relates only to the first.

Second option seems more reasonable. After all why to involve a micro in something that the regulator or even the op amp are naturally able to do?

Am I right?

Gacias for your time.

 

[MrAl]

Thanks for replying.

What I understand from your explanation:

The first option (painted red) incorporates (micro+DAC) into the feedback loop.

The second option (in bold) keeps the voltage reference (micro+DAC) out of the feedback loop and the third paragraph relates only to the first.

Second option seems more reasonable. After all why to involve a micro in something that the regulator or even the op amp are naturally able to do?

Am I right? View attachment 92289

Gacias for your time.

Hi,

You mean take over the function of the voltage reference diode?
Yes that seems reasonable, but i have to say that i havent tried this yet, and keep in mind that when you do that you take control over the entire output, and that means the voltage reference will depend on measurements made with the A to D on the micro, so it will be in effect still in the feedback loop. That is, the micro makes a measurement, determines if it is above or below the set point, then adjusts the 'voltage reference', then makes another measurement, makes another determination, then adjust again. So it's a non stop measurement and control scenario where the micro does all the measurement and control.
This would be in any case, except maybe with the pot, because with the digital pot the circuit would work on it's own even if the micro does not do any more measurement, as long as the initial setting was right or at least close (like a hand turning the pot, then letting go).
So it's up to you what you want to try, and test fully. Maybe try both ways if you have the time and have any digital pots around.

 

[atferrari]

Hi,

You mean take over the function of the voltage reference diode?
Yes that seems reasonable, but i have to say that i havent tried this yet, and keep in mind that when you do that you take control over the entire output, and that means the voltage reference will depend on measurements made with the A to D on the micro, so it will be in effect still in the feedback loop. That is, the micro makes a measurement, determines if it is above or below the set point, then adjusts the 'voltage reference', then makes another measurement, makes another determination, then adjust again. So it's a non stop measurement and control scenario where the micro does all the measurement and control.
This would be in any case, except maybe with the pot, because with the digital pot the circuit would work on it's own even if the micro does not do any more measurement, as long as the initial setting was right or at least close (like a hand turning the pot, then letting go).
So it's up to you what you want to try, and test fully. Maybe try both ways if you have the time and have any digital pots around.

Thanks for replying.

Sincerely, a little bit confused. I will build and test eventually both. The diode replacement first in any case.