Wanted Circuit Design Ideas To Construct CC CV Power Supply

[audio]

Hi to all,

               In my previous Thread I have asked Replacement component for General purpose power supply design published by NI. It is  really Hard to construct of such high current. So I decided to construct Variable power supplies of lower Ratings(1.2V to 30 V and Minimum Current to 1 Amps). Hence maximum power obtained from the RPS 30Watts and Maximum power dissipation rather would be 30W.

                      I went through the websites for Regulator datasheets which could solve my Problem. I found 317 IC could be used up to 40V.
but the problem is how to control current with these three terminal voltage Regulators?

                               Ideas given in data sheet require lower resistance between adj and vout terminal. such low resistance variable pot of High wattage is required to control current . What could be the better alternative for this Problem? How to control Current in three terminal Voltage regulators? Thank you all..

[Zero999]

Using the LM317 is a bad idea. The current limiting circuit requires a voltage overhead of at least 4V, which is added to the usual 3V, of the voltage regulator.

Use the LT3081 which doesn't have any of these limitations and the current limit adjustment potentiometer doesn't pass the load current.
https://www.analog.com/media/en/technical-documentation/data-sheets/3081fc.pdf

[audio]

what should be the power rating of 1 ohm resistor in the circuit?

[WattsThat]

P = I^2 * R so max dissipation is about 1.2 watts.

Since this is a current sensor, you want the resistance to be as stable as possible so your current limit does not drift when you’re working with higher current/power. That power dissipation means temperature rise so a larger resistor with a low temperature coefficient is desired.

I’d go with a 1%, 10W part with +/-50ppm TCR.

https://www.digikey.com/en/products/detail/vishay-dale/RS0101R000FE73/1166364

[audio]

I came through web about classic 723 which could do both voltage and current regulations. But the datasheet doesn't contain any circuit which could control voltage and current control. Is there any method to construct to control 2 to 30v and minimum current to 1 amps?

[Zero999]

I came through web about classic 723 which could do both voltage and current regulations. But the datasheet doesn't contain any circuit which could control voltage and current control. Is there any method to construct to control 2 to 30v and minimum current to 1 amps?

The LM723 has the same problem as the circuit in your original post: the current sense resistor needs to be variable. It also has very rough current limiting, which is dependant on the V_BE drop of an internal transistor, which varies with temperature.

The circuit you posted in your second reply is the right way to do it. Use a 2W 1 Ohm resistor, so it doesn't get too hot.

[mvs]

LT3080/LT3081 are quite expensive components, LM723 is obsolete...

Look for a discrete solution with separate voltage reference, error amplifiers and pass element.

[audio]

Yes. Truly Said, cost of LT3080 is much Higher than transformer.

It would be better to check with Opamp based design for Example 358 ,Dual Opamp. One would be used for Voltage Regulation and other opamp would be for current regulation. Similar to my previous Thread.

You can share design ideas with Opamp based CC CV power Supply circuit..

Thank you..

[Kleinstein]

The LT3080 and LM317 are voltage regulators meant to power a fixed circuit. They have a limited accuracy because the power dissipation heats up the circuit and reference. In addition they have limitations on the load impedance and may oscillate with the wrong output capacitance / load, which can be annoying for something like a lab supply where you don't know upfront what is connected.

For better accuracy it is preferred to have the power transistor(s) and control separate, like with the LM723. There is no much magic inside the LM723 and one can as well use 2 OPs and a voltage reference for a similar circuit. There are plenty of circuits out there in the web - just too many poor ones, so that it is not that easy to find a working one.
AFAIK the xx723 are not yet fully obsolete - just some manufacturers and versions.

Don't get fooled by first page numbers: they often don't apply at the same time and only under favorable conditions. So no 40 V and 1 A from the LM317, and the LM723 has 37 V maximum input voltage (not output) which makes 30 V out already not so easy.

[xavier60]

Yes. Truly Said, cost of LT3080 is much Higher than transformer.

It would be better to check with Opamp based design for Example 358 ,Dual Opamp. One would be used for Voltage Regulation and other opamp would be for current regulation. Similar to my previous Thread.

You can share design ideas with Opamp based CC CV power Supply circuit..

Thank you..

Is your memory really that short?

[audio]

Exatly, This was in my head.

many Circuit were present but to find the one which is working needs skill with opamp circuit.

Choosing 723 IC is better but Variable Current limiting will be a problem .

[xavier60]

Exatly, This was in my head.

many Circuit were present but to find the one which is working needs skill with opamp circuit.

Choosing 723 IC is better but Variable Current limiting will be a problem .

The LM723 has an opamp inside which needs compensation like any other opamp.
The LM723's opamp does not work properly with its input close to ground voltage, meaning that PSU design wont regulate down to zero volts
without having to add tricks to the circuit.
 For good CV/CC operation, two opamps are needed. There is a circuit about that has a pair of differential transistors tacked onto the LM723 for CC.

[exe]

The LT3080 and LM317 are voltage regulators meant to power a fixed circuit. They have a limited accuracy because the power dissipation heats up the circuit and reference. In addition they have limitations on the load impedance and may oscillate with the wrong output capacitance / load, which can be annoying for something like a lab supply where you don't know upfront what is connected.

Well, the datasheet for lt3080 has performance curves. Looks like reference current is is ~600ppm/c, which is about 2.5% for 50C raise. I'd say it's fine for an analog power supply.

As of oscillation and stability, what makes lt3080 special? Just put 10uF ceramic cap to the output. LT3080 is a very fast regulator, so I can imaging stability issues may happen (got that with a DIY load), but that's the price to pay for speed.

LT3080 is expensive, but if you take into account time and effort engineering a cheaper circuit it's actually not that bad. Unless taking an existing design. Also, I bet there will be much more parts.


PS this is my LT3080 power supply https://www.eevblog.com/forum/beginners/please-critique-this-linear-psu/ , and this is a picture of it: https://www.eevblog.com/forum/testgear/show-the-homemade-equipment-you-are-using-now/msg1984904/#msg1984904 . It has a horrible temperature drift thanks to voltage set circuitry. The problem is not with lt3080, but with lm334 which has a horrible drift. It is so bad that the datasheet proposes to use lm334 as a temperature sensor...

[audio]

"Their is a circuit about that has a pair of differential transistors tacked onto the LM723 for CC" ,where could i find the circuit on WEB?

[xavier60]

I saw it in a thread on this forum lately.

[xavier60]

https://www.eevblog.com/forum/beginners/philips-pe-1535-power-supply-modification/msg3096175/#msg3096175

[themadhippy]

Might be worth while watching daves series on designing a power supply
https://www.youtube.com/playlist?list=PLBF35875F73B5C9B5

[exe]

BTW, if you want circuit ideas, you can check this thread: https://www.eevblog.com/forum/projects/how-to-design-fast-bench-supply-with-cc-and-cv/ .

If I was building a power supply, I'd go with xavier60's power supply: https://www.eevblog.com/forum/projects/how-to-design-fast-bench-supply-with-cc-and-cv/msg3010016/#msg3010016 . Or any other power supply from a reputable forum member .

You can design your own power supply, but beware that proper design takes a lot of efforts and knowledge. Esp. when it comes to performance and stability. Beware of many broken and unsafe circuits on the net. Incl. on this forum. So, make sure the design is reviewed by a skilled analog engineer. Of course, skilled engineers are opinionated, but a good opinionated power supply is still a good power supply

PS Dave's power supply doesn't touch upon stability and compensation (afaik). So, while it can be inspiring to design an own power supply, you have to learn a lot more to do that. That was my experience 5 years ago when I tried to make my own PSU based on his educational videos.

[MikeK]

To answer OP's original question:

I found 317 IC could be used up to
40V. but the problem is how to control current with these three terminal voltage Regulators?

The datasheet for the LM317 contains a simple circuit for a "precision current limiter".

[mvs]

You can design your own power supply, but beware that proper design takes a lot of efforts and knowledge. Esp. when it comes to performance and stability. Beware of many broken and unsafe circuits on the net. Incl. on this forum. So, make sure the design is reviewed by a skilled analog engineer. Of course, skilled engineers are opinionated, but a good opinionated power supply is still a good power supply

The problem with skilled analog engineers is that they don't bother with basic designs. You can look for instance on PSL3604. It is a very good supply, but also very overloaded with features. At least used parts are more or less generic and cheap.
https://www.eevblog.com/forum/projects/diy-bench-power-supply-psl-3604/

[Zero999]

To answer OP's original question:

I found 317 IC could be used up to
40V. but the problem is how to control current with these three terminal voltage Regulators?

The datasheet for the LM317 contains a simple circuit for a "precision current limiter".

But it's crappy, as mentioned above.

The original poster should just go for the LT3081. It's not cheap, but is much better than faffing around. It's possible to do it with a few op-amps, but it's difficult to stabilise.

Here are a couple of threads about this. There are many more.
https://www.eevblog.com/forum/projects/oscillation-in-psu-simulation/
https://www.eevblog.com/forum/projects/how-to-design-fast-bench-supply-with-cc-and-cv/

[audio]

Just Check this Circuit For CC CV Circuit.

Will it work as intended?

[xavier60]

It can be made to work. It will have poor load regulation because the B-E drop of Q1 and the drop across R3 are outside of the CV regulation loop.
CC wont work properly at low output voltage. The CC opamp needs a voltage reference and maybe some compensation.

[Zero999]

Just Check this Circuit For CC CV Circuit.

Will it work as intended?

It can be made to work.

As you only need 1A, do away with Q1.

Connect the top of RV3 to a 1.2V reference and make R3 1.2 Ohm. If it's not critical, the LM317L can be used as a cheap voltage reference.

Add a -5V negative power supply rail for the op-amp, so its output voltage can go below zero.

[rstofer]

You can design your own power supply, but beware that proper design takes a lot of efforts and knowledge. Esp. when it comes to performance and stability. Beware of many broken and unsafe circuits on the net. Incl. on this forum. So, make sure the design is reviewed by a skilled analog engineer. Of course, skilled engineers are opinionated, but a good opinionated power supply is still a good power supply

The problem with skilled analog engineers is that they don't bother with basic designs. You can look for instance on PSL3604. It is a very good supply, but also very overloaded with features. At least used parts are more or less generic and cheap.
https://www.eevblog.com/forum/projects/diy-bench-power-supply-psl-3604/

It could be that supplies designed by analog engineers might actually work whereas most of these low-parts-count DIY supplies have fairly large warts.

The PSL3604 linked above is way up on the parts count scale.  I'm guessing that it actually works as intended.

Look at one of Dave's teardown videos and see how many components are in a commercial supply.  Dave did a teardown of the Rigol DP832 for example.  Look at the parts count!

Don't worry about 30V at 3A, see what happens at 1V at 3A given a 36 VDC source and a linear design.  Commercial supplies use custom toroidal transformers for this very reason.  It is unlikely that the hobbyist is going to source such a transformer.

[exe]

Don't worry about 30V at 3A, see what happens at 1V at 3A given a 36 VDC source and a linear design.  Commercial supplies use custom toroidal transformers for this very reason.  It is unlikely that the hobbyist is going to source such a transformer.

Well, a custom toroid costs 60-100euro, which is I'd call affordable (even less if bought from a "third-world" country). Apart from that, what I do is I just use a pre-reg, like the one from here: https://www.eevblog.com/forum/projects/very-low-noise-preregulator-for-benchtop-power-supply/ .

If one wants to go really cheap, one can make a custom tranny by themselves. Or add extra windindings to an existing toroid.

[mvs]

Dave did a teardown of the Rigol DP832 for example.  Look at the parts count!

Parts count is not a measure of how good a design is and Rigol DP832 is a good example of this.

MOSFET in output stage of DP832 can operate outside of SOA in case of short output from high voltage levels.

Regular MOSFET are not designed for linear applications. Liv has limited Vds by design in his PSL3604 to around 12-14V. Rigol just put a beefy MOSFET (CEP80N15) with a hope that it would withstand Vds of around 40V during discharge of bulk caps (tap switching should be fast on this PSU).
Large gate capacitance of this MOSFET leads also to stability issues, so they put 1000uF caps on the outputs.

[Zero999]

Dave did a teardown of the Rigol DP832 for example.  Look at the parts count!

Parts count is not a measure of how good a design is and Rigol DP832 is a good example of this.

MOSFET in output stage of DP832 can operate outside of SOA in case of short output from high voltage levels.

Regular MOSFET are not designed for linear applications. Liv has limited Vds by design in his PSL3604 to around 12-14V. Rigol just put a beefy MOSFET (CEP80N15) with a hope that it would withstand Vds of around 40V during discharge of bulk caps (tap switching should be fast on this PSU).
Large gate capacitance of this MOSFET leads also to stability issues, so they put 1000uF caps on the outputs.

A 1000µF output capacitance makes it useless in constant current mode. It also makes it easier to fry things: set the current limit to the lowest possible, say 20mA, so the circuit being tested doesn't fry if there's an error, but oh no, there's around 100A of available instantanious current available from the output capacitor.

[mariush]

If you feel up to it, you could recreate the circuit of an old HP power supply, I've attached the PDF with details and schematic.

You could simplify the circuit and have separate transformer for the control section and a separate transformer for the power stuff and drop the lcd displays completely or replace them with your own modules.

I've also attached a 18v 2A power supply schematic ... it's also simple, just ignore the top part of the pdf with the 2 multimeter chips that just show the voltage and current, the bottom part is all the power supply.  The values are in schematic... they use the standard notation as in 821 = 82 x 10^1 = 820 ohm  , vr3 and vr4 are potentiometers, k1 is 12v mechanical relay (you can simplify schematic with smaller transformer and not use 2 secondary windings) .. and so on..

And the 3rd article is from a magazine, contains schematic list, part values, circuit board drawings etc

Also, remember that if you need some funky power ratings for resistors, you can parallel resistors to spread the dissipation over multiple resistors. For example, if you need 1.2 ohm, maybe get 4 x 4.7 ohm resistors in parallel and you'll get very close to 1.2 ohm

[Simon]

Dave did a teardown of the Rigol DP832 for example.  Look at the parts count!

Parts count is not a measure of how good a design is and Rigol DP832 is a good example of this.

MOSFET in output stage of DP832 can operate outside of SOA in case of short output from high voltage levels.

Regular MOSFET are not designed for linear applications. Liv has limited Vds by design in his PSL3604 to around 12-14V. Rigol just put a beefy MOSFET (CEP80N15) with a hope that it would withstand Vds of around 40V during discharge of bulk caps (tap switching should be fast on this PSU).
Large gate capacitance of this MOSFET leads also to stability issues, so they put 1000uF caps on the outputs.

A 1000µF output capacitance makes it useless in constant current mode. It also makes it easier to fry things: set the current limit to the lowest possible, say 20mA, so the circuit being tested doesn't fry if there's an error, but oh no, there's around 100A of available instantanious current available from the output capacitor.

i blew an LED up this way trying to run it in constant current on a supply.

[Zero999]

723 obsolete!,your  taking out of your ass!!

No need to be like that. The DIP and SOIC packages are no longer manufactured. The metal can variant is still made but is horrifically expensive, so it might as well be obsolete. The LT3081 is much cheaper, than the LM723.

Dave did a teardown of the Rigol DP832 for example.  Look at the parts count!

Parts count is not a measure of how good a design is and Rigol DP832 is a good example of this.

MOSFET in output stage of DP832 can operate outside of SOA in case of short output from high voltage levels.

Regular MOSFET are not designed for linear applications. Liv has limited Vds by design in his PSL3604 to around 12-14V. Rigol just put a beefy MOSFET (CEP80N15) with a hope that it would withstand Vds of around 40V during discharge of bulk caps (tap switching should be fast on this PSU).
Large gate capacitance of this MOSFET leads also to stability issues, so they put 1000uF caps on the outputs.

A 1000µF output capacitance makes it useless in constant current mode. It also makes it easier to fry things: set the current limit to the lowest possible, say 20mA, so the circuit being tested doesn't fry if there's an error, but oh no, there's around 100A of available instantanious current available from the output capacitor.

i blew an LED up this way trying to run it in constant current on a supply.

The risk can be minimised by connecting the LED first, then applying power.

I've being toying with the idea of making a CV CV power supply, with a fast current limit. The key is to use minimal, ideally no output capacitance, but it compromises stabilty and operating in the CV region, which is where it'll spend most of its time. On the other hand, most circuits which require good transient response, will already have decoupling capacitors, so it's less important than most think.