Adjustable DC Regulated Power Supply (0 - 30 V 2 mA - 3 A)With Digital Control

[Sideshow Bob]

sir i can't understand how to unlock password protected zip file

The name of this forum is sonsivri. Try that

[manglam]

 

The name of this forum is sonsivri. Try that

i m so stupid....

[zac]

I purchased a few of these power supplies recently and found they were surprisingly good, especially for the price.  They also include the unusual capability of limiting power output.  The 0-30V 5A version was US$60 and the 10A version was $100 (but 10A version is sold out).  I haven't tried the USB software remote control features yet. 

Model: HM305P/HM310P

Output Voltage: 0-30V four digit display

Output Current: 0-5A four digit display

Power Effect:

--Voltage:≤ 0.1%+5mV

--Current: ≤ 0.03%+3mA

Load Effect:

--Voltage:≤ 0.1%+5mV

--Current: ≤ 0.01%+10mA
 Ripple:

--Voltage:≤ 20mVrms

--Current:≤ 100mArms

Display Resolution:

--Voltage: 10mV

--Current: 1mA

https://www.amazon.com/gp/product/B07SPYF8ZW/

[sadman]

I found on a Russian forum very well optimized design he also added voltage and current regulation through voltage reference  based on TL 431 the author of the power supply also publish a article in Russian magazine prakticka elektronika scan copy of article attach a youtuber made this supply and shared his experience link given.

https://www.youtube.com/watch?v=M36MDYnTRRw
  

[bobcat1]

HI All

Here is analog device latest 75 Watt, Single-Output Bench-top tracking Power Supply design!!!!

you can buy the board , add a micro , or just use a potentiometer to adjust current and voltage

https://www.analog.com/en/design-center/reference-designs/circuits-from-the-lab/cn0508.html?ADICID=some_NA_P328165_LinkedIn_IMG_CN0508

All the best

bobi


 

[Renix]

I have built a couple of the RIDEN RD6006 power supply kits using their box and my choice of switching supply to put in the box. They work as advertised for the application I wanted. (Initially as a current controlled supply for anodizing aluminum.) I haven’t actually looked at the ripple output, but I suspect it is mostly determined by the SMPS I selected.

The price on BG is ~$67USD,  so after a case & SMPS, I’m not sure the cost matches those complete supplies in earlier posts from Amazon. I think I paid a good deal less a year or so ago.

The RD6006 is here:
https://www.banggood.com/RIDEN-RD6006-or-RD6006-W-Digital-Control-Switch-Adjustable-Power-Supply-DC-Stabilized-Power-Adapter-Buck-Module-Monitoring-Power-Supply-p-1587151.html?rmmds=detail-left-hotproducts&cur_warehouse=USA&ID=6279943

One thing I always ask myself before starting a project like this is….
Do I want to design power supplies or do I want to use power supplies? An honest self-answer can save a lot of time. This also applies in many other areas. Some years ago I had a need for a CNC mill to do a paying project. I spent some time evaluating all of the “kit” stuff available for various bare bed mills. In the end I bought a complete mill from Tormach for the job. Paid off the mill with work in < 6 months and never looked back. I’m glad I realized early that I was a user not a builder of mills!

[kreutz]

I found on a Russian forum very well optimized design he also added voltage and current regulation through voltage reference  based on TL 431 the author of the power supply also publish a article in Russian magazine prakticka elektronika scan copy of article attach a youtuber made this supply and shared his experience link given.

https://www.youtube.com/watch?v=M36MDYnTRRw
  

The main problem of this design and derivatives, including similar power supplies with 723 ic, is the negative auxiliary summing voltage ( used to achieve a 0 volts output). The zener diode determines the whole temperature coefficient of the output voltage.
 
An improved option would be to use a low-dropout regulator 3.3volt or 5 Volts IC instead of that zener.

Regards,

kreutz

[Sideshow Bob]

The main problem of this design and derivatives, including similar power supplies with 723 ic, is the negative auxiliary summing voltage ( used to achieve a 0 volts output). The zener diode determines the whole temperature coefficient of the output voltage.
 
An improved option would be to use a low-dropout regulator 3.3volt or 5 Volts IC instead of that zener.

Regards,

kreutz

During the worst Covid boredom I did make a discrete LM723 type power supply.
http://www.sonsivri.to/forum/index.php?topic=68701.msg199426#msg199426
However I did not come around to build a working model of it. But the simulation show that I at least was on the correct track (perhaps)

[Wizpic]

During the worst Covid boredom I did make a discrete LM723 type power supply.
http://www.sonsivri.to/forum/index.php?topic=68701.msg199426#msg199426
However I did not come around to build a working model of it. But the simulation show that I at least was on the correct track (perhaps)

I did play with your design and like you say it works well in simulation, The last time I worked on mine I have it all working laid out all the PCB’S got all parts here just need to get back into it and order the PCB’s For it and finish off the software for it. If I get time I may have ago at building and playing around with your design in the link you posted. Need to give myself a kick up the back side and get motivated

[kreutz]

Please, see: https://www.youtube.com/watch?v=2w4cR8D8WJ4

[Sideshow Bob]

Please, see: https://www.youtube.com/watch?v=2w4cR8D8WJ4

I agree with the author in this video. The idea is kind of correct. But poorly implement. I have built and used many LM723 based adjustable power supplies. The fact that have not been able to go lower than two volt has not bothered me at all. 

[PM3295]

I am sure this problem can easily be fixed by dropping in a standard voltage (2.5 V) TL431 in place of the zener. Alternatively, using a low-voltage  regulator as mentioned. As long as the 431 shunts at least 1 mA, it will regulate very well.

[Sideshow Bob]

I am sure this problem can easily be fixed by dropping in a standard voltage (2.5 V) TL431 in place of the zener. Alternatively, using a low-voltage  regulator as mentioned. As long as the 431 shunts at least 1 mA, it will regulate very well.

Not that the LM723 is a bad regulator. But you can find a lot modern linear regulators that can go down to zero volt

[PM3295]

In its time, we used the 723 frequently in RF VCO designs because of its very low noise properties, which can help reduce phase and amplitude noise. The noise output of the chip is just a few uV's (100 - 10 kHz BW) with the recommended Cref capacitor. This noise can be reduced even more by using an external PNP pass transistor with the 723. The idea is that the noise of the PNP will be lower than the internal NPN follower.

[Checksum8]

One way to get a zero volt output. Use a ICL7662 or LT1054 negative charge pump to get the negative reference voltage to feed your zener or negative regulator. A ICL7660 would work but does not like more than +10v

[Sideshow Bob]

Eh no...with that approach you will ending up injecting high frequency noise into your setup. The high frequency noise will be spawn from the internal oscillator charge pump. What you could do is to use a second low ampere transformer to create a the negative voltage and use a proper low noise voltage reference. For the negative bias. In a hobbyist setup this will be perfectly fine. Or just jeez realize that the LM723 although not a bad regulator ( it make even me feel young). It is outdated if you really need need power supply that need to go below 2 volt. Just use a more modern approach. It is plenty of ICs that can do the job job.

[kreutz]

............ Just use a more modern approach. It is plenty of ICs that can do the job job.

Some examples, please.

[Sideshow Bob]

Some examples, please.

Here are some suggestions. You should be able to insert a pass transistor(s) in all of these I think. They do not go down to zero volt except one I think. But They are able to go down to in the range of 0.6 to 0.4 volt. Do you really need to have lower output than that? Non of them are super expensive either
https://www.analog.com/en/parametricsearch/11459#/p5576=Output%20Current%20Monitor&p5574=36|45

[Manuel]

The one attached can be easily improved providing the right Vin with the right Current and the right heatsink.

Based on 2N3055 or TIP3055 can arrive at 7A and I suggest 35V as VMax.

I suggest to add 10K load on the output.

Just for info: From ST the TIP3055 is easier to be found.

'njoy & take care,

X!

[Sideshow Bob]

The one attached can be easily improved providing the right Vin with the right Current and the right heatsink.

Based on 2N3055 or TIP3055 can arrive at 7A and I suggest 35V as VMax.

Dude it is because of the shortcomings of this kind of voltage "regulator" . That they come up with the monolithic, or integrated, linear voltage regulator ICs eons ago. 

[PM3295]

One way is to use the higher current LM338T and combine it with some op-amps to form a current feedback control loop.
The circuit below show one possibility of implementing this idea. The current scaling gives 1V/Amp so the current limit setting can be  monitored on TP1 and the instantaneous current feedback on TP2. The scope traces shows the transient response with a 3 Amp current pulse load on the output.

R14 is not needed, but Proteus kept crashing without this in place. Something with the TL431 model.

[Sideshow Bob]

One way is to use the higher current LM338T and combine it with some op-amps to form a current feedback control loop.
The circuit below show one possibility of implementing this idea. The current scaling gives 1V/Amp so the current limit setting can be  monitored on TP1 and the instantaneous current feedback on TP2. The scope traces shows the transient response with a 3 Amp current pulse load on the output.

R14 is not needed, but Proteus kept crashing without this in place. Something with the TL431 model.

Have you made this in real world or just simulated it. I have only taken a sort look at it. But I have problems with your current sense setup. Mainly that that the voltage on the current sense resistor seems to have lower potential than the LM358 opamp grund point

[PM3295]

At this stage, it is just a design suggestion. The latest LM358's have been upgraded from the older design. You may improve the low end more by using the LMV358 (rail-rail), but it only works with 5.5 V rail, so it needs additional 5 V regulator.  With differential input configuration using the LM358, it will just limit the low current sensing to about 100 mA. Below that it becomes non-linear with about 70 mV error at zero. Naturally, using a negative supply of a few volts, will correct this error.

The diff amp needs well matched resistors of at least 1% tolerance or better for good performance. Alternatively, one of the feedback resistors should be adjustable.

[PM3295]

As a follow-up for those interested, I tested the current sense, and read-back function using a standard LM358. The circuit actually performed much better than predicted by the simulator, and I could get accurate readings down to as low as 13 mA through the 0.1 ohm sense resistor.  

[vern]

in my opinion there is a flaw in your design:
you are measuring the total current with R3, including the currents that are used by the LM358 / LM393 und the current through Q1.