Lab bench power supply

[Zygimantas]

Hey guys,

I want to build lab bench power supply, I found one in internet. http://yeoldetransistor.tumblr.com/post/134262046794/a-homemade-power-supply
What you guys thinking about it ?

And I saw one in youtube with LTC3780.


I want to know whick is better ?

[Kleinstein]

The first one from the link is not good. Its linear, but two lm317 in series is not good. It is also only rather low power and might hit the thermal of the LM317 quite fast. The better solution would be the standard circuit around the LM723 with a power transistor.

The circuit from the video is a ready made switched mode supply board build into a case. It likely has quite some noise, so its more for powering motors and similar things, but not that good for sensitive circuits. Also regulation is likely relatively slow - so quite some overshoot on fast load changes. This might be a problem for some circuits.

[Zygimantas]

The first one from the link is not good. Its linear, but two lm317 in series is not good. It is also only rather low power and might hit the thermal of the LM317 quite fast. The better solution would be the standard circuit around the LM723 with a power transistor.

The circuit from the video is a ready made switched mode supply board build into a case. It likely has quite some noise, so its more for powering motors and similar things, but not that good for sensitive circuits. Also regulation is likely relatively slow - so quite some overshoot on fast load changes. This might be a problem for some circuits.

Okey, I realized what you talking about. In internet I find couple scheamtics with lm723, but I still do not really understand much about electronics, so I want to ask you, whether it will work ? https://www.google.lt/search?q=LM723+with+a+power+transistor&biw=1280&bih=918&tbm=isch&source=lnms&sa=X&ved=0ahUKEwiR2b_4nMTPAhUF2ywKHX3sDaMQ_AUICCgB&dpr=1#imgrc=WRCysOU6IiQtvM%3A http://thebackshed.com/windmill/articles/VariablePowerSupply/TheBackShed723Reg.gif

[Kleinstein]

The circuit is in principle ok, but not really good. 

There are a few strange things:
The reference filter cap should be towards GND - so the other half of the pot.

The number of output transistors depends on the needed current/ supply voltage. You may not need so many.  At higher voltage the BD137 is rather weak to drive 4 of the 3055 - it's ok for 1 or 2 or with a < 15 V supply. The emitter resistors (0.1 Ohms in the plan should be larger (0.22-0.3 Ohms) for the 3055 unless you have well matched and thermally coupled ones.

The two parallel resistors set the current limit. The pot to adjust the current limit is not working well this way, as this will allow to increase the limit all the way to infinite. There is a better way, with having the ends of the pot between the base and emitter of the output transistors. The wiper goes to pin 2 and pin 3 is at the output. This will use the pot to reduce the current, maybe to 1/10 of the maximum value.

The filter cap at the DC input is suitable for about 0.5 A if behind transformer and rectifier, the current limit and the power transistors are more for a current in the 2-8 A range.

p.s.: A better plan would be here, under #11:
https://www.eevblog.com/forum/repair/300w-inter-m-r300-audio-amplifier-repair-stuck-in-'protection'/

[Zygimantas]

The circuit is in principle ok, but not really good. 

There are a few strange things:
The reference filter cap should be towards GND - so the other half of the pot.

The number of output transistors depends on the needed current/ supply voltage. You may not need so many.  At higher voltage the BD137 is rather weak to drive 4 of the 3055 - it's ok for 1 or 2 or with a < 15 V supply. The emitter resistors (0.1 Ohms in the plan should be larger (0.22-0.3 Ohms) for the 3055 unless you have well matched and thermally coupled ones.

The two parallel resistors set the current limit. The pot to adjust the current limit is not working well this way, as this will allow to increase the limit all the way to infinite. There is a better way, with having the ends of the pot between the base and emitter of the output transistors. The wiper goes to pin 2 and pin 3 is at the output. This will use the pot to reduce the current, maybe to 1/10 of the maximum value.

The filter cap at the DC input is suitable for about 0.5 A if behind transformer and rectifier, the current limit and the power transistors are more for a current in the 2-8 A range.

p.s.: A better plan would be here, under #11:
https://www.eevblog.com/forum/repair/300w-inter-m-r300-audio-amplifier-repair-stuck-in-'protection'/

Thanks for helping me.

[oldway]

This is a good schematic of a bench power supply with LM723.

For more details, look for service manual of PE1542 Philips power supply.

[Microcheap]

I've once built this PSU: http://www.electronics-lab.com/project/0-30-vdc-stabilized-power-supply-with-current-control-0-002-3-a/  I like its performance and the fact it has current limiting, a must have feature for me. There is a forum with more than 72 pages and thousands of messages discussing some problems and improvements to this design but the true is that I have built it more than 8 years ago and it still works.
You can even get a chinese clone version on aliexpress for less than US$6 delivered: https://www.aliexpress.com/item/Adjustable-0-30V-2mA-3A-DC-Regulated-Power-Supply-DIY-Kit-Short-Circuit-Current-Limiting-Protection/32674464028.html.

[Kleinstein]

That circuit / kit is in principle a good circuit, but has a few limitations:

The main one it that the OPs voltage rating limits the safe operation to about a 20 V output voltage with something like a 20 V transformer. With slight adjustments one might get a little higher, but no 30 V without going beyond maximum ratings of the cheap OP. The original circuit (before cost cutting) used a more expensive OPs, that can stand a considerably higher supply. Just don't be surprised of you blow the OP if you try with higher voltage.

The other point is like with the cheap Chinese supplies you can buy, the current rating is rather optimistic. So better limit the current setting to something like 2 A.