Banggood PSU Enhancements

[Fabius_89]

There are simpler ways to reduce the negative supply (e.g. just change D7 to a 3.3V or 3.9 V zener). A small BJT wired as a VBE multiplier could also work.

I went with the Lm337 approach by Paul´s blog because the original zener is a bit noisy and as he found in constant current mode there is a 1.7V peak to peak ripple at the output which in some way the negative charge pump is responsible. Being driven by mains frequency. So eliminating D7 and changing R3 to and LM337 reduced some mains hum in CC mode. Still I must say that I did not experienced that much ripple testing my supply with the mentioned modifications (LM337 as R3, 470uF cap at output and 2.5A max output limit) as Paul did, and he limited his PSU to 1.5A (for more information go to http://www.paulvdiyblogs.net/2015/05/tuning-030v-dc-with-03a-psu-diy-kit.html?m=1)in his, maybe I got a better version of the PCB or who knows?

[Kleinstein]

Ripple in the CC mode is to a large part due to the poor board layout, not properly sensing at the shunt. This could be fixed with a bodge. Cut at the blue line and connect the green.

For less negative side ripply one could increase C3 a little.

[PinheadBE]

I intend to use the modified kit with a 2x12 VAC transformer, and with the Arduino controller described on Instructables, modifying the code to switch a relay to change the input voltage to either 12 VAC or 24 VAC depending on the output voltage needed, so that the power dissipation in the output transistors could be limited for low output voltages.

Do you think it might work as such, or do I need to foresee other changes to the already modified schematics ?

Any opinion appreciated.

Thanks

[Kleinstein]

Reducing the AC voltage to 12 V may result in only relatively low current for the charge pump. So one may have to increase the capacitors there or still run the charge pump from 24 V in some way. At high current and thus high ripply 12 V AC may not be enough to power the OP used to provide the 10 V reference - in some suggestions this part of the circuit has changed anyway. So it could be a problem with some versions only.

There are quite a few modified / improved versions of the this supply around. As there may be a significant drop in the supply, when switching the AC voltage, one may have to have sufficient buffer capacitance. It may help to have the separate filter cap for the OPs, as in some versions.

[montyx]

Hello,
I've just bought this PSU from Aliexpress, and since then I've found this topic about. You have mentioned there are different and updated versions of the PCB itself. Can you help how can I find them please?
Thanks in advance

[Kleinstein]

The main weakness of the circuit it that some of the OPs (TL081) that come with the kit get too much voltage - so either change to other OPs that can stand the higher voltage or reduce the transformer voltage (e.g. 18 V AC). With changes to the circuit (some could be done as a bodge) one could reduce the voltage to the OPs a little, but this type of circuit is limited to output voltages smaller than the output range of one of the OPs. So 30 V from the TL081 is very hard.

The PCB is a main part of the kit. So there is little use in buying the kit and get a new board. The main weakness of the PCB layout is the ground connection that leads to ripply in CC mode. This can be fixed (see the picture a few posts before).
AFAIK the kits still come with the old layout.

[montyx]

Honestly this kit is a good sandbox to play with. The price indicates that, so I haven't got so big expectation
But also I'd like to purchase a low noise psu for a affordable price, and I prefer kits, because I can learn during the assemble. This was the most important reason why I've purchased this kit.
So if you have better ideas about more reliable kits, or any linear adjustable PSUs which are affordable, I could really appreciate any small help to a beginner guy.

[Urshurak776]

I’m
Building one but using a UA741CN chip instead of the TL081’s.  Transformer is 24V/2A.  Not sure how well this will work but giving it a whirl 🤣

[m3vuv]

Hi all,im currently building one of these,im checking voltages as i go, here goes,dc input from rectifier is 28v,u1 pin 7-4 is 42v,pin 4 against rectifies ground is -15v,u2+pin 4+rec -ve =-5.1v pin 7+4=33v, u3 pin 4+rec -ve = -5.1v,pin 7+4 =33v,why do i get  42v on pin 7+4 on u1?,any ideas,i presume this is wrong?,this is without the ic's fitted,just the sockets,cheers in advance.

[Zero999]

Hi all,im currently building one of these,im checking voltages as i go, here goes,dc input from rectifier is 28v,u1 pin 7-4 is 42v,pin 4 against rectifies ground is -15v,u2+pin 4+rec -ve =-5.1v pin 7+4=33v, u3 pin 4+rec -ve = -5.1v,pin 7+4 =33v,why do i get  42v on pin 7+4 on u1?,any ideas,i presume this is wrong?,this is without the ic's fitted,just the sockets,cheers in advance.

That sounds right to me. The rectified voltage is equal to the RMS AC voltage, multiplied by √2, giving around 40V.

Did you read the first page of this thread? The TL081 included in the kit has too lower voltage rating of jus 36V and should be replaced with an op-amp rated to at least 44V. A few alternatives have been suggested.
https://www.eevblog.com/forum/beginners/bangood-psu-enhancements/

[Kleinstein]

Hi all,im currently building one of these,im checking voltages as i go, here goes,dc input from rectifier is 28v,u1 pin 7-4 is 42v,pin 4 against rectifies ground is -15v,u2+pin 4+rec -ve =-5.1v pin 7+4=33v, u3 pin 4+rec -ve = -5.1v,pin 7+4 =33v,why do i get  42v on pin 7+4 on u1?,any ideas,i presume this is wrong?,this is without the ic's fitted,just the sockets,cheers in advance.

The description is a bis confusing. It is not clear where the voltages are measured, but it sounds like there is something wrong. One point to check may be the shunt. The circuit has a negative supply of some -5 V, so the -5.1 V are OK. If needed one may be able to reduce this a little (e.g. -4 V), but this still does not help with the main weakness of the circuit: it gives too much voltage to some OPs.

Are the 28 V from the rectifier measured without a filter cap ?  The capacitor gives the high DC voltage close to the peak voltage.

[m3vuv]

yes thats after the big input cap,basicaly i get 33v  to a coupe of opamps and 40 odd to the other,thats between pin 4 and 7 on all opamps.

[floobydust]

Then I would think your shunt resistor R7 0.47R 5W is open. With no IC's in place, there might be stray charge on C7 but ultimately U1 should have lower rails than U2, U3.

[m3vuv]

forgot to mention the shunt resistor wasnt fitted as thats off board,was just checking supply voltages to the opamps.