cheap chinese adjustable lab power supply kit on ebay ( 3AG1 , xiaolin )

[Axel1973]

Hi everyone,

since its dirt cheap and i wanted to upcycle some very old crappy linear fixed voltage power supply a little bit i recently bought this kit from an ebay seller:

https://www.ebay.de/itm/172830484802

i thought, for this price i cant do anything wrong really. but still i was expecting at least "some grade" of an professional trader or supplier, offering some proper schematics, a bill of material incl. all names, specs etc. to download. but no.

all they provide are crappy low resolution screenshots/web graphics some poor chinese guy just googled and downloaded from somewhere  and the low rez product photos they have on the ebay site.

no building , calibration or ANY instructions or manual at all.

The PCB and the cheap arse schematics they googled for me mentioned a "design by 3ag1" , xiaolon and two date codes: 20170510 on the pcb top side, 20150520 on the pcb bottom side.

for later documentation purposes i scanned the two sided pcb with a flatbed scanner in high res color. files are very large (50+ MB a side), so i cant upload them here in the original form. therefor here are two heavily downsized versions of the pcb + the schematics they provided. im not sure if this schematic does match the pcb. didnt checked it yet.

i hoped to find some good documentations / hight res schematics , manuals , bom or further material on this kit, but yet failed to find anything.

Does anyone here knows this kit too and does have additional documentation or "heads up" to it ?

[Axel1973]

oh yeah, not to mention that the product description is AWFUL !!! im no english speaker, so having trouble understanding proper english sometimes. but this nonsense-chinglish ??!??! 

that is why im looking for better documentation material in the first place.

regards
Axel

[Old Printer]

From the pictures in the listing it looks pretty straight forward, assuming they are the same. As you have not checked that yet,  that would be a good place to start. For $10 I doubt there is much more available. If you have a specific question you will probably find some help if you can learn how to post a good picture. Match up the components with the picture and work your way through. Double check before soldering. If it is beyond your skills right now, put it away until you are ready.

[Axel1973]

this is what i have got so far (low res only)

[voltlog]

I also got this power supply recently and it will be featured in my next InTheMail video. I was doing a bit of research with the info found on the PCB to find out the original source of the schematic and maybe learn more about this power supply but this thread which was created the day before was all the relevant info I found  Anyway I will be assembling this unit in an upcoming video.

[Axel1973]

im looking forward to watch it and hear your opinion and experience with it.

[oldway]

At first glance, the greatest problem of this power supply is that the current limitation is far too slow to protect the mosfet in case of output short circuit.

[floobydust]

Four TL431's? They can't afford an op-amp or zener diode in china. IRF3205 to 5A, although some SOA circuitry.
Looks like a cheap design that will have many quirks.

[xavier60]

This design would have a very low dropout voltage, not the many volts of dropout that many of the op-amp designs have.
U3 should have been made to control the MOSFET's Gate via a more direct path.

[floobydust]

I see it has advantages in that no aux. supply is needed for op-amps and to drive the N ch. MOSFET.
I find TL431's noisy (fig. 7,8) and compensating them is a hassle, sometimes stability is a bit of work.

We'll see how the PSU does in the reviews.

[voltlog]

I assembled this power supply and also did some basic tests. I think there are some things that are not right with this power supply so I don’t think I will be using it in any project. For the full story please watch the video below.


For those who don't want to watch the full video I'm gonna list the problems I found: First I don’t like the overshoot behavior were seeing on power-on and I certainly don’t like that double drop waveform were seeing on power off.  Then there is the 50Hz oscillation that we can see on the output and I’m not sure if that’s caused by insufficient filtering on the rectified input or if it’s something else but I don’t like having that on the output of the power supply.

[qu1ck]

For those who don't want to watch the full video I'm gonna list the problems I found: First I don’t like the overshoot behavior were seeing on power-on and I certainly don’t like that double drop waveform were seeing on power off.  Then there is the 50Hz oscillation that we can see on the output and I’m not sure if that’s caused by insufficient filtering on the rectified input or if it’s something else but I don’t like having that on the output of the power supply.

So why not rule out insufficient filtering by measuring with a scope what's after the bridge rectifier and try and eliminate that part by providing stable DC from good power source? You will also have to provide another voltage for the upper rail, measure what is it's potential with standard AC input first.

[voltlog]

Here is a measurement taken at the output of the bridge rectifier right on the 4700uF capacitor.

[schmitt trigger]

Sorry for resurrecting an old thread but...
What are those oscillations?

[exe]

Sorry for resurrecting an old thread but...
What are those oscillations?

Looks like ringing to me. Normally due to excessive inductance (including your oscilloscope probes), incorrect routing, and too sharp signal edges.

Or stability issues, as ringing tends to fade. Hard to say without zooming in and know how measurement was done.

[mikerj]

Those small bursts may be caused by diode commutation; rectifier diodes switch can on and off pretty quickly which causes high frequency current spikes.  This can cause ringing in the interconnects between diodes and capacitor.  For this reason you will commonly see low value capacitors (~10n) connected across each rectifier in a full bridge when used in noise sensitive applications (radio, audio etc.).

[sairfan1]

For cheap chines power supplies best option is RD Tech https://rdtech.aliexpress.com/store/923042
Dave also reviewed some of their power supplies they are tested and well rated,
You will find options from something like $10 to $75 even comes with PC software to analyze data.

[floobydust]

Sorry for resurrecting an old thread but...
What are those oscillations?

That trace is apparently ripple across the 4,700uF filter cap.
It's probably power line hash or rectifier-recovery ringing with the transformer leakage inductance.
6A10 has no reverse-recovery spec. Thanks Diodes Inc., Rectron, Kingtronics and all you copycats. It's probably a slow diode.
Or Voltlog has dirty mains full of SMPS noise.
TBH I'm used to Rigols adding nonsense to waveforms, if there is EMI on the scope ground.

[Cliff Matthews]

About Voltlog's video: Was there a problem wiring that tiny CMI filter on the secondary?

[schmitt trigger]

Flooby;
I believe that your list of possible causes is correct. A combination of the factors listed.

[exe]

Just to be sure, what power supply are we talking about? Because if it's a linear psu with a diode bridge and a capacitor then it's not a waveform I'd expect. I'm also not entirely sure that small caps across diodes do much due to very slow voltage frequency in mains (50-60Hz).

[floobydust]

It's this linear PSU.
In the video, VoltLog mentions he does have noise on mains in his lab, so he added a CM choke after the PSU transformer secondary for his measurements to help filter some out.
But as you can see there is still some present. Nothing to do with this PSU kit.

[exe]

If it's a linear psu, then why does it have 72mV p-p 100HZ noise? It's also strangely shaped. I'd say something wrong in there. Not entirely sure the noise comes from the outside. At least not all of it. I've built like 5 linear power supplies, never had issues like that in the final product (did have stability issues, etc, but that were problems in schematics, wrong parts, etc). I'd expect noise and ripple to be well below 10mV under full load and no high-freq components (although, there are some nasty loads...).

So, there are two problems: 1) 72mV p-p noise 2) ringing at the top of waveforms. I'd check drive of pass transistor to be sure there is no stability issues.

[Cliff Matthews]

He should have measured those caps before installing. Maybe there's cheap 470uF caps inside of bogus cans.

[mikerj]

If it's a linear psu, then why does it have 72mV p-p 100HZ noise? It's also strangely shaped. I'd say something wrong in there.

You should expect to see 100Hz ripple on the main filter apacitor of any linear supply that uses a full bridge rectifier.  I don't understand why you think this is wrong?  The person that posted the waveform didn't mention anything about how much load was on the output of the supply when it was measured so it's impossible to say if the amount of ripple is too high or not.

Capacitors across bridge rectifier diodes certainly are helpful.  Despite the mains frequency being low, the rectifiers still switch on and off quickly, generating high frequency harmonics.  As I said, you will find these capacitors in linear supplies where low noise is important.