[Solved] Schematic of a 25.2V (6S Li-Ion) AC charger from AliExpress + Question

[John_doe]

Greetings and salutations.
Somehow it didn't seem right to post this in the beginner forum but do bear in mind that I am an amateur and thus beginner mistakes are not unlikely.

First, some background: I bought this 'TangsFire' 25.2V 5A charger from AliExpress to charge a cordless handheld vacuum cleaner. It cost me $12 on the 2024 Singles Day sale:
https://www.aliexpress.com/item/1005007409292612.html
I only wanted about 1.5A to 2A charge for my vacuum cleaner but this charger outputs 5A (and does actually supply around 5A, I checked) so I set out to reduce the current by opening and modifying the charger.
Initially, I thought I could just visually see how the current is limited by looking at the circuit but was unsuccessful in doing so, so I started doodling on the "Typical application" diagram from the SC2569 datasheet with MS Paint thinking it shouldn't take long before I figure it out. Alas, I drew about 75% of the secondary side before I finally understood how to reduce the current (yes, still with MS paint. I was too lazy to switch to LTspice at this point). So I replaced R29 with a 1.5K resistor and the current dropped to 1.8A and I was content. But I thought if I was already almost done with the schematic, why not finish it and upload it somewhere, maybe it'll help someone in the future (assuming I didn't make any serious errors) and also maybe get help with a part of the circuit I don't understand.

This charger was pretty much stripped to the bare minimum that keeps it safe (I guess you get what you pay for), namely the fuse and the NTC. Many components related to filtering and noise reduction were removed to save costs. Googling the model number shows that there are many products that use the same board, for example:
https://www.reddit.com/r/AskElectronics/comments/1eov5xf/bike_charger_add_missing_filter_and_safety/
https://mysku.club/blog/aliexpress/91531.html

A note on my schematic: I didn't know what Q5, which is a SOT23-3 part, was. The marking on it is simply '84' and I didn't find anything with Google. I thought it was an NPN but poking it a bit with a DMM disproved that theory, then I thought it might be a "Digital transistor" so I googled "digital transistor sot23 marking 84" and got many hits on the DTC114WCA So it might be it but I'm not 100% sure so I've drawn it as it's package.

My analysis of the circuit: It's a bog standard flyback power supply with a 431 and optoisolator, but added to it is an additional 431 (U5) used as a 2.495V reference for the LM358 dual op-amp. One of the op-amps (358_2) compares the current on the output shunt (R20) to 5A and if it exceeds it, the op-amp lowers the output voltage by increasing the current of the optoisolator LED thus the output is always 5A or lower. The other op-amp compares the output current to around 400mA and if it is lower, turns on the green LED and stops the 12V fan. It also has some hysteresis and will not reset unless the current of the shut exceeds about 800mA.

There are some choices that I found odd, like the choice of the audio transistor for the secondary 12V power supply (Q2), or the needless(?) complexity of the SC2569 supply voltage, but my main question is about the Q3 NPN transistor and it's supporting resistors. I don't understand what it does... It seems to do nothing ?
Here is the part of the circuit that baffles me:

So if anyone is bored enough to take a look, I'd appreciate an explanation.


P.S. I was looking for the SC2569 datasheet but could only find it on sites that require payment or some form of Chinese registration, so if anyone has a copy I'd be very happy if you could upload it.

[Edit Dec 21st 06:28 - Marked solved]

[Edit Dec 24rd 2024 01:34 UTC]
I just wanted to update this message with some information which I didn't have time to post earlier:
This charger is capable of 150W (Which is roughly the feasible maximum from a flyback power supply) and is sold in many configurations under the brand names 'TangsFire' and 'TANGSPOWER' with models TS-150WF and TP-150WF (respectively):
https://www.tangspower.com/charger/tangspower-150w-charger
The different models' names include numbers following the TP-150WF in accordance with voltage and current, E.G. TP-150WF-2525000 (25.2V 5A), TP-150WF-4204000 (42V 4A), TP-150WF-5463000 (54.6V 3A) and so on. They all use a board named JN-150WF-XXXYYY however, some of the boards with the exact same name and version are actually different. Here are links to more photos of the boards I found googling (including the 2 links I already posted):
https://www.reddit.com/r/AskElectronics/comments/1eov5xf/bike_charger_add_missing_filter_and_safety/
https://mysku.club/blog/aliexpress/91531.html
https://mysku.club/blog/aliexpress/100367.html
https://www.ixbt.com/live/car/obzor-byudzhetnogo-zaryadnogo-ustroystva-tangspower-dlya-akkumulyatorov-13s-48-v.html

Also correction to what I wrote before - The LED (Marked LED1) is not actually green but green+red in a two pin package. This is achieved by putting 2 LEDs in antiparallel inside one package and I did not know that such components existed until now (example of an antiparallel model: LTL-293SJW).

And finally, two more photos I took, one is the sticker on my charger, the other is the IC with the "84" marking. Also in the second photo you can see the modification I made by changing the resistor.





[End edit Dec 24rd 2024 01:34 UTC]

[PlainName]

I only wanted about 1.5A to 2A charge for my vacuum cleaner but this charger outputs 5A

The vacuum cleaner should only take as much current as necessary, so it should be unnecessary to limit the output of the charger. Unless the vacuum cleaner is very cheap and carries the warning "Only use with included power supply" and then I'd think twice about using it anyway 

[youngda9]

R20 is the current limiting resistor, it is 0.02 ohms.  If the supply outputs 5A and you want it to output 2A, multiply the resistor value by 5/2 and replace it with a 0.05 ohm resistor.

[youngda9]

R20 is the current limiting resistor, it is 0.02 ohms.  If the supply outputs 5A and you want it to output 2A, multiply the resistor value by 5/2 and replace it with a 0.05 ohm resistor.

Or you could do either one of the following modifications to get the same result if you don't have a 0.05 ohm resistor for the R20 location.
Increase R28 from 100K to 249K
Reduce R29 from 4.12K to 1.62K

[John_doe]

The vacuum cleaner should only take as much current as necessary, so it should be unnecessary to limit the output of the charger. Unless the vacuum cleaner is very cheap and carries the warning "Only use with included power supply" and then I'd think twice about using it anyway 

Fair comment, but I tested the charging  current with a lab power supply before I bought the charger - It behaves quite strangely, it will take any current you give it up to about 2A and if you exceed that, it will enter some weird protection mode where it still draws current, but only a few hundred milliamps. I did not check if the protection current actually goes to the battery or not.

The vacuum cleaner is not a cheap model at all. Well, it was cheap for me because I found it in the garbage, but an internet search shows it was a mid-range priced device.

R20 is the current limiting resistor, it is 0.02 ohms.  If the supply outputs 5A and you want it to output 2A, multiply the resistor value by 5/2 and replace it with a 0.05 ohm resistor.

That is true. However, R20 is a big, high power, surface mount resistor with an uncommon value. I found it much easier to replace R29 with a 0.25W through-hole resistor that I already have than order a pack of SMD resistors which I will probably never use.

Or you could do either one of the following modifications to get the same result if you don't have a 0.05 ohm resistor for the R20 location.
Increase R28 from 100K to 249K
Reduce R29 from 4.12K to 1.62K

Thanks. I already replaced R29 with a 1.5K non SMD resistor. It was easier than replacing R28 because R29 has more room on the board (I used the leg of the C9 capacitor as ground).


Soooo.... No one knows the answer to my question? 

[D Straney]

So, if Q3 is the DTC114WCA, then it looks to me like it's turning on a "dummy load" on its collector, once the 12V secondary-side control supply gets to a certain level.  There's the external voltage divider (effective low-side resistor is ~6KΩ, due to Q3's internal resistors) with 1.4:1, and then the internal voltage divider on Q3 of ~3:1.  This means that Q3 will turn on only once the 12V supply reaches ~2.5V.

I don't know why you'd want to put a dummy load on the output in the first place, and key it off the 12V supply - but if all this is true, then I'd guess it's a bandage over some weird startup behavior, where it needs a minimum load during normal operation, but doesn't start up well into even a small 1KΩ load.  Best guess I've got so far.

Edit: realized I mixed up Q3 and Q5 - dummy load would turn on only after the very beginning, when the 12V supply reaches ~0.8V.

[John_doe]

I don't know why you'd want to put a dummy load on the output in the first place, and key it off the 12V supply

I heard about some SMPS designs requiring a small dummy load to work properly. What I couldn't understand is why not just put a resistor across the output, and what you suggested, that it is a startup issue, just doesn't sound right to me because I have powered up the charger with a load already connected and it worked fine...


Actually, I just had an idea just now - Maybe it's to not discharge the battery?
If there was a 1K across the output, then if someone disconnected the charger from the power outlet but left the battery connected to the charger it would discharge it at a rate of around 24mA. Maybe that's what they were trying to avoid? What do you think?

[Edit Dec 19th 04:48 UTC - Spelling]

[youngda9]

To answer your question, Q3 turns on when 12V_1 comes up to provide an extra load to the output through the R18||R19 1K resistance.  Q3 is set to turn on at 0.6*12.2/10=732mV or so...which is immediately.  I don't see why they don't just tie the R18||R19 1K resistance directly to ground and remove Q3 along with R43-R44.

You could solder in a short across Q3 to mimic that it is always ON, effectively connecting the 1K resistance to ground at all times.  Try doing that and capturing the turn-on profiles of 12V_1 and the output voltage.

Run the test and report back showing before and after scope plots of the two voltages.

[youngda9]

You can also run the test where you remove R43.  This will hold Q3 off, resulting in removal of the pre-load from both 12V_1 and the output.  That might learn you something.

[D Straney]

Actually, I just had an idea just now - Maybe it's to not discharge the battery?
If there was a 1K across the output, then if someone disconnected the charger from the power outlet but left the battery connected to the charger it would discharge it at a rate of around 24mA. Maybe that's what they were trying to avoid? What do you think?

Good point, that would make a lot more sense than keeping the dummy load off only at the very beginning of startup - keeping the dummy load from draining a connected battery when powered off.

[John_doe]

Good point, that would make a lot more sense than keeping the dummy load off only at the very beginning of startup - keeping the dummy load from draining a connected battery when powered off.

Thank you, I'll go ahead and mark this solved but if anyone can come up with a better suggestion than anti battery discharge, I'm all ears.