EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: robzy on March 28, 2022, 10:12:59 am
-
So I bought a 10S lithium ion charger (photos in link below), and I cracked it open, but I've never seen a lithium-ion charging toplogy like this before, and I wanted to ask if you have?
Basically, the secondary side of the main transformer (which sounds like it runs at circa 15kHz) is pretty much connected directly over the 10S LiIon batteries, with a single rectification diode, and 2 x 0.061ohm shunt resistors (in parallel for 0.031ohms).
The secondary side circuit has a dual op-amp, but it looks like its sole purpose is to measure the voltage across the shunt resistors and turns on/off a green and red LEDs.
I don't see any evidence of any feeedback (e.g. optocoupler) from the secondary side to the primary side.
https://imgur.com/a/E7e5Yaf
Does this charging topology look familiar to anyone? Can this topology do a reasonable job of charging a lithium ion battery when it only looks at the primary side of the transformer?
-
Eeep, does noone know, or was it a bad question?
-
I've seen this topology when the charger is designed to feed a battery pack with built in charge control/cell balancing. The power supply simply indicates charge done by detecting charge current to control the status LEDs. The battery pack itself is worrying about properly charging each cell.
-
The secondary voltage can be also sensed at the primary, thus eliminating kind of feedback.
Wasn't very common years ago, but it's being used more and more.
More info:
https://www.youtube.com/watch?v=n1PYEM4DIug (https://www.youtube.com/watch?v=n1PYEM4DIug)
Make better pictures of U1, where the marking is readable, if the datasheet is found it'll be much easier.
-
Thanks both for the responses :)
The secondary voltage can be also sensed at the primary, thus eliminating kind of feedback.
Wasn't very common years ago, but it's being used more and more.
[...]
Make better pictures of U1, where the marking is readable, if the datasheet is found it'll be much easier.
Oh interesting! Maybe that's whats going on! Unfortunately the markings on U1 are impossible for me to read, I'm going to have to try with different light and some magnification.
I've seen this topology when the charger is designed to feed a battery pack with built in charge control/cell balancing. The power supply simply indicates charge done by detecting charge current to control the status LEDs. The battery pack itself is worrying about properly charging each cell.
I do have a seperate BCM modue (if thats the right word) thats meant to handle balancing and fault conditions, but I wouldn't have thought that it would handle the CV/CC requirements of charing Li-Ion batteries.
-
this circuit doesnt have primary side feedback circuit unfortunately. primary side feedback usually is used in LED power supply
CC reuse primary side protection circuit
CV sensed PWM IC supply and not bias winding
so CV/CC accuracy ~10%
by looks 20..30Wmax SMPS
repaired ton of similar SMPS and rebuilt them so they dont explode next time
not that design for to take good example
-
It probably has indeed primary side current and voltage sensing. Look at how many terminals the transformer primary has, if it's more than two, it is likely doing that. I'm on the fence, if this is an elegant solution, or a dirty hack that shouldn't be used for battery charging, that can end explosively.
-
mistake not as bad, it is not hacked but regular LED driver utilized as battery charger. accuracy ~5%
expected voltage sag when loaded. because transformer, diode and capacitor losses. will give soft CV/CC curve(voltage and current will decrease and increase charging time at the end of charging)
temperature stability??
but dont expect getting more from made in china 50..100EUR charger, bad circuits, bad thermals, bad over current protections, shorted transformer windings....
-
It probably has indeed primary side current and voltage sensing. Look at how many terminals the transformer primary has, if it's more than two, it is likely doing that.
Y'know, I thought the primary side only had 1 coil (2 terminals), but looking at it closer I can see there are 2 coils (4 terminals).
it is not hacked but regular LED driver utilized as battery charger. accuracy ~5%
Yeah, if that's the case, I don't want to use it on Li-Ion batteries. My understanding is that you need much better accuracy than 5%.
-
biggest concern could be if secondary diode gets shorted (overheated) and BMS is not turning off