SC6038 or a DIY charger for a 7.4 V battery.

I seem to have drifted away from reviewing the various DIY projects that are my usual subject.
I decided to make up for that in this review.
So, here’s a charger with balancing for a battery made from two lithium cells.


As always, I’ll start with the unboxing. Here’s how it arrived.
I know some readers aren’t keen on unboxing photos, but I just had to show this.
The chips came with tracking. I don’t remember how long they took to arrive, but they got caught up in the holidays, so it definitely wasn’t quick.
When the tracking showed that the chips had arrived at my local post office, I stopped by after work to pick them up. I went in and said I was there to collect a small parcel—or probably an envelope, since how much space could ten chips take up?
But the seller amazed me. He’d packed the chips in a cardboard box.
You can see the box dimensions in the photo.


Inside was a bubble-wrap pouch. I’ve had parcels that were packed well, but this level of packaging takes some effort. The seller definitely deserves a thumbs-up.
My order traveled like royalty.


Somewhere inside that mountain of packing material, I finally found my order.
Everything was neat and tidy: a resealable bag and a printed insert listing the chip part numbers.
If the seller had included the datasheet too, I probably wouldn’t have been able to take it—but apparently he knows where to draw the line :)))))


As you can see, the full order is there, with the chips on standard tape.


The markings on the chips are clear but very minimal; there’s no manufacturer listed at all.


For some reason, I thought these chips were supposed to have a thermal pad on the underside for soldering. But there’s nothing on the bottom. The datasheet doesn’t clarify this either.


I needed to confirm the order, but how could I do that when I had no idea whether the chips even worked?
So I quickly threw together a little charger board.


Then, specifically for this review, I drew up the schematic.
Maybe someone will want to build one.


I won’t go through the whole PCB printing process; it’s described in detail in my review of the paper I use for this.
Here’s what I got in the end.


Then I gathered all the parts I needed. The photo doesn’t show everything, though; I added the LEDs and capacitor later.
Rather than two separate LEDs, it’s better to use a two-lead, bi-color LED.
Red indicates charging; green means the battery is charged.


This is the little board I ended up with.
I set the charging current to 400 mA, since I didn't need any more, and the chip gets quite hot (around 100°C). The better approach would have been to make a double-sided board and connect the two sides with plenty of vias, but I did things a little differently.
You can see two jumpers on the board (and especially in the PCB layout); they act as a heatsink.
Soldering a small piece of copper-clad fiberglass board to these jumpers on the other side would improve heat dissipation. The jumpers themselves do a pretty good job, though.
So even though they don't actually connect anything, you still need to solder them in place.

The charging current is calculated using the formula IOUT = Vcs/RSET
where Vcs is the internal reference voltage, which is 200 mV, and RSET is the current-sense resistor's resistance.
So in my case, the current is 200/0.5 = 400 mA.


And here's a closer look at the whole thing.
I used 12 kΩ resistors instead of 10 kΩ ones because I'd run out of 10 kΩ resistors.


The actual charging current turned out to be a little lower.
The initial charging current was 370 mA


About 20 minutes later, I measured the charging current again and also checked the chip's temperature: it was 100.1°C.
That's not actually too bad; it can operate at this temperature for a long time without any problems.
The temperature is also this high because the board is powered from 12 V. It drops noticeably when the input voltage is reduced to 10 V.
So, you can get a higher current, but only if you keep the input voltage down and improve heat dissipation.
After warming up, the current dropped to 350 mA
I also checked that the chip isn't bothered by shorting the input wires.


I should say a few words about one feature of this device as a whole: its battery balancer.
It's no secret that cells connected in series can become unbalanced, since battery capacities are never exactly the same.
I described the balancer circuit in my review of a Kenwood radio battery modification.
I used the same solution here, but this time I decided to run a small experiment.
I got two 18650 batteries with a capacity of about 1800 mAh. As it happened, one was almost discharged and the other almost fully charged.
I connected the batteries to the board and periodically measured their voltages, recording the readings in a table. I didn't take the measurements at fixed intervals, since I couldn't always do that, and the whole process took about two days (the difference was quite large).
Don't be surprised that the battery's total voltage increases—this isn't a perpetual motion machine :)
The charged battery was simply on its “plateau,” where its voltage barely dropped, while the voltage of the discharged one rose quickly because it hadn't yet reached the part of the curve where the voltage creeps up very slowly.
The board was transferring energy from the charged battery to the discharged one. You could speed up this process, for example, by replacing the storage capacitor with one of higher capacitance, but I wouldn't do that.
Anyway, here's the table I ended up with.


That's probably it.

So, to sum up.
Pros.
The chips work
The packaging is excellent.
The seller is decent.

Cons.
None found.

Sure, I could have bought a ready-made device, but it's not always possible to find one in the size you need. And sometimes you need to use these chips in devices you're developing.

I hope you found this information useful and interesting. Sorry, no cat today :(

And as always, downloadable files, datasheets, PCB layout, schematic.