12V Portable Auto Car Jump Starter Power Bank Booster Battery Charger 50800mAh (WHL #54)
Some random power bank, we haven’t had those for quite some time now, have we…
This heap of trash actually came from eBay, which nowadays rarely has the best offers for China-shopping. I bought this back in December 2018 for 22.99€ including shipping, and as far as I can recall, I arrived so quickly that it must have been sent from an European warehouse.
The reason for buying a “jump starter” type of power bank was that at the time I had some project running that needed a ~12V supply. In the end, we decided for a different solution, but a 3S lithium cell configuration would have done the job. Those starter packs usually come with EC5 connectors, which are rated 60A continuous (competitor to the XT60/XT60E system used in my APC external battery mod), 120A for 2 minutes or 180A for peak load. That’ll do for a car starter in the average winter if the onboard lead acid battery isn’t completely dead yet. It might not fire up your Leopard 2 near Stalingrad, but you know, choose your tools accordingly.
Well, on the very first test run after a similar unit from a popular German brand succeeded, it, umm, failed miserably. And that’s not only because of the runtime nowhere near matched the estimated capacity, “50800mAh” obviously being China marketing numbers. Nope – it also failed in terms of cell matching. You see, if one really wants to jump start a car with such a thing (which is entirely possible, I did that myself with the other unit), it’s reasonably important to have a) similar cells and b) a similar state of charge in all of them. This unit failed both tests.
As it was sitting for quite some time, voltage readings did drop quite a bit further down from the rough 9.5V cutoff, which would have been close to 3.2V per cell in no-load condition. Voltages are now as follows:
Cell 1: 3.59V
Cell 3: 3.61V
Cell 2: 1.04V… (wa wa waa)
No surprise combined cell voltage went south quickly when one of the cells is significantly worse than the others.
Alright, so the battery itself is crap. What about the general construction?
Clamps: Well, looking fine – until you take a closer look. There’s coarse teeth with gold surface finish on three sides (looking good actually!), but the center part is recessed, so contact will only be made with the tips of the teeth instead of area contacts once snapped back. Furthermore, the entire upper jaw isn’t connected to anything. So contact area with the car battery would be a few mm² best-case. Who’d need more for a couple hundred of amps anyway…
The clamp leads do have diode protection, which only serves to keep the dumbest of the dumb inside of the gene pool for a bit longer. It consists of a separate PCB inside a really cheap in-line plastic box on the positive terminal, hosting four paralleled IRF 42CTQ030S D²PAK Schottky diodes with heavy additional solder on the top side. Bottom side is blank despite tons of via stitching…
These diodes feature a 10W power loss at just 25A of forward current, so their use is clearly a gamble of usage time vs. thermal mass, as prolonged use would melt the thing. Furthermore, who knows if those are genuine after all?
And back to the battery itself. I think the general construction of the pack is pretty nice, the big EC5 connector and the PCB do fill the envelope perfectly, and they are fixed in place pretty well. The battery also does use the available space just fine, it’s not wasting space but it’s also not a Samsung-like tight fit that would lead to self-destruction, given the cells ARE poorly chosen and will puff up over time.
The cell block is wrapped in a thick sleeve and has several layers of tape underneath, so there’s really no danger of an accidental short or the cells moving relative to each other. The one from the Russian dude linked below even says 22.2Wh/11.1V on this sleeving, clearly contradicting the 50.8Ah (!) figure on the product description. 2Ah capacity (each) seems reasonable to me, given the cells are fairly wide but pretty thin.
While the external wiring is said to be AWG8, the internal is AWG10. As it is extra fine stranded aluminium wiring in silicone insulation, it does meet and even exceed the expected diameter. Not sure why they’re gilding the lilly here and then cheap out on the battery itself 
but this wiring is perfectly suited for the job.
Soldering is…fine, I guess. The extra fine wiring does soak up a lot of solder, and every connection is soldered, none is welded. Having very short tabs for the cells could have killed the center cell.
The solder spots do look crusty because of lots of remaining and burnt flux, but are not that bad underneath. I don’t mind that at all.
And now for the board!
Starting from the back side, the DTM4606 (between the two mounting holes) is a combined N and P channel MOSFET package, SOIC8 footprint, 30V drain-source, 4 amps continuous drain current at 70°C ambient. Typical pinout for a SMD MOSFET, easy to identify.
Not so much the 8810 TFZ8SC TSSOP8 package nearby. I had no luck finding the second part (no manufacturer logo to be seen), but the 8810 matches an Alpha&Omega part (AO8810). That one also matches pinout of the chip – it is a 20V common drain dual N-channel MOSFET, having a common drain (duh) on pins 1 and 8, separate sources on 2/3 and 6/7, and their gates on 4 and 5, respectively. Both of these MOSFET packages have sub-100mΩ sense resistors nearby and are close to the battery input terminal, which makes some sense for switching applications.
Now the important one that wins the in the obscure chip lottery: The “JSY-3050PA 00S7K02”. 20-pin, 0.65mm pin spacing, narrow body and elongated so that it would fit 24 pins as well. There are a couple of hits to Russian sites listing this one as jump starter IC in similar or identical battery packs. Nobody has a datasheet for it, I don’t even know the manufacturer. If you have some data on it, please share it in the comments!
The chip has to be pretty capable, as there’s a lot going on in that power bank. Of course there’s a buck regulator for the 5V USB supply with some timeout, there’s a LED that can be toggled via the soft power switch, and there’s an indicator for the jump start function. Furthermore, as this is a 3S cell configuration in contrast to the regular 1S with step-up, there’s also at least a simple BMS included, as every single cell is connected to the PCB over the small connector opposite to the USB connector. It obviously didn’t do the job in getting the cells to an equal state, but that could be the fault of the cells or cell selection, not the IC. In regular operation I’d also expect a shutdown of the entire circuit if those cells drift too far apart. Ideally there would be a balancing circuit that can equal out those cells, but without a datasheet on the IC that’s just a guess.
Overall this could have been a pretty capable product, as I’ve seen from an European competitor charging about double the money for a valid product. Unfortunately the Wan Hung Lo cheapass cells, likely further compromised by soldering and therefore thermal damage to the batteries, did not live long enough to even come to the point of jump-starting a car – they died from a simple 5W load that was cut off way too late for the weak center cell of the 3S pack. If the others are okay they will be re-used individually with a standard boost converter circuit with cell protection, but this shouldn’t be the reason for buying this thing. There’s better ones available for both regular 5V device charging and jump starting your car.
Merry Christmas everyone! 
can you tell me the value of the can capacitor that is wrapped with shrink tube
thank you.
Not sure which one you’re talking about.
doesn’t have a shrink wrapped capacitor and the main battery pack doesn’t have caps at all.
The center electrolytic in the above picture is a 220µF/16V/105°C unit, and the wrapped part is just a 1W-style LED chip in a big clear package.