Jackery Battery Bank FAIL Investigation - Part 1 

I would say solder a fuse or a thin copper wire (like 0.3 mm) over the battery protection ICs and test it again. When it works, you know the protection ICs went defective and you can try to source new ones. When it still does not work or it blows the fuse/wire, you know there is some major fault and you just salvage the cells.

A tip regarding the velcro... always put the FLUFFY part of the velcro on the object you're likely to drop on the floor/ground. If you put the hooks on that part then when you drop it you'll likely end up with all sorts of dirt/debris attaching itself to the hooks and it'll not only look awful but won't work as well. Stick the hooks on the tripod/whatever that you won't drop and the fluffy on the stuff you will drop.

That top protect thing looks cracked.

That is most certainly a DMG5802LFX-7 by Diodes Incorporated. The markings vary, so it’s hard to find. ME = Product Type Marking Code, this doesn’t change YM = Date Code Marking that has a decoder Y = Year (ex: F = 2018) M = Month (ex: 7 = July) So that is was made in July of 2018! Hope this helps :)

Did you check that the FETS weren't off because the BMS was requesting them to be off? It's possible that they were open because they were supposed to be. For what's worth, i'd have checked the gate voltage. Just my thought. I'm in the process of trying to fix a pack myself at the moment, but can't get the BMS to unlatch 😬

I posted the link to that 6 pin part but it was automatically removed. The part number is this DMG5802LFX-7 and it is just a dual mosfet and unlikely to be the problem.

You're right that it could possibly be a latchup situation, but I don't think you left it disconnected long enough. Might want to try disconnecting and shorting the inputs of the BMS ICs.

Maybe try charging the cells using a bench supply, bypassing the protection chip, and see if the chip comes alive once the cell voltage rises?

I think that mosfets might be okay, there's two of them to share current. You might check for chips on the other side of board, as there might be battery protection IC there that causes problems, and keeps mosfets open.

The top battery protection mosfet has a crack right through the top length of it as seen just underneath the white fuzz on it in the closeup at the end. I(t look like it shattered exploded. Note that this may also mean that the other boots converter IC may be shorted out and it blew up the protection ICs.

What's the voltage on the gate? before you assume something is off because it's broken, first check it isn't off because that's what it's being told to do.

DMG5802LFX is a dual mosfet in a W-DFN5020-6-EP package likely controlled by the controller under that inductor. Unlikely that those failed is more likely that is the controller.

The interesting part for me is, I repair low vision devices. They have battery packs with 2x3500mAh from Sanyo. They fail a lot. But I always disasamble the packs to safe the cells and they test good. Now the point: The protection PCB uses the same FETs in the same package....

I would use a current limited bench supply set anywhere within a li ion cell voltage and power the circuitry after the protection mosfets and if it goes CC use the flir if there are mlcc caps or the main PD chip is dead. I have encountered "dead" powerbanks where an mlcc went short circuit and the battery protection stayed tripped. Removed and changed the cap and it worked again.

You could replace with an SMD fuse if you're stuck finding a replacement, not ideal but better than a link.

WAG = wild ass guess , SWAG = scientific wild ass guess --- Since the batteries are charged , maybe the failure is from the output side .

My last broken power bank had open circuit current sense resistors. From the outside they looked absolutely fine 😂 I found the problem in the end and it worked again. It was a special power bank with a PV input not in production anymore so it was worth the repair cost for the customer. But working with a power bank isn't something I want to do on a regular basis. I not comfortable working with theses high power LiIo battery's.

I recon there must be a dedicated battery protection IC which controls the MOSFETs. I would check it's inputs and outputs first. You could also buy a generic dedicated protection PCB and just bridge the internal one.

14:40 Thanks for the functional diagram Dave

They may be dual mosfets, without any logic builtin. So maybe searching just mosfets filtered by package would get you something. There may be something on the other side of the board controlling them or it may be the main processor. Maybe the main processor turned them off but it won't turn them back on via just USB power? Maybe try shorting the mosfets to power it from the cell side?

Fun watching your videos over the years! Wonder what you do with those nickel strips that pile up? Is there any way to recycle? Thanks!!

An expensive brand, but once opened just contains the usual 'one-hung-low' brand components. Basically, a turbo-charged standard USB power bank. Those 'parts' look like two back to back MOSFETS (same as in standard protection but the high powered version), the small device nearby is probably the protection IC.

Short out the protection for a second and see if there is a sign of life.

You can use any dual smd mosfet for that. Or just glue one on the side and jumpwire the pins.

What multimeter are you using that allows you to see a separate display like you have at the top?

Both mosfets are cracked. And the same package is used on some battery protection boards which are mounted on tablet or phone batteries Or on batteries which the protection circuit is included in them. If you have collected some of those battery protection boards now it's time to use it.

Just solder a 4-inch nail across that part of the circuit. We used to use those nails as fuses on the old workshop fuseboard when we'd run out of wire. 🙂 Battery protection is for girls!

at 5:54, what are the 2 small red wires for on the bottom used for?

Dave, if you Google "mef7 marking battery protection", I believe the first result is the IC: MJ3542 series

Did you measure the ”gate” of the protection devices?

I've never seen this particular model until I explicitly searched for the model number, and I've had to source Jackery stuff 3 years ago. Their website also makes no mention of this power bank, which probably means they never sold it in the US. We have a couple of the big chonky "generators" that are still trucking.

What are the most bit closer to the negative contact looks like it has a crack across it

Check the battery cells from the start. BMS would be poor. You kill me, Dave.

If there's a current surge after the cell protection ICs, they'd trip open instantly when you reconnect the cells. Assuming there is an issue there, this possible fail case has not been checked, has it?

Try shorting from the battery ground to the circuit ground. It can reactivate some circuits. My guess...

Did you take a look on the other side of the pcb?

15:16 is there a crack in the lower protection unit on the right side? Looks like it wanted to inflate somehow 😄

Put a jumper over the MOSFET and see if it powers up or if something blows up.

Use thin metal to disengage the clips and it pops right out eazy peazy. That’s how mine came apart.

are those mosfets being help open?

one of the protection IC's look to have a crack on the right side 3 pins ?

Fully charged would be 4.1 Volts. 3.58V is far from that.

Hey dave, ive played with numerous power banks over the years, stress tested and even destroyed a few. In my experiances, the failures are always on one of two spots. The battery protection/mosfets and the main IC or boost chip itself. Not to say othe rhtings have gone wrong but generally in the 20+ power banks ive destoyed in testing pepouses, it usually one of them two. Try bypassing the protection mosfets. These are usually the death of most power banks that stop working with nominal battery voltages. (not charging or going straight to full when plugged in is a sign of this.)

Remove the top ME1, it’s cracked. Lower MOSFET is probably still functional. It’s only showing open because it’s in parallel with the damaged MOSFET.

Voltage on the gate of the fet? I assume there isn't any due to the fact that both of them are open. Broken NTC? Maybe it's actually trying to protect the cells due to a wrong temperature value... My two cents.

spudgers, i use a guitar pick, dunno if that's an acceptable tool but, its always worked a lot better for me than those kinds

For the sake of testing, would it not be possible to remove and bridge-out those positions? That would let you confirm that the rest of the battery bank is working correctly.

LOL, Dave has run up against the same problems that I have when trying to fix Chinese electronics, lack of information! I've been working to fix a AC-DC power brick for a laptop, the unit is only 11 years old, and all of the chips inside have been obsolete for nearly 8 years, meaning they only had a life span of about 2-3 years. I have not yet found any replacements for these parts, other than getting some used and recycled parts (currently waiting the 3ish months for delivery). I'll be watching to see how this ends...

Have a theory if you were not meant to take it apart they would not have put screws or clips in it. Good tare down and review.

Did you see the ceack

xiaomi power banks also very common for them to fail for no reason (even if the battery is still good), and someone found out that there's some sort of counter that if you reach maybe some sort of charge cycle on the power bank, it stops functioning.

That was one funny ass tear down Dave! Thanks for the entertainment as always.

Why dont you just carefully short the protection FET's (like in a controlled way in case there is a short)? This way you at least know if th rest is working...

I love how you can't even read the specs label on most of these products. I have to use my phone to zoom in and take a macro pic.

What about the thermal probe, if it has gone bad, the MCU could be turning on the protection.

this probably should be on the main channel ;)

Charge the lipos directly, so the cells have higher voltage and the dis and reconnect the circuit maybe? Than connect the charger to reset the safety circuit

May be its the NTC causing it to go in a protection mode!

For the price of a new one I would say "murder it a bit more by taking out the cells and be done with it" ! 😂

Can't you try jumpering those mosfets?

One of the resistors on the left looked like it had a cold solder joint.

Film accumulators remind me of circular breathing.

some power banks are reset by plugging the output to the input.

Keep the cells and swap out the case+pcb with an eBay diy power bank

You didn't measure the gate voltages to see if the controller is trying to switch them on, so can't jump to the conclusion that they are dead transistors.

I believe it is the SON-6 Package

Hahaha. I love the Chinese English translations. Poke any battery with sharp object for very delicious fireworks display.

"Fuel gauge"? What, does this thing run on petrol?

why not attempt a bypass for now to see if it comes alive at all? then you could use some sort of fuse or thermistor to replace it not the perfect solution but for testing why not.

MEF7 on the top has a hole in it i can see parts inside

one of those looks cracked

Mine went on fire laying on my dresser, different brand

EVE makes some of the best cells on the market. On par with cells from Samsung, Sony, Molicel, etc....

Replace is to easy I think you should replace the whole thing with a New Product. Maybe a Local Design or a Up and coming Player that has good Reviews . Or a RU-vidr that has designed there own Pruduct.😅

no doubt it's just 3 days out of warrantee lol

USB-C port. Someone plugged poorly (poor USB-C design anyway.. lots of burned ports and pcb traces after it launched) and created spike.

Quite often those power banks are designed with management/protection ICs that cannot recover from their under-voltage lockout, when they enter the lockout condition it disconnects the management IC from power to prevent further drain on the cells. So it's a chicken before the egg situation, the management IC needs to power the protection FETs in order to power the management IC, proper implementations will allow the USB input to power the management IC to recover from the lockout but this device appears to use the same USB C port for input and output so thus the management IC needs to power a FET before it accepts power from the USB input, so no chance for the consumer to recover it. Solution is to just bridge the protection FETs with a low value resistor, some DIY battery bank PCBs even have a push button on the PCB to do exactly this as removing the cells or when installing cells for the first time will usually trigger the under-voltage lockout.

If you already know how it works why you need the datasheet? That's nonsense.

3.5v is not a fully charged 18650. Just sayin', it's charged at 4.2v.

chock full of cheap garbage parts... and Jackery is the "premium" brand??

EVE are a respectable brand.

Jackery Jobby

Batteries are the key characteristics that make any portable devices 'portable', or else they're just 'eWaste'. It could be worth exploring ways to revive dead battery banks - whether by repairing & reinitialising the controller boards (interesting tip to disconnect power - but would it have helped to 'purge' the circuit as well by pressing & holding the little button before reconnecting power?), or swapping out bad cells with better ones, or else salvaging & repurposing the cells for other devices & projects.

To chime in with what someone else said, BMS chips these days often have permanent undervoltage lockout (to prevent the "dangerous" condition of charging a battery pack from a very low voltage). If you disconnect the actual battery for any reason, the BMS chip can lock you out. For some laptop batteries, this requires fairly expensive hardware/software to reprogram the BMS ICs, and other times you need security keys that you can't get etc. Bypassing protection may still prevent being able to charge the cells. Looking up the BMS chip's datasheet should tell the story.

Jackshittery, huh? It's always those dead MOOSEFET jobbies... first thing to check when I see a failed BMS. Usually sends the unit off to a "beyond economical repair" bin :/ I'd still try desoldering them and checking with a component tester. Otherwise the board looks okay to me.

Trick... Stainless steel. I have a guitar pick shaped one that lives on my keys. Opens almost anything

as usual, 10.000$ worth of tools, but unable to repair a powerbank

Jackery is garbage. #Z

Oh I love how "engneers" don't do logical diag on domestic products it makes me LMAO, you didn't even see if the gates were pulled high not mention moving the big arse inductor & actually seeing WTF the chip was doing, appearently it's all about chip numbers rather than probing pins & diag'ing what's going on... bloody amazing diag skills NOT!