Need help with understanding what happened to my LiPo

Other Hardware General Electronics
1

I have a question. I have recently started working on electronics projects.

I had a Seed Studio nRF52840 connected to a BNO085 and a 3.7V 300mAH (i think) LiPo battery. It was all bundled into a plastic-case.

The firmware code, puts these IC's to deep sleep, turns off BLE, operates in low power mode. (Although I didn't have any brownout protection code) .
The serial monitor used to show: 4.2V percentage: 30-40-60-90-100 percentage. And it used to survive a full day and night atleast.

Then i moved to a bit more humid location, because of some reason, the thing wasn't charged for around a 1 month. Now the battery feels a bit puffy/mushy.

And the serial monitor just shows 5.28V 100% charging=1 . And once I disconnect after saying charging for 3-4 hours, the battery probably lasts around 30min. While charging the battery is relatively cold compared to the heat on the metal casing on the nRF52840.

At first I thought:

  • if its showing ~5V then probably the circuit is bypassed.
  • PMC or whatever I thought might have tripped, but then if that was true, then it would not have survived the 30 minutes.

I don't have access to a multimeter, (will update based on what I find) .. but the more pressing question is:

If this is the case, how does these smart watches, feetbits can be kept without charge for 2-3 months, and then it can still be charged back up? Is it a different battery??

Looks something like this: RUNE - iko's logs

2

As a first step disconnect it and take it to battery recycling point. Then we can try to find out what went wrong. Take a foto of the battery though.
And don't carry it in your pocket...

3

That's way to high for a lipo. You are lucky it did not burst into flames.

4

Battery quality... followed by circuit engineering.

A known manufacturer's original product is the only way to guarantee battery quality and longevity.

I have worked with thousands of LiPo battery packs from the early 2000s, mostly manufactured for the device, with hundreds I built from parts. I have seen these batteries physically abused - thrown, slid, dropped, splashed with water, overcharged during storage, undercharged during storage, over amped, discharged-dead-and-recharged, left on the charger overnight/many-nights... and not one single problem... regardless of the "boogieman" stories and shameless insurance company propaganda. I have had exactly one battery combust in my hands, which was a not-good quality electric drill BUT it was probably my fault for over-torquing the motors.

The "bad" batteries that might die on the shelf, or start a fire, are the illegally produced batteries (one can not "poorly" produce these, because strict guidelines ensure quality), or overly abused batteries.

That's hydrogen gas buildup. A sure explosion if released near a spark.

5

Many LiPo´s have a protection circuit that, among other things , prevents a deep discharge of the LiPo. I consider that a must have.

Still, relying on the deep discharge protection can be bad for the LiPo. An AI once pointed out the scenario for an esp32 board that resets the MCU due to the brownout detector, then when it starts up again and e.g. starts the radio for WiFi the brownout detector is again triggered. This loop goes on until the deep discharge protection of the LiPo kicks in. This affects the LiPo´s health negatively.

6

I, too, have had many batteries over the years. The ONLY one to go bad (it broke a fairly strong plastic case) was my best. It was an Anker.

7

My experience has been different, unless the only thing you classify as a problem is catching fire! I've seen Lipo fires - never my own - it's some scary stuff. And I am talking ~1000 mAH territory. I can only imagine a scooter battery fail.

@ikouchiha47 - you have been advised but it bears repeating:

Retire and properly dispose of any puffed or damaged Lipo cell or battery of cells immediately.

Do not take Youtube advices for how to recover from mistreatment around charging and discharging. Life is too short.

Do not attempt repair, do not rebuild packs from cells that look OK but came from packs with damaged cells. Life is too short.

I use Lipos with no protection circuitry. I would never deploy anything like that if there was any chance of discharging without limit.

There are better chemistry choices for batteries meant to run something unattended for long periods and be charged in situ w/o supervision.

You can probably "get away" with some things, but once again, luck will find you and (wait for it) Life. Is. Too. Short.

a7

8

Problems can be: short usage time on a full charge (weak cells), intermittent outages (see below), up to "fire" (never had a fire in the big batteries, but I caused a drill fire from misuse).

Some of my battery time was with an app-based e-bike rental. 36vdc systems, 250 watt motors. The rentals were abused to the worst of ones imagination (thrown off bridges was a favorite). I repaired most. Cannibalized the rest. We saw expected degradation of the batteries over the years to 85% of capacity, which is when we returned the batteries to the manufacturer.

The place where I built the 15Ah batteries, the owners were responsible for charge/discharge, but no returns were requested. We supplied their charger, and it was programmed. There are two (three, really) camps on charging: "Charge to 100% and use to 0%" (in real-charge, the batteries usage was from 58VDC/100% to 40VDC/0%) and "Charge to 80% and use to 20%" (and 70/30, but we discouraged that because it only lets you use 40%... and charge cycles skyrocketed, which is not good). We advised 80/20 because the real usage would (statistically) never use the full range before needing a re-charge.

Another place, I maintained the shop batteries, and I was the 100-to-0 camp, and in five years (expected end-of-life for a rechargeable battery), out batteries were above the the 85% cutoff.

When others, outside the shop, were responsible for battery care, most had years of use, even with frequent use. You could "smell" the problems when people were in for a complaint.

  • Weak cells. New standards not only forbids battery maintenance, but requires about a 2lbs/1kg of rubber potting compound to be injected into the battery housing. I have an argument against this, but smarmy insurance runs the industry, not the people who know what is happening. Not too long ago, and with properly manufactured and graded cells, it was only a matter of ripping some solid nickle strip welds and spot-welding in new cells. The cells all had their capacity and resistance documented. No more of this now, if you want to stay insured.

  • intermittent outages. This is caused by too-small gauge wires, which also have too-thin/wrong-compound insulation. The small-gauge wires used to get power from the batteries to the load is usually engineered for space, not current and heat dissipation. The high current is resisted by the small gauge causing heat, and the too-thin insulation is defeated by tight spaces and physical pressure. Sometimes the connector housing melts onto the connector pins causing gaps. In a bad case, the too-hot wires broke through their insulation on the "bends" and exposed bare, wire to whatever might touch it. Again, I documented this with hundreds of photos, suggestions of the wire gauge, insulation properties, and connector properties, but all I got was hundreds of, "Shut up and look pretty." (this is when I stepped away from batteries).

I enjoyed maintaining broken batteries. It was a little physical work and a little mental work, and it preserved resources. I am not into throwing a 15Ah battery away for a single cell or group of three cells. Again... never had a fire... and the "bad battery" issues were just the poor engineering allowed on the wiring that passed large current. Not the battery. Ever.

For all these battery measurements, I have been ising the West Mountain Radio CBA V

9

The puffiness is due to hydrogen and other extremely inflammable gases generated by improper discharging and charging. The battery is destroyed, so recycle it properly. A proper battery management system will help prevent this dangerous situation.

It is extremely important that the cell voltage never be allowed to drop below about 3.0V, which usually means disconnecting a discharged battery either by a switch or discharge protection module.

10

Seeing the voltage I thought that it might be bypassing the the battery. Because I didn't feel the battery getting hot.

11

Okay, I have to look at discharge protection module. Barring the state of the battery rn, I was looking for information on how does actual production devices do it. Something I could replicate with dev boards.
And to understand, why this happened.

I think i have a bunch of answers to go and look around. The discharge protection part, didn't know this would come up as an issue, was mostly of the opinion that, once the lipo is discharged, there is no current/voltage going to the whole board, so its "off"

As for AI. I mean i can use it for software dev, because I can look and read code to understand, and cross reference, but then for things I don't know ----- aaah, its like, when new born birds have their eyes shut, and open mouth to eat, but they don't know if the food is from the mom's mouth or butthole. Its difficult like that. I would rather read some case-study, engineering blog or talk to someone real

12

Xiao nRF52840 is well optimized for low power applications and average current draw is typically in microamp range.
But low power operation of BNO085 board might be very challenging (or impossible).

13

Indeed. You may want to read about Lipo chemistry at

Lipo batteries are amazing and can deliver prodigious current even if you didn't intend provoking them to do.

Like severing both the red and black leads with one snip of your wire cutters.

If I did not need real watts (amps * volts), I would not choose Lipo.

As you see, YMMV.

a7

14

The board that OP is using is designed for a LiPo cell.
The mistake that OP made is to ignore battery safety.
You can't just leave a project to fully drain the battery.

The project must have

  1. a protected cell (built-in circuitry to stop full discharge)
  2. code and/or circuitry to shut off the processor when the battery gets too low (say 3.5volt)
    or
  3. a simple battery disconnect switch if the project is not used for a while

This problem wouldn't have happened if the board was permanently connected to USB.
Because that Seeed nRF52840 board has a built-in charger.

This is the type of battery OP should have used.
Note the protection circuitry under the yellow plastic

1Ah protected cell
1Ah protected cell800×800 36 KB

15

The consequences of getting these LiPo's wrong, can be dire.
It can wipe your property out and you.

I reported here a case of a man getting life threatening 25% deep burns after a LiPo in a shed set off a chain of events leading to a gas cylinder exploding. A simple accident that cost him dear. His precious collection of RC model aircraft and vintage motorcycles destroyed.

You can avoid a lot of problems by not putting low cost as the primary reason for choosing particular items.

Best treat LiPo's as an accident waiting to happen. Maybe not so much good quality branded, tested stuff like our phones and tools and increasingly, our vehicles.

16

One issue with those so called LiPo 'protection circuits' is that a lot of them have a very low cutoff, often 2.4v.

So if the cutoff is ever needed, then the LiPo has gone so low it really ought to be safely disposed of.

17

Indeed happened to me once, I was charging a 3S 1300mAhr LiPo from a model plane and it exploded right next to me into a fireball about 1m diameter. Close call.

18

I think!
It might be a good idea to know what kind of battery you actually have before going any further. Maybe it's not even Li-ion.

19

The protection circuit cut-off (usually 2.7V) should be last line of defence.
That's why I also recommended monitoring battery voltage, and take action when needed.
You should not leave a Lithium battery circuit unsupervised.

I once wanted to see how a flat 18650 (unprotected) battery reacted to damage.
I took a hammer/nail, an axe and a torch to it, without anything happening.

20

I built my own protection circuit based on an FS312F-g which cuts off significantly before that.