Author Topic: Hotspot battery getting overcharged  (Read 5677 times)

0 Members and 5 Guests are viewing this topic.

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Hotspot battery getting overcharged
« on: April 23, 2020, 10:32:16 pm »
I have a netgear 810s portable hotspot that seems to be overcharging its internal battery.

In the past 3 years of ownership, I had to change the battery twice already and it probably only gets less than 50 cycles during its lifespan since I use it docked in its station (ethernet to desktop) most of the time and only rarely ever use it as a portable hotspot.

The reason for changing the batteries is because they always seem to swell and that gets worse with time so for the sake of safety I keep changing them when that happens.

Last time I took the battery out and measured its voltage, it was reading 4.28V which is quite a bit higher than the 4.2V norm to which Lithium batteries are usually charge to so I'm guessing that the constant overcharging is causing them to bulge.

Now my question is, when I get the battery replaced for the third time now, is there anything I can do so make it stop charging beyond 4.2 V? Maybe a resistor or some nichrome wire to drop the voltage "seen" by the battery?

 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9322
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Hotspot battery getting overcharged
« Reply #1 on: April 23, 2020, 11:09:15 pm »
Can you find the datasheet of the charging chip? I once had a cheap pocket AP with a similar design flaw, plus it had the battery right up against the hot CPU.

As for an easy fix if the chip does not have a documented way to change the voltage, a pair of back to back Schottky diodes in series with the battery would work. Even better is to bypass the diodes with a MOSFET when operating on battery power.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Online amyk

  • Super Contributor
  • ***
  • Posts: 8526
Re: Hotspot battery getting overcharged
« Reply #2 on: April 24, 2020, 12:02:00 am »
You need to check the charger IC/regulator; it may have drifted out of tolerance.

Alternatively, fit a 4.3 or 4.35 volt cell --- 4.28 won't fully charge one, but those can tolerate the voltage safely.
 

Offline Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 4004
  • Country: nl
Re: Hotspot battery getting overcharged
« Reply #3 on: April 24, 2020, 01:54:03 pm »
4.2V is the absolute maximum for (apparently most) Li-Ion chemistries.

For long time storage it's usually recommended to save them half charged.

Also, It used to be that when you go above around 4.1V, the battery lifetime (recharge cycles) start diminishing.

Usually batteries are pushed to (or just over) the maximum for the manufacturer's sake.
They get a bit more charge (10 to 15%) out of the battery for your new gadget, and when the battery dies after several years, they are also happy to sell you a new gadget.

But there are 100's, maybe 1000-s of (slightly) different battery chemistries which makes it difficult to give a general answer.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2248
  • Country: us
Re: Hotspot battery getting overcharged
« Reply #4 on: April 24, 2020, 02:38:09 pm »
It may not make any difference, but you might see if there's new firmware for the hotspot.

Since swelling batteries are a safety issue, it might be worth a call to Netgear to see if they'll replace your hotspot.  4.28V is definitely too high, particularly if it continues charging all the time instead of terminating the charging.

In any case, you don't want to be using this thing as is.  It could catch fire.

Actually, one option might be to get two batteries, and find a way to charge them using another charger.  Then you could switch them out as needed, and never apply charging current to the hotspot itself.  I'm not sure how the docking station would fit into this.  Is it getting power over Ethernet?


 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #5 on: April 24, 2020, 03:46:32 pm »
It may not make any difference, but you might see if there's new firmware for the hotspot.

Since swelling batteries are a safety issue, it might be worth a call to Netgear to see if they'll replace your hotspot.  4.28V is definitely too high, particularly if it continues charging all the time instead of terminating the charging.

In any case, you don't want to be using this thing as is.  It could catch fire.

Actually, one option might be to get two batteries, and find a way to charge them using another charger.  Then you could switch them out as needed, and never apply charging current to the hotspot itself.  I'm not sure how the docking station would fit into this.  Is it getting power over Ethernet?

Unfortunately, they haven't released a new firmware in over a year now and contacting them for a replacement is not an option as there are no Netgear distributors in my country (bought from amazon Spain).

As I am using the hotspot docked in its station most of the time, which is basically a Wi-Fi router and charges the hotspot's battery, It's not much of an issue especially now that I've noticed that It works just fine without the battery so I will only be needing it when I use it on the go so from now on, I will be using it without the battery inserted and only insert it when I leave the house and take the device with me.

That being said however, I could charge it with an external charger by 3D printing a housing for it with some copper or gold platted pins to make contact with the battery and charge it that way.

But then again that defeats the portability aspect of the hotspot if I need to carry a dedicated charge and battery enlosure whenever I need to recharge it.  :-//
 
« Last Edit: April 24, 2020, 03:51:48 pm by drummerdimitri »
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #6 on: April 24, 2020, 03:49:41 pm »
It's common for portable device manufactures cheating by charging batteries to 4.3V, 4.35V or even 4.4V.

There are different chemistries for Li-ion batteries, and some can be safely charged to a higher voltage at a cost of reduced lifetime.

50 cycles is ridiculous, though, you might get a faulty product where the manufacture bought 3.6V batteries (4.2V cutoff) for 3.7V batteries (4.3V~4.35V cutoff).

Usually a properly designed 3.7V battery can last hundreds of cycles at 4.35V cutoff.

I wish there was a way to probe the contact pins while it is charging to understand the behaviour of the internal charger.

Will have to take it apart to see what charging circuitry is on the board. I'm guessing the think black strip attached to the battery is just a protection board for OC, OV, OT etc. correct?
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #7 on: April 24, 2020, 04:35:58 pm »
Will have to take it apart to see what charging circuitry is on the board. I'm guessing the think black strip attached to the battery is just a protection board for OC, OV, OT etc. correct?

Yes, but the factory could be ordering the "correct" 4.35V board for 4.2V batteries.
Supply chain management for small players in China is tough.
I just got a sample book of 0805 capacitors, with one type being rated 104 1kV and just 0.5mm thin :palm:, right this afternoon.
You never know what you get unless you order from known distributors or the factory directly.

Never mind, I just applied a voltage to the protection circuit (I have taken the battery appart) from my DC power supply and drew current from my load and it let the "battery" go up to 5V without cutting off power to the load.

Maybe that's controlled internally?

 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #8 on: April 24, 2020, 05:49:18 pm »
Can you find the datasheet of the charging chip? I once had a cheap pocket AP with a similar design flaw, plus it had the battery right up against the hot CPU.

As for an easy fix if the chip does not have a documented way to change the voltage, a pair of back to back Schottky diodes in series with the battery would work. Even better is to bypass the diodes with a MOSFET when operating on battery power.

I opened up the unit and all the components are shielded via a soldered case so I will not bother with that.

I like the idea of using diodes to drop the voltage. Seems like a cheap and reasonable fix.
 

Offline Renate

  • Super Contributor
  • ***
  • Posts: 1460
  • Country: us
Re: Hotspot battery getting overcharged
« Reply #9 on: April 24, 2020, 06:25:54 pm »
Well, at least it's not one of the Verizon "Elipse" model.
They intend portable hotspots to be used portably and charge as quickly as possible when they see power.
All the ones I've seen will swell up if you leave them on power.
Some of them work without batteries in them, some don't.
My current piece of junk doesn't require a battery so I just pulled it.
The previous one I just wired a USB cord to a dropping diode to the terminals for the battery. (i.e. no battery).
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9322
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Hotspot battery getting overcharged
« Reply #10 on: April 24, 2020, 06:56:41 pm »
Then your protection board only does OCP and dead short detection, which is the bare minimum. Those are hard to come by as newer chips almost all do UVP/OVP as well.
A bit silly to use active components just for overcurrent protection in a battery pack when a much cheaper fuse will do just that. Most likely it's just defective in some way.
The previous one I just wired a USB cord to a dropping diode to the terminals for the battery. (i.e. no battery).
A trick from Perk coin mining back in the day, put into use soon after those running some mining racks had the batteries expand from running the (cheap) phones 24/7 under heavy load. Some just hardwired the battery connection to 5V, hard to tell if that causes any early failures since it wasn't at all unusual for those cheap phones to break down after 6-12 months of mining use.
« Last Edit: April 24, 2020, 07:01:42 pm by NiHaoMike »
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Online amyk

  • Super Contributor
  • ***
  • Posts: 8526
Re: Hotspot battery getting overcharged
« Reply #11 on: April 25, 2020, 12:02:58 am »
Then your protection board only does OCP and dead short detection, which is the bare minimum. Those are hard to come by as newer chips almost all do UVP/OVP as well.
A bit silly to use active components just for overcurrent protection in a battery pack when a much cheaper fuse will do just that. Most likely it's just defective in some way.
DW01-based battery protection boards are cheap and ubiquitous, and offer far more protection than a fuse.

Also, about letting the battery go to 5V, that would definitely damage it irreversibly and could've resulted in a fire. Lion cells should be treated with extreme caution.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #12 on: April 25, 2020, 10:55:44 am »
Then your protection board only does OCP and dead short detection, which is the bare minimum. Those are hard to come by as newer chips almost all do UVP/OVP as well.
A bit silly to use active components just for overcurrent protection in a battery pack when a much cheaper fuse will do just that. Most likely it's just defective in some way.
DW01-based battery protection boards are cheap and ubiquitous, and offer far more protection than a fuse.

Also, about letting the battery go to 5V, that would definitely damage it irreversibly and could've resulted in a fire. Lion cells should be treated with extreme caution.

I never said I charged the battery to 5V! I applied 5V to the protection board detached from the battery and used a power supply and load to simulate a battery being drained and it did nothing to stop the "battery" from working up to this voltage so there is no OVP on the board itself.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #13 on: May 15, 2020, 12:54:58 pm »
I will be buying a new li-ion battery pack that fits the footprint of the original battery by keeping the plastic case and just replacing the battery section.

Will be using two diodes back to back to drop the voltage to around 4.2V but currently only have Schottky diodes that drop the voltage by 0.17V so the battery will only be charged to 4.11V thus reducing the effective capacity.

Are there any diodes that I can use with a voltage drop of 0.8-1V? I could use four diodes to do this but I'd like to keep the component count low as I am not sure how much I can fit in that case when the new battery arrives.
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9322
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Hotspot battery getting overcharged
« Reply #14 on: May 15, 2020, 02:40:33 pm »
Will be using two diodes back to back to drop the voltage to around 4.2V but currently only have Schottky diodes that drop the voltage by 0.17V so the battery will only be charged to 4.11V thus reducing the effective capacity.
That's close to perfect! If you need more capacity, buy a slightly bigger battery.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #15 on: May 15, 2020, 03:27:15 pm »
Will be using two diodes back to back to drop the voltage to around 4.2V but currently only have Schottky diodes that drop the voltage by 0.17V so the battery will only be charged to 4.11V thus reducing the effective capacity.
That's close to perfect! If you need more capacity, buy a slightly bigger battery.

Theoretically speaking, yes close to perfect but now that I have it installed it's far from perfection!

It is reading 23% charged and wont go above that with the diodes in the circuit and it dies when it drops to 0% so I basically cut my capacity in 4  |O

Not sure why that's the case since the voltage drop when the battery is fully charged (4.2V) is only 0.6 V!

Maybe I'm doing something wrong here. A circuit diagram might be useful in this case  :-//
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9322
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Hotspot battery getting overcharged
« Reply #16 on: May 15, 2020, 03:48:58 pm »
Where did you get the new battery from? A lot of the cheap ones are rejects with low capacity. Try bypassing the diodes for a test.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #17 on: May 15, 2020, 03:50:42 pm »
Where did you get the new battery from? A lot of the cheap ones are rejects with low capacity. Try bypassing the diodes for a test.

It's a 900 mAh battery and I tested it's capacity and it was around 900 mAh so right on the money until I find a suitable replacement for the original capacity of 2930 mAH.

Works fine without the diodes so they're definitely the culprit here.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2248
  • Country: us
Re: Hotspot battery getting overcharged
« Reply #18 on: May 15, 2020, 04:02:55 pm »
Does the charger have to read the battery voltage at any point during the charging process?  It seems it would have to do that at the very beginning to make sure it's above 3V before applying full charging current.  And then after it terminates charging it will initiate charging again if the battery falls below 4.05V or so.  Diodes would block any reading of battery voltage.

 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9322
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Hotspot battery getting overcharged
« Reply #19 on: May 15, 2020, 04:26:56 pm »
The ESR of a much smaller battery could result in the usable capacity being greatly reduced if there's a high discharge current.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #20 on: May 15, 2020, 04:58:56 pm »
The ESR of a much smaller battery could result in the usable capacity being greatly reduced if there's a high discharge current.


It's not the capacity that is the issue, it's that the hotspot is not reading the voltage correctly and it seems that every time I take it off the charger and put it back on, it starts charging from 2%.

Will need to investigate and test this further...
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2248
  • Country: us
Re: Hotspot battery getting overcharged
« Reply #21 on: May 15, 2020, 05:13:12 pm »

It's not the capacity that is the issue, it's that the hotspot is not reading the voltage correctly and it seems that every time I take it off the charger and put it back on, it starts charging from 2%.


Is this with the diodes in place?  How can the hotspot read the battery voltage at all through two reverse-biased diodes?  Well, maybe I just don't understand what you're doing.
 

Online amyk

  • Super Contributor
  • ***
  • Posts: 8526
Re: Hotspot battery getting overcharged
« Reply #22 on: May 16, 2020, 03:44:46 am »
Will be using two diodes back to back to drop the voltage to around 4.2V but currently only have Schottky diodes that drop the voltage by 0.17V so the battery will only be charged to 4.11V thus reducing the effective capacity.
That's close to perfect! If you need more capacity, buy a slightly bigger battery.

Theoretically speaking, yes close to perfect but now that I have it installed it's far from perfection!

It is reading 23% charged and wont go above that with the diodes in the circuit and it dies when it drops to 0% so I basically cut my capacity in 4  |O

Not sure why that's the case since the voltage drop when the battery is fully charged (4.2V) is only 0.6 V!

Maybe I'm doing something wrong here. A circuit diagram might be useful in this case  :-//
In a newer post you said the battery was ~900mAh, and the original 2930. It's probably  using a Coulomb-counter to measure the state of charge, and disconnecting when it reaches 0% to prevent overdischarging, since 23% is awfully close to 900/2930. The diodes are going to confuse it, however, and that's probably where the rest of the missing percentage went --- they're dissipating the voltage drop as heat.

Also, a 900mAh cell in a circuit designed to charge a 2930 may be charged at over 3x its rated current. That will definitely shorten its life too.

The best solution is to find a 4.3V cell (they do exist), if you can't figure out how to modify the charging circuitry to lower the termination voltage.
 

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Re: Hotspot battery getting overcharged
« Reply #23 on: May 16, 2020, 08:56:16 am »
Will be using two diodes back to back to drop the voltage to around 4.2V but currently only have Schottky diodes that drop the voltage by 0.17V so the battery will only be charged to 4.11V thus reducing the effective capacity.
That's close to perfect! If you need more capacity, buy a slightly bigger battery.

Theoretically speaking, yes close to perfect but now that I have it installed it's far from perfection!

It is reading 23% charged and wont go above that with the diodes in the circuit and it dies when it drops to 0% so I basically cut my capacity in 4  |O

Not sure why that's the case since the voltage drop when the battery is fully charged (4.2V) is only 0.6 V!

Maybe I'm doing something wrong here. A circuit diagram might be useful in this case  :-//
In a newer post you said the battery was ~900mAh, and the original 2930. It's probably  using a Coulomb-counter to measure the state of charge, and disconnecting when it reaches 0% to prevent overdischarging, since 23% is awfully close to 900/2930. The diodes are going to confuse it, however, and that's probably where the rest of the missing percentage went --- they're dissipating the voltage drop as heat.

Also, a 900mAh cell in a circuit designed to charge a 2930 may be charged at over 3x its rated current. That will definitely shorten its life too.

The best solution is to find a 4.3V cell (they do exist), if you can't figure out how to modify the charging circuitry to lower the termination voltage.

Actually, it is able to charge to 100 percent however, as soon as I turn it off and back on, it thinks the battery is empty displaying 1 or 2 % and dies in a few minutes even though the battery remains fully charged (4.16 V).

Not sure why that is but surely it is not the right solution.

Maybe I'm connecting the diode up incorrectly so if anyone can draw a simple circuit diagram that would be helpful in troubleshooting this issue.

I'm sure there are 4.3 V batteries but without any proper designation, it would be impossible to differentiate them from the "regular" 4.2 V batteries as those are always marketed as 3.7 V nominal  :palm:

Another solution would be to replace the battery with the largest one that would fit the original plastic housing and let it charge to 4.28 V only when being used on the go.

I used to keep the battery in the hotspot while in use so that battery would remain at 4.28 V for months at a time.

Now I will just insert it when I carry it in my pocket to extend its lifetime and just replace the battery ever year or so for the next couple of years and get a new hotspot then which hopefully wont have this issue.



 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2248
  • Country: us
Re: Hotspot battery getting overcharged
« Reply #24 on: May 16, 2020, 02:42:27 pm »
Looking at this page on the Netgear site:

https://kb.netgear.com/24865/AirCard-Mobile-Hotspot-Battery-Information

I see that the 810S uses the W-7 battery.  The picture of the W-7 shows it's a nominal 3.8V battery.  In fact the earlier batteries on that page all say 3.7V, while the later ones say 3.8V.  That may just be marketing, but it also might mean that the 3.8V batteries charge to 4.3V.  If that's the case, your 4.28V charged battery would be ok.  This link indicates that it the case:

https://electronics.stackexchange.com/questions/244732/why-are-3-8v-lithium-ion-batteries-used-in-mobile-devices-rather-than-3-6v-or-3

However, the answers on that page indicate that charging to 4.3V may still reduce the service life of the battery.  An answer for that would be to find or build a 4.2V charger for your batteries.

In any case, you should make sure your battery is a W-7 which is labeled 3.8V.  If you're buying 3.7V batteries, that would explain the issues you're having.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf