Author Topic: Leaking AA and AAA batteries  (Read 46131 times)

0 Members and 1 Guest are viewing this topic.

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #75 on: December 13, 2018, 11:39:34 pm »
I run uC's below their specified voltage. I can get away with 1.5v on an Atmel chip but 1.2 gets a little tricky.
What I do when voltage level isn't an issue is to run rechargeable batts like the Eneloops which I find excellent.

I'm using Lithium AA's now and they seem to be good so far.

Rob 
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #76 on: December 13, 2018, 11:59:38 pm »
I've never had a lithium leak. Not once! EVs are full of them, did you ever hear about them leaking? Not me, my EV has never had a leaking Lithium. Panasonic 18650's, never had a leak.

I think it's time to roll out the old saw:
Quote
The plural of anecdote is not data.

Just because you personally have never seen a Lithium battery leak doesn't mean that they don't.

I've seen several leaking CR2032s.

That may be brand related. I've not had a leaking 2032 either. Got pics?

I came across a Mac IIfx that was completely destroyed by a leaking lithium PRAM battery. It ate some traces right off the motherboard and badly rusted the metal parts of the chassis, it was a near total loss.

I've also had a leaking lithium coin cell a bit larger than a CR2032 on a Mortal Kombat arcade board which leaked and rotted contacts off several IC sockets.

Oh and I've had a CR123 leak.

Lithium batteries are less leak prone than typical alkalines but they certainly can and occasionally do leak. I suspect some chemistries are far more leak prone than others.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #77 on: December 14, 2018, 12:03:17 am »
Oh I also had an energizer lithium AA leak after I put it in a cheap solar pathway light to see what would happen. I'm not sure that counts though, the battery clearly said not to recharge it.
 

Offline MrMobodies

  • Super Contributor
  • ***
  • Posts: 1906
  • Country: gb
Re: Leaking AA and AAA batteries
« Reply #78 on: December 14, 2018, 03:05:22 am »
There was one time when I had alkaline leak in a low power led light about 6 years ago. I put in new Duracell I can't remember which one but I paid for the "best". I think it switched on in the bag and went flat or I forgot to turn it off. A couple days later when I went to use it it wouldn't turn on so took the batteries out they all leaked.

I took home, took it apart to spray with alcohol and scrape off any corrosion I can find.
One of the wire was corroded to the terminals and the terminals so I cleaned it out and replaced the wire but I couldn't find any other damage on the led board.

I put in some of the Ansmanns batteries and it worked, left it for a few days and it was fine no leaking.
 
A couple of weeks later I took it with me and the light wouldn't turn on. Okay no problem they have gone flat and I'll swap them over with some I charged up, I took them out and the batteries were corroded and I opened the light and the board and everything inside was corroded.

The alkaline was still in there despite spraying it with alcohol and scraping it many times and it ruined the four rechargeables inside and they were expensive at the time.

It was a blue 72 led lamp that I got from Sainsbury's similar to the Rolson 72 LED Inspection lamp.


I should have known better to put cheap alkalines in to see if it would continue to corrode.
« Last Edit: December 14, 2018, 03:16:46 am by MrMobodies »
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #79 on: December 14, 2018, 04:26:28 am »
I took home, took it apart to spray with alcohol and scrape off any corrosion I can find.

I opened the light and the board and everything inside was corroded.

The alkaline was still in there despite spraying it with alcohol and scraping it many times

Maybe you have learned from experience, but this is not the right way to clean leaky battery damage. You need to use the best universal solvent--water--and lots of it. You need to rinse with copious amounts of distilled water. Submerge the board, scrub it with a soft brush, and rinse with more running water. After rinsing, leave it to dry in the open air in a warm place for several days. Where to avoid using water is in the light assembly and reflector. But assuming you can disassemble the light and remove the board, all the disassembled pieces should get the thorough wash treatment.

Cleaning anything is like cleaning clothes. You need to get the contamination into solution and then remove the contamination from the scene by rinsing it away.
 
The following users thanked this post: MrMobodies

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6923
  • Country: ca
Re: Leaking AA and AAA batteries
« Reply #80 on: December 14, 2018, 05:48:36 am »
Isopropyl alcohol is not very good for cleaning alkaline battery corrosion. You have to neutralize the white potassium carbonate with an acid. I use a q-tip and vinegar to clean that up, then a rinse with water or alcohol.

Asphalt (bitumen) was originally used as the seal on AA and AAA batteries. It survived the chemicals.
I guess it was expensive or environmentally unfriendly. Kodak AA's used a hard epoxy glue seal and that worked too.
Whatever Duracell uses now is total garbage, they all leak.
 
The following users thanked this post: tooki, MrMobodies

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #81 on: December 14, 2018, 06:17:33 am »
Oh I also had an energizer lithium AA leak after I put it in a cheap solar pathway light to see what would happen. I'm not sure that counts though, the battery clearly said not to recharge it.

But they are not ionic, right? As long as they don't eat copper trace, they are far less troublesome.

Honestly I don't know, I only mentioned that they can leak, it may well not be as corrosive.

The 3V lithium batteries used in the old Macs sure are corrosive though, I've never seen battery leakage that corrodes like those do. Rots out all the metal it touches and the vapors rust everything else.
 
The following users thanked this post: cdev

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #82 on: December 14, 2018, 09:22:40 am »
But they are not ionic, right? As long as they don't eat copper trace, they are far less troublesome.

They sure are ionic. All chemical cells contain electrolyte, and all electrolytes are ionic by definition. However some electrolytes are more corrosive than others. In the case of lithium cells there is very little electrolyte present and it is absorbed in some kind of separator material so there is not really very much liquid to leak. This is why lithium cells are safe from a leaking point of view.
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3632
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #83 on: December 14, 2018, 11:24:01 am »
Some computers used CR or BR type cells for memory retention, others used LiSOCl2. The latter contain highly corrosive liquids (the cathode of the reaction is a liquid, thionyl chloride).

Some electrolytes are corrosion inhibitors:
"LiBF4 appears to have by far the lowest corrosion rate. There is little difference in the amount corroded... stored for 40 hours and 140 hours."
"Depression of corrosion... was attributed to the formation of a stable passive layer on the surface."
"... the LiBF4 is the most effective for preventing the corrosion."

All chemical cells contain electrolyte, and all electrolytes are ionic by definition.
That is true, but not all chemicals added to a battery formula to act as electrolytes are ionic—in the case of lithium ion batteries, the electrolytes are aprotic. It is only by solvating lithium ions that they become ionic complexes.
« Last Edit: December 14, 2018, 12:02:38 pm by helius »
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Leaking AA and AAA batteries
« Reply #84 on: December 14, 2018, 11:45:41 am »
Isn't the voltage lower in hydroxides?
It's around 1.45V freshly charged. Then it drops to around 1.3V after some time (without external discharging). But generally this is not issue. It's just that non rechargeable batteries have steeper discharge curve. Voltage when most of the energy is spent is around the same for both types. So unless device is a garbage which can no longer work while there still is a lot of energy left in the battery, there won't be any issue using NiMH.
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #85 on: December 14, 2018, 03:14:59 pm »
It's not about how much energy is in the battery, it's about the working voltage of the circuit. I've tested ATTINY12V's down to 1.2 volts but ATTINY5's cut out below 1.5v. So I can design a single battery app that uses under 100uA and it will run fine on an Alk but not on a NiMH using 12V's and even get away with ATTINY5's in some apps.
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Leaking AA and AAA batteries
« Reply #86 on: December 14, 2018, 04:18:31 pm »
It's not about how much energy is in the battery, it's about the working voltage of the circuit. I've tested ATTINY12V's down to 1.2 volts but ATTINY5's cut out below 1.5v. So I can design a single battery app that uses under 100uA and it will run fine on an Alk but not on a NiMH using 12V's and even get away with ATTINY5's in some apps.
It's about you should not use attiny directly powered from a single cell to begin with. If it cuts off at 1.2V, your design is a piece of crap. If you look at discharge curves, NiMH actually stays above 1.2V for more time than alkaline, especially at higher loads.
Edit: and using them below 1.8V is not up to spec at all. It may cause problems, especially with using EEPROM.
For such use there exist adequate MCUs like this https://eu.mouser.com/ProductDetail/Silicon-Labs/C8051F921-G-GM?qs=sGAEpiMZZMtjWZqwEMjY%2f787M7OqHS%2f9yWzbyePZbJY%3d
« Last Edit: December 14, 2018, 04:30:44 pm by wraper »
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #87 on: December 14, 2018, 06:33:43 pm »
I run the designs at 1.55 volts, the voltage of Alks, not at 1.2 volts, that's why I don't use NiMH. I can run down to 1.2v to test my overhead but 1.55v is where my chips work with no problems.

But I suggest you don't do it because it seems to make you angry or something that I actually find quite funny.

And the 32 pin chip you linked cost multiple times more than a TINY with 8 pins or SOT23-6 with 6 pins which runs in the picoamps when in sleep mode and I can program in AVR assembly. I also have to use an oven to put your chip on a board which makes prototyping a pain.
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #88 on: December 14, 2018, 07:26:58 pm »
I think you're absolutely right and I'm never going to do it again......... feel better? I certainly do.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #89 on: December 14, 2018, 08:09:32 pm »
I run the designs at 1.55 volts, the voltage of Alks, not at 1.2 volts, that's why I don't use NiMH. I can run down to 1.2v to test my overhead but 1.55v is where my chips work with no problems.

The open circuit voltage of a brand new, freshly manufactured alkaline cell is about 1.55 V. But how long do you think it remains that way? It doesn't remain there for very long.

http://data.energizer.com/pdfs/e91.pdf

The voltage remains above 1.5 V for about 2% of the service life of the battery. So if you design something that relies on an alkaline battery having a voltage above 1.5 V your design isn't going to work.

Nominal voltage alkalines: ~1.5V
Nominal voltage NiMH: ~1.2V

The "nominal" voltage of a battery is exactly what the word says,  the "named voltage" printed on the side. Dictionary definition: "nominal" = "in name only". A nominal value is not the same as the actual value.
« Last Edit: December 14, 2018, 09:30:29 pm by IanB »
 
The following users thanked this post: Marco1971

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Leaking AA and AAA batteries
« Reply #90 on: December 14, 2018, 08:32:22 pm »
And the 32 pin chip you linked cost multiple times more than a TINY with 8 pins or SOT23-6 with 6 pins which runs in the picoamps when in sleep mode and I can program in AVR assembly. I also have to use an oven to put your chip on a board which makes prototyping a pain.
It costs 20-30% more than attiny12V and is way more capable chip. 24 pin QFN variant is only a little bit larger than SOT23-6.
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #91 on: December 14, 2018, 09:11:29 pm »
It depends on the current. At under 100 picoamps it's going to stay around 1.5 volts a long time.
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #92 on: December 14, 2018, 09:22:48 pm »
I'm using an ATTINY5 at 34 cents vs $1.62 (Mouser). Why would I want to pay more for a chip than I need it to do? Your chip is a pain to prototype and program. I have to keep the chips in sealed bags so I can heat oven them without damage. I can mount an SOT23-6 on a simple adapter and program it on a simple interface. If I want larger chips I still us AVR's they're my favorite chips, I have assembly down pat and I can do anything I want on them. I stock many AVR chips and it makes life easier.

 
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Leaking AA and AAA batteries
« Reply #93 on: December 14, 2018, 11:18:06 pm »
It depends on the current. At under 100 picoamps it's going to stay around 1.5 volts a long time.
100 picoamps is when MCU is shut down and does nothing, and actually that current is a bit higher than 100pA at room temperature. As you wasting 98% of the battery anyway, you could just use 2 smaller batteries in series and use full amount of their capacity. Even single 3V CR2032 would last longer than 2% of AA alkaline. Not to say it would be reliable design. Not something that works only with lucky draw in MCU die lottery and in narrow temperature range.
« Last Edit: December 14, 2018, 11:25:17 pm by wraper »
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #94 on: December 15, 2018, 12:28:28 am »
That's wrong:
First the active supply current of the uC is under 10uA, on interrupt (31uS on state for uC in my present design) it changes state and then goes back to sleep. The total current of the circuit is the 31uS acitive current period and the remaining current at <60pA for the rest of the remaining time. IF the uC only needs to wake up on interrupt every 10 minutes for 31uS then the current is basically 60pA.

Two: AND since I am running the uC at 1.5v both the active current and the power off sleep state are lower. (hint, I can do this because I'm using a very low frequency oscillator and the lower the freq the lower the uC operating current and the lower the operating voltage it's capable of.)

Three: No, that's wrong again (see 2), when you raise the operating voltage you raise the operating current in all uC states. At 3v the active and sleep current is about twice as high.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #95 on: December 15, 2018, 12:44:17 am »
I run the designs at 1.55 volts, the voltage of Alks, not at 1.2 volts, that's why I don't use NiMH. I can run down to 1.2v to test my overhead but 1.55v is where my chips work with no problems.

The open circuit voltage of a brand new, freshly manufactured alkaline cell is about 1.55 V. But how long do you think it remains that way? It doesn't remain there for very long.

http://data.energizer.com/pdfs/e91.pdf

The voltage remains above 1.5 V for about 2% of the service life of the battery. So if you design something that relies on an alkaline battery having a voltage above 1.5 V your design isn't going to work.


Maybe we finally found a use for the Batteriser  :-DD


Seriously though, that's a piss-poor design if it relies on the higher terminal voltage of an alkaline cell and won't work well with NiMH. Maybe for a quick hobby hack that's fine but there's no excuse for a commercial product to have such limitations these days, it's so easy to avoid them.
 
The following users thanked this post: wraper

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #96 on: December 15, 2018, 12:57:29 am »
You're absolutely right! I'll never do it again, feel better? I certainly do.

Is a common quartz wall clock a "commercial design"? See how long it runs on a NiMH battery.
« Last Edit: December 15, 2018, 12:59:33 am by Robaroni »
 

Offline Robaroni

  • Frequent Contributor
  • **
  • Posts: 433
  • Country: us
  • Retired EE
    • Design Specialties
Re: Leaking AA and AAA batteries
« Reply #97 on: December 15, 2018, 01:05:06 am »
No Ian, I've tested the uC's down to 1.2 volts before they cut out. So an Alk at 1.45 volts will run my circuit. Again, the lower the operating frequency the lower operating voltage of the circuit AND the lower the current.

But, hey, don't do it, I couldn't care less.
« Last Edit: December 15, 2018, 01:07:06 am by Robaroni »
 

Offline Gary350zTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: us
Re: Leaking AA and AAA batteries
« Reply #98 on: December 15, 2018, 01:07:41 am »
Nominal voltage alkalines: ~1.5V
Nominal voltage NiMH: ~1.2V

The "nominal" voltage of a battery is exactly what the word says,  the "named voltage" printed on the side. Dictionary definition: "nominal" = "in name only". A nominal value is not the same as the actual value.

Words have many meanings. The definition I used here is Merriam Webster 3B (see attachment); relating to theoretical size, or approximate, which is what is typically used in engineering. In my example above I used the symbol for approximate (~), which seems to fit this definition and situation. The actual alkaline voltage can range from ~1.6V down to 0.8V (or 0.0V ;D).

In this case, "nominal" meaning "named voltage", or "theoretical voltage" seem to be both right, even though the battery voltage can vary greatly. Not a big deal.

What surprised me recently is that the Energizer Ultimate Lithium L91 battery spec says it has a nominal voltage of 1.5 Volts, but the application manual says the voltage can range from 1.74V to 1.83V.  (I learned this from this thread. :-+)

This caused a problem for me recently. I have several wireless thermometers and weather stations. The manuals warns against using lithium batteries because they have "to much power" (bad description from the manual writer). This didn't make any sense to me because the thermometers have very low power consumption. The thermometers worked good with the lithium batteries, until recently when I got a new thermometer that would not work with lithium batteries, but worked with alkaline batteries. The higher voltage of the lithiums (1.8V) is too high for the circuitry in that thermometer.
« Last Edit: December 15, 2018, 01:10:06 am by Gary350z »
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6923
  • Country: ca
Re: Leaking AA and AAA batteries
« Reply #99 on: December 15, 2018, 01:38:33 am »
I have to use lithium batteries in wireless thermometer/weather station transmitters.
It gets too cold during a Canadian winter, alkaline batteries freeze up.

Some wireless transmitters I have killed using low internal resistance batteries or a bench power supply.
I find the fast rise time dV/dt causes the IC to latch up. I had to add a 47R series resistor to stop it.
This might be "too much power".
The voltage didn't matter, I think it's all 3.3V IC's anyhow.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf