Rejuvenating NiMH batteries.

[Pommie]

I've got some NiMH AAA batteries that have been stuck in a cupboard for a few years and are basically useless.
Does anyone know how to rejuvenate these batteries?
I've currently got one hooked up to an Arduino Nano doing the following,
charge until the voltage is 1.2V
discharge until the voltage is 0.75V
Both charge and discharge are done via a 100Ω resister connected to a digital pin so this will charge and discharge at ~40mA.
It's currently taking 4 seconds to charge and around 14 seconds to discharge.
I intend to leave this for a couple of days to see what happens.
Anyone got any ideas how this could be improved?

Mike.

 

[augustinetez]

Up the voltage and charge at C/10?

 

[Pommie]

I've doubled the current and increased the charge voltage to 1.3V.

Mike.

 

[Pommie]

Now,
Charge time 2 seconds.
Discharge time 9 seconds.

Looks like no capacity at all.

Mike.

 

[Nigel Goodwin]

They are knackered, dump them!

 

[Pommie]

They are knackered, dump them!

I think you're right. However, I'll keep up (probably foolish) hope for a few days just in case. I vaguely remember when I was in my teens that the local R/C shop offered a similar service for NICAD cells. Might have been snake oil but time will tell. And one thing I have lots of is time.

Mike.

 

[Nigel Goodwin]

I think you're right. However, I'll keep up (probably foolish) hope for a few days just in case. I vaguely remember when I was in my teens that the local R/C shop offered a similar service for NICAD cells. Might have been snake oil but time will tell. And one thing I have lots of is time.

Mike.

NiCd were prone to 'whiskers' which you could sometimes blow off with high current pulses.

If you want old batteries, we've got a big box full - mix of NiMh and Li-Ion - many are probably fairly good, I occasionally grab myself a few

 

[augustinetez]

At 1.3V & 80mA, it's going to take all night to get a charge in to them (assuming their not knackered).

Ni-MH need to be force fed to accept a proper charge - the charger I've got here (nothing fancy) is 3V @ 300mA per cell and yes they do get a little warm.

 

[Tony Stewart]

1.2V to 0.75V via 100Ω resister at ~40mA.
It's currently taking 4 seconds to charge and around 14 seconds to discharge

C=Ic dt/dV 10 * Tau=10* RC = 4s ( to 99% ) (est.)
Cch = 356 mF = 0.04A * 4s/0.45V , R = 10* Tau/C = 10 * 4s / 0.356 V = 110 with 100 in series. ESR = ? 10 ohms ? , ESR*C = Tau= 3.56 s to 63% target
Cdis = 14/4 * 0.356 = 14/4*.356 = 1.2 Farads
Double-electric layers have multiple ESR*C in parallel with short term and long term Tau's. With low current rates, they merge as one capacitance with an A-h or A-s value at constant current. C*dV=Ic*dt [A-s]

The sure sign of an undercharged battery is the rapid rise of ESR. A permanently damaged battery is similar but worse in that the ESR does not improve much due to the thickness of oxidation on the electrode-electrolyte interface. The oxidation rate increases when undercharged.

The best way to eliminate oxidation is to pulse high current without heat. Pavg <0.1W Yet heat damages the battery. Heat is the normal way to detect fully charged using a thermistor near the anode (?) to latch off a charger.

This would look like an inductive flyback pulse with the current battery polarity modulated by a vary low repetition rate to maintain a low avg. power dump.

This also creates a high enough voltage to bridge tiny slivers or dendrites in the electrolyte and possibly evaporate them. This can be detected by the change in flyback voltage. You can also use a very narrow current limited pulse from a slightly higher voltage such as 5V then monitor the change in average current using an RC filter to measure the average voltage rise.

 

[rjenkinsgb]

Does anyone know how to rejuvenate these batteries?

Charge to 1.5V, until the current has dropped to under C/20
Discharge via a low value resistor until absolutely dead; at leat 24 hours.

Do each cell separately!

Repeat 2 - 3 times.

If they will not reach 1.5V they are internally shorted. Discharging a large cap into a cell to zap the needle crystals may provide a (temporary, or not) fix to that.

 

[audioguru]

Some posts in this thread talk about older NiCad batteries, not newer Ni-MH batteries.
Many Ni-MH are rated for 500 charge-discharge cycles (about 1.4 years in a solar garden light).

 

[christinescoms]

1.2V to 0.75V via 100Ω resister at ~40mA.
It's currently taking 4 seconds to charge and around 14 seconds to discharge

C=Ic dt/dV 10 * Tau=10* RC = 4s ( to 99% ) (est.)
Cch = 356 mF = 0.04A * 4s/0.45V , R = 10* Tau/C = 10 * 4s / 0.356 V = 110 with 100 in series. ESR = ? 10 ohms ? , ESR*C = Tau= 3.56 s to 63% target
Cdis = 14/4 * 0.356 = 14/4*.356 = 1.2 Farads
Double-electric layers have multiple ESR*C in parallel with short term and long term Tau's. With low current rates, they merge as one capacitance with an A-h or A-s value at constant current. C*dV=Ic*dt [A-s]

The sure sign of an undercharged battery is the rapid rise of ESR. A permanently damaged battery is similar but worse in that the ESR does not improve much due to the thickness of oxidation on the electrode-electrolyte interface. The oxidation rate increases when undercharged.

The best way to eliminate oxidation is to pulse high current without heat. Pavg <0.1W Yet heat damages the battery. Heat is the normal way to detect fully charged using a thermistor near the anode (?) to latch off a charger.

This would look like an inductive flyback pulse with the current battery polarity modulated by a vary low repetition rate to maintain a low avg. power dump home improvement lynnwood wa.

This also creates a high enough voltage to bridge tiny slivers or dendrites in the electrolyte and possibly evaporate them. This can be detected by the change in flyback voltage. You can also use a very narrow current limited pulse from a slightly higher voltage such as 5V then monitor the change in average current using an RC filter to measure the average voltage rise.

I have a Logitech s-00116 bluetooth speaker with a NiMH battery pack that dates back to 2009 (about 12 years old). It didn't take a charge at first, but the charging indicator was blinking. Today, after leaving it to charge for 2 days, it held enough charge to run the bluetooth speaker for about 15 minutes. If I keep deep-cycling the battery will the charge capacity keep improving? I don't want to bring it back to the original 6-hour run-time, just something reasonable until the new battery pack I ordered arrives.

 

[Tony Stewart]

The purpose of this charger is to pulse charge with very low power so as not to age the battery with high energy flow or heat rise, but restore the possible oxidation and dendrite growth. Feel free to increase the current. R/L ratios ought to be 100~200 for realism. This dumps < 0.5A from 100 us inductor charge every 500 us for a 20% d.f. Discharge time depends on ESR/L ratio. A slider is provided for ESR simulation.

There are probably much better designs than this, but I wanted to demonstrate how to pulse charge a battery cell with say < 0.5A and flyback to whatever the battery ESR dictates with less than 1mA steady state charging from a higher voltage.

Using any 0.1 to 1 Ohm, Vt< 2V, NFET and any 1 A diodes to rectify the flyback voltage as an ESR indicator while monitoring the actual Vbat average.

Speed up simulation to make changes in response quickly then slow down or stop to read the plots.
The Darlington just drops the emitter voltage below Vbat to avoid steady charging.

https://tinyurl.com/2ybqq5q9 my SIM