NiMH Not Charging as Expected

[Bo Deadly]

I made two 14.4V NiMH battery packs out of 12 2/3AAA cells. The batteries used to make the packs came charged. So I drained them at ~50mA each for about 5 hours. Then I tried to charge them with a current source. The charging went like this:

Left axis is voltage
Right axis is current in mA
X axis is minutes

Va - Voltage of Battery A
Vb - Voltage of Battery B
Ia - Charging current of Battery A
Ib - Charging current of Battery B
SVa - SMPS supply voltage for Battery A
SVb - SMPS supply voltage for Battery B

So the batteries pretty much charge up to over 14V within 5 minutes. Then the supply starts to run out voltage and the charging current drops. This is not what I expected. I expected the battery voltage would remain relatively low for hours. If I got 5 hours @ 50 mA out of it, I would expect it to consume the full 34mA the supply can provide for 10+ hours or so. But the batteries just go to well over the 14.4V nominal within 15 minutes.

What's going on here? What am I misunderstanding?

 

[abbey road d enfer]

So the batteries pretty much charge up to over 14V within 5 minutes. Then the supply starts to run out voltage and the charging current drops. 

 
When fully charged NiMH batteries reach 1.78V per cell, so you need about 24V, for 12 cells.

 

[totoxraymond]

How did you setup your PSU?

I remember a video on the eevblog about charging batteries with a bench psu.

Basivally you need to start charging in constant current mode, and switch to constant voltage at a certain point.

I know it might be obvious to you, your graph doesn't seem to show constant current.

Cheers

Thomas

 

[Bo Deadly]

Well that's a huge drag. It's amazing how many pages on the Internet talk about charging current and nothing about the actual voltage.

Fortunately multiple companies make the little 1-2W SMPS w/ +-24V out. Clearly I'll have to completely redesign and use a PIC.

But at least I now know my batteries probably aren't garbage and I can test other things like the AC filter.

I know it might be obvious to you, your graph doesn't seem to show constant current.

It's constant current at 34mA so below that it degenerates to a voltage source. The little SMPS (DETN01L-15) has a bit of a current source behavior to it such that I could have probably just connected it directly to the battery through a schottky diode.

 

[Gold]

I don't know if this helps but here is a link to the manual of the battery charger I use for the AGM solar batteries I use. There is a chart that shows the different charging modes it goes through.

https://www.cotek.com.tw/support_dataSheet.php?FileLevel=JCU0IyE=

 

[Matador]

Constant current only works if the terminal voltage can rise along with the cells to maintain the constant 50mA.  If you run out of voltage, current will fall as the battery impedance rises during charging.  It's essentially converting to constant voltage charging once the supply cannot rise any longer (aka. it'll take forever to charge, actually can cause destruction of the cells). 

You should budget 1.8V per cell of charging headroom, which is closer to 22V for 12 cells.

 

[Bo Deadly]

It looks like for low current charging, which is what I'm doing, a lot of datasheets [1] [2] [3] suggest using only 1.45-ish charging voltage.

So I removed one cell from my pack and now I'm getting a decent charge. It's dropped down to only ~10mA after 2 hours but for my purposes that might be ok for now. Previous tests have shown that the DETN01L-15 SMPS puts out 15.9V @ 1.7mA which comes out to 15.6V at the battery after the schottky diode. So for 11 cells that 1.42V / cell which should get me within 80% charge or so. Of course it will take a looong time and the pack will only put out a mean of ~13.7V. But +-13.7V is still decent for audio circuits.

[1] GP Batteries Selection Guide
[2] YUASA NiMH Datasheet
[3] Ansmann NiMH AA Datasheet