Surrender to the AA power - AA battery pack 10S11P (27Ah max.) with 50W solar

Greetings solar enthusiasts,

I wanted to show something I built half year ago and which I felt shy to show given its rarity and to some extend, lack of sense.

Surrender to the AA powerwall!!! A set of 10S 11P NiMh 1,2V cells with a total of almost 27Ah capacity able to supply (as experienced for now) a constant 70W AC while using a DC/AC power inverter and a constant 300W when directly drawing from DC.

This rather than a "usable" installation is a model or toy. Thanks to the low voltages and stability of NiMh AA batteries I have been able to make mistakes without overheating or exploding components. However, nowadays I have permanently attached a small solar panel to it (50W); removed the inverter and added a distribution port with multiple 5V USB connectors and 12V connectors and I use the battery for my office corner to charge electronics, use lights (12V bulbs) and a small 150W-300W heater for those days where the temperature is on the edge. Obviously, when I use the heater, the battery is drawn in no time (30min), as well as cells get a bit warm. NiMh cells can handle decent currents >1C but are not meant to be used this way. Ideally no more than 0,5-1A.

The model includes a small LCD display with a switch that I can use on demand to verify the input voltage of the panel and the incoming current, so I can not only verify that the battery is being charged but I can optimize the panel inclination and positioning.

At this moment I am exploring ways to utilize 12V DC as for finding more uses to the battery. My lightning consumes only 5W per bulb, phone 10W and laptop 50W via inverter which I do not often need. With summer arriving in Europe and the longer days, I often have surplus of energy. My next addition is a 130W heating element for warming up water and prepare tea.

No BMS, aiming to manually balance the pack maybe once a year with a AA battery charger. The battery is full at 14,4V and discharge has to be stopped at 10V. NiMh cells allow to be discharged down to 5% theoretically with no sever damage.

Why I mentioned it makes not much sense; is because it is a heavy and bulky battery pack compared to its capacity and power. Also more expensive per Ah compared to LiFePO alternatives and mainly pursued due to donation of the cells.

All comments and suggestions are welcome. Ask what you like.

Holy cow! I never think I'd ever see a bank made out of AA nimh like that. Cool for testing. Thanks for pointing out that this is not "production ready". Fun.

Of course many would say to reduce cell count (or increase your capacity) by using larger cells, like the 10,000ma types. That's the largest size you can get as a consumer due to the patent-encumbrance of NiMH: (aka "Who killed the electric car" - classic)


Paranoid Shop Teacher time just to play it safe:
Do NOT cut open any of those cells on a flammable surface. When exposed to air, Nimh particles have a nasty habit of setting flammable materials on fire. Kind of like hot burning cigarette ash.

My thought here is to invest in a hobby-charger with balance leads that can handle nimh, do capacity tests, internal resistance and so forth. Unless you are just going to trickle-charge the bank with a 0.1C current for 12-14 hours, which in the long run is unhealthy, but many pundits of low-and-slow like the 1980's will argue.

For single-cell consumer level testing, charging, the Maha Powerex MH-C9000 is the go-to reference. Lots to choose from of course - this is just one classic example.

Because of the difficulty of doing nimh charging right - (not missing the usual dv/dt terminations etc) with simple solar cc/cv, if you want to be "totally solar", you might consider a two-step process: Use standard panel and and LFP or small agm battery to charge up first, and then use THAT when charged up, as your dc source for the hobby-charger, which most run on dc or ac as a source - but check. Just some thoughts.

Thanks for sharing - never thought I'd see anyone go that far with AA nimh!

Thanks for your input Substrate. I am keeping my design very simple for now. I have the battery permanently attached to a 20W behind the window solar panel (so effectively 10W) and from time to time I connect it 1 day or 2 to a 50W outdoor solar panel when I used too much capacity at once. The PWM charge distributor cuts off the current at 1V per cell on discharge (perfect) and the charging can go as far as 1,44V per cell (a bit too much, better just 1,34V). Theoretically full charge voltage can be set; but this aliexpress controller I have seem to not obey what I input.

With such low charging currents (10W <<0,1C and 50W max. 0,3C) I do not expect to see overcharging problems.

The only thing I am concerned about now is safety. So far it seems the batteries are not getting warm or deforming but trying to foresee If there are safety gaps from where things could go wrong.

Cool - I see the logic behind the low-rate charge. But 27ah of low impedance nimh is nothing to sneeze at either.

Maybe put some fusing here and there, because as of now, it looks like your interconnecting wires will act as temporary flammable fuses inside that wood/cardboard desk drawer.

How about some small DROK or similar voltmeters across each cell so you can watch in real-time? Fun, but maybe informative.

Maybe get an inexpensive handheld IR thermometer with laser pointer at the home-improvement store to baseline your cell temps under load (since by hand you can't tell minute differences) while under load?

Those holders springs can themselves be a source of high-resistance, and looks like 4 of the bays went bad and you had to put 4 of them in their own holder?

Just some fun options - no major critique. I'm still blown away by the use of the AA's for fun, because at that level, that's not cheap!

Connecting Nimh batteries in parallel is tricky or generally to be avoided as the battery nearing full charge starts to warm up and hog more current.
LSD (low self discharge) batteries like yours might behave better in this regard as they dont get unbalanced as fast as ordinary old-school nimh did.

They have decent DoD and cycles (ref):

Using NiMH batteries, more than 3000 cycles at 100 % Depth of Discharge (DOD) have been demonstrated.

Click to expand...

But the charging scenarios always worried me... for example, this:

It is difficult, if not impossible, to slow charge a NiMH battery. At a C rate of 0.1C to 0.3C, the voltage and temperature profiles do not exhibit defined characteristics to trigger full-charge detection, and the charger must depend on a timer. Harmful overcharge can occur when charging partially or fully charged batteries, even if the battery remains cold.

Click to expand...

...and...

• Nickel-based batteries are best fast charged; a lingering slow charge causes “memory.”

Click to expand...

@svetz You are completely right. And you got me fired up in a good way..

This project avoids the use of properly detecting the dv/dt signal from the cell(s) which demands a much higher current charge, like 0.5c or higher to do properly to avoid missed terminations. For that, we'd need much smarter charger, balance leads and so forth.

So the other option is to charge at the 16-hour trickle rate, preferably always starting from full discharge (0.9v / cell individually, or for a bit more protection in a multi-cell environment 1.0v) But yeah, that's a lot of cells, so the op started out at least by charging each to full individually in a smart charger.

The funny thing that most pundits of "low and slow" and no heat don't realize is that the 16-hour charge rate or "standard charge" printed on the side of consumer nimh is purely a TEST CONDITION, and not the best way to charge the cells in practice! Ie, this is the only technique a manufacturer can guarantee that a cell is charged to full when someone decides to do a discharge capacity test, because the manufacturer has no way of knowing that using the higher-current charge techniques by an end user will get it right! Ie, not stopping short, nor missing a termination. The manufacturer doesn't know if the end user has a crap charger, so the trickle charge method is the most guaranteed way to do it for determining capacity on the fly to them.

But as you pointed out this standard charge does NOT mean it's the best way for longevity! For one, nimh is "exothermic" and is supposed to get warm! In fact, if it doesn't, that means that the charge is inefficient, and actually contributes to degradation!

So what is warm?
When you can put your fingers on the cell, and not pull away in pain. Oh it may be very warm, and highly uncomfortable, but if you can keep your fingers on it for any length of time, that's ok.

So what is too hot?
When you simply cannot keep your fingers on a cell, usually pulling away immediately.

But consumers don't see the actual REAL spec-sheets from quality manufacturers recommending at least 0.5 to 1C to obtain an efficient charge. It just so happens to coincide with the ability of a cell to emit the dv/dt signal (temporary voltage drop), but charging at trickle rates slow-roasts them to death, even if you can't feel the heat.

Much of the consumer myth stems from using super-fast chargers that are poorly designed destroying cells, and from those who want to keep their remote-control AA's, or batteries from their baby-swings working much longer than they were designed to - nevermind poor maintenance, balance, and high internal impedance from such abuse in toys, flashlights, and other sundry items.

Long story short:
Use at least 0.5C to provide an efficient level of charge AND force the cell to produce a quality EOC dv/dt signal. It's a two-part reason. Many who agree only do so because of the dv/dt signalling, forgetting that it also means efficiency during charge.

The sad thing is that there is too much shade-tree backyard-mechanic myths about proper nimh charging to overcome from a consumer standpoint. The tide of misinformation is just too much. So much so, that even manufacturer's bend to the will of the "low and slow and no heat" consumer is always right desires in marketing materials, and the proper education about nimh gets lost.

In this project, simplicity was the goal, with accepted levels of degradation from doing so, just for fun apparently.