Charging 12 NiMh Batteries in Series?

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
This is a simple enough question I hope.

What are the downsides, if any, to charging 12 AA sized NiMh cells in series at 0.1C?

My initial thoughts were that different internal resistances of the cells might be a problem. However after 30min of searching both here and Google I haven't had much luck finding anything definitive.

Thanks in advance for your help :)

-Capo
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
I will be using a constant current setup based off the design from this website.

The schematic for it is shown below.
charger.GIF


The only key difference between whats seen above and what I will have is a few of the resistor values to allow for a different charge current.
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
At the lower end of charging currents, the 12S cells will tend to self balance, but this assumes that some are not starting out fully discharged and others almost already charged when you start.

I am not a charger expert, so your setup might be fine, just be careful with over charging. A more advanced charger will detect charge termination, but you can substitute a timer like other lower end chargers.

In the end, it comes down to how much you have invested in your cells. A lower end charger will charge them up to some % of "full". A better charger will charge them up more optimally and the number of cycles they last will be higher. At the point that I started having a couple of hundred dollars of R cells / packs around, I invested in a Triton II.
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
the 12S cells will tend to self balance, but this assumes that some are not starting out fully discharged and others almost already charged when you start.

This is what I was hoping to hear. The 12s cells will be a closed pack so that it is also discharged in series. So with a bit of luck they should all come close to even after after use.

I have a Maha Powerex (MH-C9000) charger already for the every day cells. This charging in series business is more of a side project aimed at some bike likes im making.

I pulled the bits out tonight and got the circuit shown a few post above going. Its quite easy to vary the current with a pot but i cant seen to get the current high enough.

I will post a how to once i get it polished and condensed if anyone is interested.
 
Last edited:

tabetha

Newly Enlightened
Joined
Apr 12, 2009
Messages
184
Location
NORFOLK UK
Not 12 but 8, is what my radio gear takes for my models, and I always charge them in situ, makers even give a plug for doing this, never had a problem, and always use NIMH, so would happily charge 12 in a row.
tabetha
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
so would happily charge 12 in a row.

Glad to hear, provided it charges well it should be quite a compact and useful pack.

A few stats:
  • Froward Voltage of 14.4V (16.8 fresh off charge)
  • 2.2Ah Real Capacity at 14.4V or 31.68Wh (Average cell capacity of 2,200mAh from a MAHA analysis)
  • 12-16 Hour Charge time, but up to 20 wont do any damage.
  • 380g in weight including battery holders.
  • Cost of $28.7USD for the cells and $2 for 2 6 Cell square holders
  • Under $5USD for components for charger, no PCB required.

I will get some pic's up by the end of today if i get around to it.
 

VidPro

Flashlight Enthusiast
Joined
Apr 7, 2004
Messages
5,441
Location
Lost In Space
not to change your ideas or anything, but a good combo of fast charge and final charge tapering with topping for full charge balanced finish, is somewhat voltage based, not just current limited.

meaning a LAME simple linear transformer charger of the correct voltages, will do:

1) a Faster charge when the voltage is Low, allowing for a quick charge to a lesser percentage, so you can get SOMETHING fast, even if it doesnt last

2) will slow down as the voltage differential (between charge and battery) goes down, matching the ammount of chemicals that still need to be "charged" instead of continuing the assault on lesser items that can change up.

3) will slow way down and finsh up as the voltage on the batterys reaches fully charged. and allow for full topping, balancing, and slow continuous charging without going beyond the slow charge rate specs.

as Long as the end of charge is below the 1/10th C rate , the inital rate can be faster AND still have a DUMB charger. this can help the situation of charging all "slow" and would be better for things like hard hitting incan hotwires and stuff like that.

that can often be done with the stupidest lamest cheapest dc powering source that one can dream up
charging FAST untill reaching a specific voltage, then as the voltage reaches the power supply voltage by default it slows down.
or
using more intelectual microprocessor things trying to accomplish the same goals by monitoring and changing, then screwing it up :whistle:

its like a "Teter Toter" of voltage vrses current, via the two power sources.
as long as you take into account all the possible voltages, and simply manually check the peaks at both ends, then toss in some leeway for things not being exact.

say like a nice 2 amp charge that tapers down to 1/20th C as the voltage of the batts is Holding about 1.4v each item.

usually you can find a transformer that directally meets that, and just add in a bridge rectifyer and a cap , about $12 and it will do the best of all worlds OLD SCHOOL. if it doesnt fit perfectally adding in a bit of resistance or changing the diodes for different types, gives you the rest of the leeway you need. so you get close, and check its progress, most often you will find that this stuff handles itself fine without so much techno crud.

as batterys weaken and dont work so well the internal resistance goes Up, and therfore the voltage seen on final charge goes up, so the LAME method also covers the weak battery pack , by cooling out at the higher voltages too, and when the cells are new, the exact opposite. so the thing still keeps plugging away, new or old.

plus its reasonable power efficient, vrses burning up power in something other than charging a battery, like the heat from a regulator.

Bat - Charge
:crazy:----:eek:
:)----;)
:D----;)
:D----;);)
:D----;););)
:D----;);););)
:D----;););););)
:D----;);););););)

ok someday i will make the right picture for it too
 
Last edited:

VidPro

Flashlight Enthusiast
Joined
Apr 7, 2004
Messages
5,441
Location
Lost In Space
also this can be done with a voltage controled Current Limited, simple voltage supply, that just cant go any faster.
when the battery voltage is LOW, the charger pushes out its max possible current (not nessisarly good for cheap wall warts) , and again because the voltage is controlled (somehow) the differential between the two power sources continues to get smaller, and so to does the current flow between them.
So the same lame method can be done by controlling the voltage, if you cant find a transformer of the correct voltage, which is rare, because it SEEMS like this is WHY the voltages of the transformers exist, to charge batteries and replace power on devices that use them.
 
Last edited:

Greg G

Enlightened
Joined
Oct 17, 2007
Messages
772
I use a Triton2 charger that I picked up on ebay for my 12AA Mags.

They're not very expensive used.
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
Thanks for all the input, its given me a bit to think about.

I decided to sick with the current regulator for one key reason; heat. Essentially the cells will be contained as a single pack in a small box which doesn't allow a great deal of heat dissipation. So if i was to fast charge the cells using a matched voltage I would run the risk of the cells getting to hot and ruining them. I'm hoping that by only supplying 0.1C they will keep at a reasonable temperature.

Anyway, I have quite a few 1.5A 16V AC plug packs around so I decided to use one of them as my supply.

As its AC i needed to add in a bridge rectifier and capacitor (as was mentioned above) before leading on to the rest of the circuit.

I first set it out on a breadboard to check it was all working.
2ebr29v.jpg


Then with some ugly ninja soldering I managed to condense the same circuit as above down to what you see below. This is the a top view.
2eciyas.jpg


..and a bottom view.
wvx4wz.jpg


Then heat shrink'ed the whole thing for strength and protection.
if3o6p.jpg


I followed the schematic show a few posts above pretty much exactly apart from three changes.

  • Added rectifier and capacitor to change AC supply to DC.
  • Did not include the indicator LED
  • Used a 2.2Ohm 1% resistor to give me 230mA constant current.

I tested the current with an DMM inline to the cells, reading was 230mA.

I have yet to leave it on to charge the cells but I will post how it goes.

-Capo
 

VidPro

Flashlight Enthusiast
Joined
Apr 7, 2004
Messages
5,441
Location
Lost In Space
the heat from charging is generally the "waste" from that which does not turn the chemicals, when all the chemicals are ready to turn the "charge acceptance" is higher , so "Power" goes where it was intended (charging), instead of running out as heat (wasted).
when few chemicals are left that even Can turn, and the current is still pounding in, that is when the most heat comes out in waste.

so the lame method still works, because the voltage of the battery is low when more chemicals are in the other state, ready to be pushed back to the charged state. Heat of any SAME current charge always gets higher and higher neerer to the end of charge, as more of the power from the charger is not doing what it was sent there for, and wasted as heat instead. so even with the controls you have there will be Some minor heat building up at the end of charge. (assuming you dont run out of voltage Anyway)

With the lame method, the "fast" part of the charging occurs when there is the least ammount of heat from waste, and it is slows down so much neer the end that there is little waste heat , when say on a smart V-Drop termination charger there would be the Most heat.

then again you can worry more about that, when its your bike generator using your human energy to make the power :) i always hated those things.

Great Pics.
with your voltage range, you might not need the resistance, check it again after the batteries are 90% charged.
you might be closer than you think to a charger voltage vrses battery voltage peaking anyway with a 16v AC transformer, and the multiple voltage drops through the components. and about then you will begin to understand the rest of the crap i wrote and why :)
 
Last edited:

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
with your voltage range, you might not need the resistance, check it again after the batteries are 90% charged.
you might be closer than you think to a charger voltage vrses battery voltage peaking anyway with a 16v AC transformer, and the multiple voltage drops through the components. and about then you will begin to understand the rest of the crap i wrote and why :)

Arh I think i see what your getting at. You got me worried for a while so i did some testing.

The 16v AC transformer has an unloaded voltage of 18V. This is RMS value, so once rectified (with additional parallel capacitor) the voltage is 1.414 times that value, or approximately 25v. Confirmed this with the DMM.

So with a voltage drop of around 3v for the LM317 and 1.2V for the rectifier bridge this still leaves approx 20v for charging, or slightly less when under load.

Just to be sure, I stuck all 12 cells into a commercial charger and got them to 100% capacity then put them back into 12s and tested the current. Reading was 200mA, down 30mA but still acceptable.

I was then worried that the charging voltage might be too high, but came to realize that the benefit of using the LM317 is that the current is constant regardless of the voltage so it should be ok. I hope.. A few test in 12s for 14 hours will be the only way to know for sure.

Thanks again for all your input, it feels like i'm making some progress.

Now my next problem to tackle is low voltage protection so that the LED driver's dont suck the cells down below 1-1.2V each.

Any thoughts?
 
Last edited:

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
Did a full charge last night from 1.05V up to 1.49V per cell in 13 hours.

After the cells had rested for a few hours they are now sitting at 1.43V per cell so i'm happy enough with that.

I didnt remove the cells right away from the charger but let it continue on for a while up to around 15 hours to see what happened. The voltage dropped fell from its peak and stayed steady at about 1.45V per cell with the cells all around 30 degrees C.

So it should be fine to leave on for much longer with no ill effects.

-Capo
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
Now my next problem to tackle is low voltage protection so that the LED driver's dont suck the cells down below 1-1.2V each.

Any thoughts?

There are some low voltage protection circuits out there, but I am not sure if they are alway so useful or not. Some of the voltage and current regulation driver / controllers have a low voltage cut out pin, but this assumes you want to build a real board.

Since they are NiMH cells, and individually replaceable, perhaps it is not worth worrying about over discharge?
 

VidPro

Flashlight Enthusiast
Joined
Apr 7, 2004
Messages
5,441
Location
Lost In Space
georges driver things have in them various ability for low battery warning, and cutoff, some of the most sofisticated and well thought out microcontroller drivers. at taskled.com, but not cheap, and not seperate low cutoff items.
 

KD5XB

Enlightened
Joined
Oct 20, 2008
Messages
432
Location
DM84jk
Just noticed this thread, it's quite interesting.

The 16v AC transformer has an unloaded voltage of 18V. This is RMS value, so once rectified (with additional parallel capacitor) the voltage is 1.414 times that value, or approximately 25v. Confirmed this with the DMM.

Does this mean the peak-to-peak voltage is in the neighborhood of 40 volts? Because I'm suddenly trying to remember how you get more voltage AFTER a full-wave bridge rectifier -- if that is indeed what you're using. It's been too long since I worked with RMS and peak to peak stuff.

As for the original question -- my first thought of charging all these cells in series was that it's going to be difficult to keep them all at the same level. And THEN I remembered how we used to charge ni-cad batteries for the F-4, F-16, and F-111 -- these batteries consist of 22 cells in series in a metal case, and each cell can be individually replaced, and -- they all self-balance when charging. Of course, we used an extra step called equalization in which each cell was shorted (after being discharged, of course) for some 8-12 hours or more. This seemed to help them all take the same amount of charge.

This idea sound FINE. No, GREAT, no -- well, it sounds like a jam-up job, no -- what's the word???

EXCELLENT
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
Well i can report that with a few more charges under its belt there still hasnt been a problem.

I'm suddenly trying to remember how you get more voltage AFTER a full-wave bridge rectifier

All that is required is a electrolytic capacitor in parallel with the DC side of the rectifier. This capacitor helps to smooth out the wave form and gives a higher output voltage.

A few pictures for your enjoyment. The rectifier on the left and cap on the right.
2m7y9lt.jpg


I just bent the DC Pin's back on the rectifier and soldered the cap straight to it.Simple as that.
1zd9181.jpg


An update of the charger. I found the heat sink i was using previously wasn't beefy enough so I bought a strip of aluminium and bent it and used thermal paste to glue it to the inside of the box its all sitting in. This now dissipates the heat well with out the need for venting slits.

The battery packs next to the box fits in neatly over the components. Its two 6cell holders heat shrinked for any one interested.

All put back together the box will also include two led drivers for 4 XR-E's.
i69mbc.jpg


I also added a "charging" led (RED) and a "charged" led (green but not on). The charging led is just connected to the DC side of a second rectifier with a resistor.

2w7pm43.jpg


The "charged" LED is just an indicator of voltage, nothing more.. It works with a zener diode and a transistor. In short the zener is in revers with a break down voltage of 14v (maybe, havent finished testing yet) so that when the pack voltage gets about 3v above that the led comes on letting me know its about time to unplug. The zener and led are in parallel with the batteries just so you know.

The transistor is to turn the "charged" led off when its all unplugged after charging. Its just in series with the led and has its base pin connected to the same separate rectifier used for the "charging" led, it loses power when unplugged.

Hmm this post got longer than I expected.

If its all a bit to confusing and any one is interested i can put up a quick circuit diagram. That is if someone can suggest a quick free program that allows me to draw one. Cheers,

-Capo
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
All that is required is a electrolytic capacitor in parallel with the DC side of the rectifier. This capacitor helps to smooth out the wave form and gives a higher output voltage.

True, but you need a significantly larger, higher capacitance value for it to do much smoothing. What you have now is more like a 120Hz pulsing charger, the capacitor should be at least a few hundred uF. Some people use a guideline of 1000uF per 1A, though it depends a lot on how much ripple is acceptable. Fortunately a NiMH battery pack isn't very picky but IMO, better to up the cap value.
 

Capo_au

Newly Enlightened
Joined
Aug 9, 2009
Messages
83
Location
Melbourne, Australia
Some people use a guideline of 1000uF per 1A, though it depends a lot on how much ripple is acceptable. Fortunately a NiMH battery pack isn't very picky but IMO, better to up the cap value..

Noted. I just took a look at the capacitor type I used and its rated for 10micro farads.

I have a couple of bags of 1000micro F laying around so i have no excuse not to use them next time just to be safe.

Thanks for the heads up :thumbsup:
 
Top