Commuting - How long do your NiMH batteries last?

In late 2001, I purchased a Marwi Kamikaze set of lights (dual beam, 20w flood, 12w spot, 6ampHr NIMH battery). I got a new 8ampHr battery free of charge from marwi not long after since the 6AmpHr battery died and it was replaced under warentee with the 8AmpHr one free). Anyway, it is now Jan 2004 and it appears that the battery is no longer holding a charge. I've kept the battery inside in the summer and fully charged and have used it a few times doing some night biking, but mostly it's been used for fall/winter commutes.

Now I have only commuted in winter, during the last years winter and this years (still in progress) but is that unreasonable for battery life? Should I expect NiMH battery packs to last more than 2.5 years when the use is about 50 minutes per day (in the morning) and mostly between Oct and Mar?

I'm debating on trying to send this unit back for service or perhaps buying a whole new set of lights or buying another 8amp hr battery.

Jay

NiMH batteries do suffer somewhat from memory effect. To get the best out of NiMH batteries you need to completely discharge them before recharging them. Any decent recharger designed for NiMH batteries should have a discharge cycle, but if yours doesn't run your lights until the battery is completely flat and recharge it. Give it a try anyway.

However, I suspect after 2.5 years you may be getting near the end of this battery's life if you've been charging it a few times a week. Check the specs on your battery, but you'll usually get about 500 recharges out of a battery

Before spending lots on a new battery, check out the Schmidt Dynamo front hub at Peter White Cycles:

http://www.peterwhitecycles.com/Schmidt-Lumotec.asp

DO NOT! Completely drain a NiMH or NiCad pack! You greatly run the risk of reversing one of the cells, or more. If I remember correctly you should only take the pack down to the combined voltage of the total number of cells in a pack. Example if your pack has 5 cells at 1.1v each I should not be taken down below 5.5Volts. My other hobby is RC modle airplanes, we use the same kinds of battries and usually lots better charger/cycling equipment than you get with light systems. I use a model hobby charger to cycle my light battries and quick charge them. You just have to know the charge rates for the different battries and the Delta Peak/cell of the different battrie packs.

You can read more than you ever wanted to know about care and feeding of battries here (http://www.rcbatteryclinic.com/)
You can also check sites like Sanyo and Panasonic.

I've never let the battery go so long as it wont turn on, I will usually notice that the light is becoming "yellowish" and that is usually the sign that I need to recharge. I see the same 8amphr battery on sale at Aardvark for $99. But have been looking around for closeouts to see if getting a new one is worth it. I do like the 8ampHr battery, I can usually go 1.5-2 weeks on one full charge.

Jay

I've never let the battery go so long as it wont turn on, I will usually notice that the light is becoming "yellowish" and that is usually the sign that I need to recharge. I see the same 8amphr battery on sale at Aardvark for $99. But have been looking around for closeouts to see if getting a new one is worth it. I do like the 8ampHr battery, I can usually go 1.5-2 weeks on one full charge.

Jay

It does sound like you're getting memory problems. Uciflylow was correct in a way: you don't want to completely drain your battery i.e. down to 0 volts, however, draining it down to under 1V per cell, recharging and repeating 3 or 4 times [i]may[i/] solve your problem (if the battery isn't already shagged). If you don't have any means of measuring voltage, just drain it until the lights don't shine any more, but not too far past that. A way to test that you haven't drained it too far is to turn it off for a minute or two and turn it on again. You should get a few seconds of dim light before it fades again.

It's at least worth trying before spending money on a new battery.

JayH,

That is why I like my homebuilt light. I use a cheap sealed lead acid battery that is very reliable. If my battery goes bad a year from now, another can be had for about the cost of two large pizzas, Coke and a movie rental (not that I'd actually eat two pizzas, it's just a comparison :D )

The quickest way to kill your batteries is to over-discharge them in an attempt to avoid the memory effect. I know this from experience... I wondered why my batteries always died so quickly when I took such "good care" of them. Once I stopped trying to discharge my batteries before charging them, they stopped dying.

Now, I know that fancy chargers can condition your batteries to extend their capacity, but I think this works more for Nicads than for NiMHs -- but this bit is just from my memory...


From http://www.hippy.freeserve.co.uk/nicdmyth.htm :
(this is about NiCd batteries, which are more susceptible to memory than NiMHs.)



The original concept of Memory Effect came from testing NiCd battery packs for use in space vehicles, and satellites in particular. Under perfectly consistent conditions, a NiCd battery was charged up and partially discharged over a number of cycles. Each time the charging and discharging was done to exactly the same points and, in this very particular case, the performance of the battery did suffer what became known as Memory Effect.

The important thing to note is that this Memory Effect was found only when the charging and discharging regime was totally consistent and to very particular points; varying the regime, charging and discharging to other points, does not reproduce Memory Effects.

Only in a regime where you have absolute repeatability and accuracy of control over the recharging regime can the memory Effect even be encountered, and while this is indeed a concern for such situations as were tested for, it can be, and is, overcome by adjusting the regime.

You can't really test a battery using a simply DC voltmeter on the +/- of the output, right? It should be under some kind of load (resistance) to be accurate. Although I guess you can prove a battery is dead, it wont prove if it's still good? Not an electronics major. I've considered making my own lights before but like studded snow tires, I can usually find sales and stuff to make it worth it, for me, simply to buy a set. I'm patient and resourceful enough to scour the net for coupons and bargains.

Well, I'm going to try to recharge it again and check it out this way. I am then going to put that light in addition to my backup the next time I bike commute. Hate to be without a decent light now that it's icy and snowy out.

Jay

The quickest way to kill your batteries is to over-discharge them in an attempt to avoid the memory effect. I know this from experience... I wondered why my batteries always died so quickly when I took such "good care" of them. Once I stopped trying to discharge my batteries before charging them, they stopped dying.

Now, I know that fancy chargers can condition your batteries to extend their capacity, but I think this works more for Nicads than for NiMHs -- but this bit is just from my memory...


From http://www.hippy.freeserve.co.uk/nicdmyth.htm :
(this is about NiCd batteries, which are more susceptible to memory than NiMHs.)



The original concept of Memory Effect came from testing NiCd battery packs for use in space vehicles, and satellites in particular. Under perfectly consistent conditions, a NiCd battery was charged up and partially discharged over a number of cycles. Each time the charging and discharging was done to exactly the same points and, in this very particular case, the performance of the battery did suffer what became known as Memory Effect.

The important thing to note is that this Memory Effect was found only when the charging and discharging regime was totally consistent and to very particular points; varying the regime, charging and discharging to other points, does not reproduce Memory Effects.

Only in a regime where you have absolute repeatability and accuracy of control over the recharging regime can the memory Effect even be encountered, and while this is indeed a concern for such situations as were tested for, it can be, and is, overcome by adjusting the regime.


The memory effect occurs because Cd forms very stable compounds with intermediate oxidation states relative the charged and discharged oxidation states. The batteries' discharge base needn't be totally the same, merely at proximate levels. Also, multiple memory pts can occur if two or three common discharge bases are regularly occurring.

With NiCd batteries the effect is quite pronounced, and for that reason the optimized recomendation is to deep condition NiCd batteries every 10 cycles or once each month. Deep conditioning entails dropping the voltage to about .6V/cell in battery packs. Many chargers will automatically do a light conditioning to .9V/cell on each cycle, the occassional deep cycle conditioning is more important. If you can seperately discharge each cell, then bringing the individual cell to 0V is thebest way to wipe out those intermediate compounds breaking the "memory." In my undergraduate days when my employ included repairing/maintaining defibrulators we pulled the NiCd cells out of their banks and individually resistered them overnight so the cells would be brought to zero. In series-packs try not to go below .6V/cells because the cells are not balanced in charge and capacity, so a margin for error is placed in packs to prevent the more deliterious reverse charging of the weaker cells. Reverse charging rapidly destroys cells with dendrite formation.

Ni also forms mildly stable intermediate compounds, so NiMH and NiZn batteries will have a mild memory effect. It is so mild that its only been widely accepted within the last couple of years that it merrits any conditioning. Current recomendations are conditioning NiMH batteries once every 30 cycles or every 3 months.

From what I've learned, you should get an accurate reading when testing off a DC voltmeter.