Appropriate Charge Rate for NiMH Batteries

[wapkil]

https://www.candlepowerforums.com/threads/187526&highlight=torture
as far as slow charging i did this test, which could be concidered far from slow, for over 300 hours, only the first few hours are in a time lapse video.
the battery still doesnt act squat different from the others

It shows that a slow charge rate can be really dangerous. It first damaged your power supply and then even the mains power. If you have a nuclear power plant nearby, please don't perform any longer tests

 

[Marduke]

They do not stipulate you must use any one of the methods they recommend to achieve the specifications for their cells.

In fact they do, and I specifically linked the reference documents above.

BTW, I don't think it can be said that manufacturers recommend fast charge rates because they are "better". In most of the longer manufacturer's documents that I read this recommendation was prefixed by an explanation that it is because of the contemporary trend to charge batteries faster or because it reduces back to service time. I haven't seen anywhere an explanation that it is actually better for batteries...

They changed their tune over trends, but not quite how you assume. Back in NiCad days, slow chargers worked just fine, as NiCads were very tolerant of it. When NiMH consumer cells were introduced, the chargers did not drastically change. If you know the precise capacity of the cell in question, you can charge at a slow rate on a timer. Since it was easier to use the same chargers and just change the time, it was an easy assumption to simply assume consumers would recharge cells only when depleted. This is also when you started to notice more strong language to the effect of "use only our cells in our charger". Put a lower capacity cell in a charger meant for higher capacity, and you can damage the cell.

Eventually, it became evident that consumers wanted to be able to charge up their cells whenever they wanted, regardless of the state of capacity. Cells of varrying capacity also became prevelant. Charging a cell of unknown capacity at an unknown state of charge is much harder than when you knew exactly how much a cell stored and assumed it was empty.

That meant the charger had to become "smart". You know have to know eactly what the capacity is, and what the state of charge is to do a timed charge and not kill the cell. Obviously that is not easy. Second best method is to give the charger some brains so it can figure out on it's own when a cell is "done" charging. There are MANY different algorithims you can use to apply a charge to a cell and monitor it for completion, many of which are discussed in the above referenced documents. Of these, a method had to be chosen that was reliable, affordable, and easy for the consumer to do. That drove smart chargers to a -dV/dT condition to sense end of charge. But to sense this signal, the charger must be able to measure it. A typical NiMH cell will show this signal more prominantly >0.5C charge. Below that and the signal is VERY difficult to measure, or the cell may not even display one at all. Above 1C charge, you can get into other issues with charging the cell too fast.

0.5-1.0C is a sweet spot where you can both measure a -dV/dT signal, and not overcharge the cell. If your equipment is capable of it, that is the current "best" method to charge NiMH cells of an unknown state of discharge.

 

[wapkil]

They changed their tune over trends, but not quite how you assume.

[...]

0.5-1.0C is a sweet spot where you can both measure a -dV/dT signal, and not overcharge the cell. If your equipment is capable of it, that is the current "best" method to charge NiMH cells of an unknown state of discharge.

And of unknown capacity. I generally agree with you, although I'm afraid that all these "smart" methods have their problems. For example a supposedly advanced Maha C9000 charger uses the simple maxV termination at a too low voltage of 1.47V probably because they couldn't implement a smart algorithm that would work reliably with all "crap" cells that users may put inside.

I don't think anything you wrote contradicts my "assumptions" (actually more like impressions, I know too little to assume anything). I believe you explained quite well why smart chargers became popular. I wrote that my impression is that since they became popular, the manufacturers started to recommend the best methods for them. It means that these algorithms are probably best for smart chargers but it doesn't mean that they are better than a correctly performed standard charge.

I haven't read now the documents from the links that you provided but I saw all of them earlier. I may have missed something but I believe that in none of them it was written that fast charging is better for batteries. Some of the documents recommend fast charging but I'm not sure if this recommendation isn't simply based on user convenience. Especially when confronted with documents that actually advertise fast charging as more convenient for users but, even if they describe the standard charging procedure, don't offer any suggestions that fast charging is better for batteries.

 

[VidPro]

if a "fast" charger is looking for a V-drop, there better BE one.
if a "smart" charger is looking for a V-drop, there better BE one.
if a "digital" charger is looking for a V-drop, there better BE one.

anybody find me a "good" slow charger this week? i bought something that was timed and had slow specs (lies on a website), and when i tested it it was at 350ma. The front of the box should have said Betty crockers battery baker.

look at the rates on the dumb chargers manufactures are putting out in packaged kits, Not slow, but great for selling more batteries i guess.

all these digital pulse chargers AVERAGING a slow rate, not slow, hard pulses.

how do you Even DO a slow charge, if it wasnt for OLD backwards chargers from the 80s, which you can still find in the antiques section of E-bay.
some of my actually slow chargers timeout before 2000ma even

how do you TEST slow rates, on chargers that dont have them?
it is no wonder, with Lacross 900s and stuff like them , that the manufactures Dont recommend slow. it Cant be Done.
i know of like 2-3 units out there still sold that do a real in rate slow charge.

 

[travelinman]

Actually Randy (my rcmp tech buddy) deal with a lot of individual cells, such as flashlight cells etc. You're right, not all are AA, some are C as well.

As a matter of fact, he is researching going to Lithium Ion. He has a couple of his techs looking into high capacity Lithium for some of his uses. Don't know what kind, maybe only available to industry or something like that. I'll let you know more when I do.

 

[wapkil]

if a "fast" charger is looking for a V-drop, there better BE one.
if a "smart" charger is looking for a V-drop, there better BE one.
if a "digital" charger is looking for a V-drop, there better BE one.

anybody find me a "good" slow charger this week? i bought something that was timed and had slow specs (lies on a website), and when i tested it it was at 350ma. The front of the box should have said Betty crockers battery baker.

[...]

how do you TEST slow rates, on chargers that dont have them?
it is no wonder, with Lacross 900s and stuff like them , that the manufactures Dont recommend slow. it Cant be Done.
i know of like 2-3 units out there still sold that do a real in rate slow charge.

Once again I completely agree. For an average user and usage fast charge rates are more convenient, easy to get in current chargers, maybe even less prone to errors. I think no one is denying it.

As I understood though the discussion is about the best method in ideal conditions for someone who doesn't care about charging convenience. If I took my hobby charger (which I don't have) and took time to select all the parameters needed for the algorithm, would it be better for the batteries future performance (assuming moderate discharge current afterwards) to use the standard charge or some fast charge method. It still seems to me that the standard charge would be better in this case.

 

[clintb]

VidPro,

If you're not afraid to spend a little bit, here's a truly wonderful constant current charger with two channels. I've had one for years, and it's served me well.

http://www.acehobby.com/ace/ACE2528.htm

 

[VidPro]

VidPro,

If you're not afraid to spend a little bit, here's a truly wonderful constant current charger with two channels. I've had one for years, and it's served me well.

http://www.acehobby.com/ace/ACE2528.htm

now thats old scool , even discontinued


When using the DDVC charger, you must keep in mind that it is not a peakdetect or automatic cut-off charger. You are the only one who can stop it. Don't forget about it, or you could damage your cells.

so it had 2 otions i guess , the old 16 hour slow charge, or fast charging while holding battery in hand, terminate on "ok its warm enough".

 

[VidPro]

, would it be better for the batteries future performance (assuming moderate discharge current afterwards) to use the standard charge or some fast charge method. It still seems to me that the standard charge would be better in this case.

many good hobby chargers have nice slowdown alogrythms , like the triton silverfox has talked about.
and most of them do really nice stuff of the proper CC/CV for li-ion.

and then a person can often adjust that V-drop detection, to very low, so the machine terminates sooner on a overcharging.
and often you can put a temp probe ON the battery, and set very tight specs for that to bail out, which can be relative to the room temps your in and charge rate, so it would actually work, because of a manuel perfecting of it..
but
they cant single channel 4 or 8 ni-?? cells, like a consumer would want.

 

[Marduke]

And of unknown capacity. I generally agree with you, although I'm afraid that all these "smart" methods have their problems. For example a supposedly advanced Maha C9000 charger uses the simple maxV termination at a too low voltage of 1.47V probably because they couldn't implement a smart algorithm that would work reliably with all "crap" cells that users may put inside.

Well, there is more to that. The C9000 also has -dV/dT IIRC, but LSD cells and Eneloops in particular (the majority of CPF's use on the C9000), the cells terminate at a higher than typical voltage vs what the c9000 was designed for. "Regular" NiMH I believe have a good change of actually triggering the -dV/dT condition.

I haven't read now the documents from the links that you provided but I saw all of them earlier. I may have missed something but I believe that in none of them it was written that fast charging is better for batteries. Some of the documents recommend fast charging but I'm not sure if this recommendation isn't simply based on user convenience. Especially when confronted with documents that actually advertise fast charging as more convenient for users but, even if they describe the standard charging procedure, don't offer any suggestions that fast charging is better for batteries.

Duracell states for "optimum performance", use this method:

• For fast charging and optimum performance, Duracell recommends a three-step procedure:
  1. Charge at 1C rate, terminated by using dT/dt = 1°C (1.8°F)/minute
  2. Apply a C/10 top-up charge, terminated by a timer after 1/2 hour charge (optional, not required)
  3. Apply a maintenance charge of indefinite duration at C/300 rate
• The recommended charge procedure should be used with a backup temperature cutoff of 60°C (140°F).

Duracell SPECIFICALLY recommends 6.3.1

Interestingly, Energizer states that "the charge acceptance efficiency for the NiMH cell is improved as the charging rate is increased." which tells us it really is more effecient to charge at 1C than at slower rates.

 

[Bones]

if a "fast" charger is looking for a V-drop, there better BE one.
if a "smart" charger is looking for a V-drop, there better BE one.
if a "digital" charger is looking for a V-drop, there better BE one.

anybody find me a "good" slow charger this week? i bought something that was timed and had slow specs (lies on a website), and when i tested it it was at 350ma. The front of the box should have said Betty crockers battery baker.

look at the rates on the dumb chargers manufactures are putting out in packaged kits, Not slow, but great for selling more batteries i guess.

all these digital pulse chargers AVERAGING a slow rate, not slow, hard pulses.

how do you Even DO a slow charge, if it wasnt for OLD backwards chargers from the 80s, which you can still find in the antiques section of E-bay.
some of my actually slow chargers timeout before 2000ma even

how do you TEST slow rates, on chargers that dont have them?
it is no wonder, with Lacross 900s and stuff like them , that the manufactures Dont recommend slow. it Cant be Done.
i know of like 2-3 units out there still sold that do a real in rate slow charge.

I'm pretty sure the old Energizer CHM39 I use for cells that none of my other chargers will recognize is a true slow charger:

http://data.energizer.com/PDFs/chm39.pdf

Its output: 130mA x 2(AA) / 80mA x 2(AAA) / 23mA x 1(9V)

Granted, its a two channel, four place, timer controlled 14 hour charger; but given enough time, it eventually even things out at a slow enough rate to minimize concern.

It's also surprisingly well built and finished and actually may keep going and going...

 

[VidPro]

It shows that a slow charge rate can be really dangerous. It first damaged your power supply and then even the mains power. If you have a nuclear power plant nearby, please don't perform any longer tests

not to worry, it's just a nuclear and Bio warfare lab nearby but they burn out more stuff than i do. A little tritium in the groundwater just makes it easier to get a glass of water at night and 6fingers, the neighbor kid is going to be a great pianist someday if they can just do something about the wart problem, and that tail thing he has.

 

[wapkil]

Well, there is more to that. The C9000 also has -dV/dT IIRC, but LSD cells and Eneloops in particular (the majority of CPF's use on the C9000), the cells terminate at a higher than typical voltage vs what the c9000 was designed for. "Regular" NiMH I believe have a good change of actually triggering the -dV/dT condition.

I know, there is also a timeout and I think another condition. Nevertheless all the NiMH batteries I have (LSD and non-LSD) terminate on the maxV. I was really surprised when I found out this behavior. I believe that when the cells age other conditions may appear earlier, but it doesn't seem to happen for good cells.

I mentioned this to show the problems with "smart" algorithms. IIRC Maha lowered the maxV level after problems with missed terminations in earlier C9000 versions. It shows that at least for this particular manufacturer smart algorithms turned out to be not smart enough to rely on them.

Duracell states for "optimum performance", use this method:

• For fast charging and optimum performance, Duracell recommends a three-step procedure:
  1. Charge at 1C rate, terminated by using dT/dt = 1°C (1.8°F)/minute
  2. Apply a C/10 top-up charge, terminated by a timer after 1/2 hour charge (optional, not required)
  3. Apply a maintenance charge of indefinite duration at C/300 rate
• The recommended charge procedure should be used with a backup temperature cutoff of 60°C (140°F).

Duracell SPECIFICALLY recommends 6.3.1

Interestingly, Energizer states that "the charge acceptance efficiency for the NiMH cell is improved as the charging rate is increased." which tells us it really is more effecient to charge at 1C than at slower rates.

Duracell writes "For fast charging and optimum performance" you can interpret it any way you want. They also recommend an indefinite trickle charge. I think it means that the recommended algorithm is not dedicated to ensure the best cells longevity (but will ensure that the batteries will be always fully charged, hence the optimum performance).

I don't think we can really find the answer to the question about the best charging algorithm directly in manufacturers recommendations. They couldn't for example write "We recommend the standard charge for our batteries. It will take only 16h. Before charging make sure to test the battery actual capacity to select the appropriate current. The capacity will change so you have to test it periodically. You need a good charger (~$100 for a single channel version). If you want to charge more than a single battery at once, you'd need a separate single channel charger for each of them"

The better charge acceptance efficiency at higher current is interesting but I don't know if it has any influence on batteries longevity or other performance factors.

 

[Marduke]

Duracell writes "For fast charging and optimum performance" you can interpret it any way you want. They also recommend an indefinite trickle charge. I think it means that the recommended algorithm is not dedicated to ensure the best cells longevity (but will ensure that the batteries will be always fully charged, hence the optimum performance).

I don't think we can really find the answer to the question about the best charging algorithm directly in manufacturers recommendations. They couldn't for example write "We recommend the standard charge for our batteries. It will take only 16h. Before charging make sure to test the battery actual capacity to select the appropriate current. The capacity will change so you have to test it periodically. You need a good charger (~$100 for a single channel version). If you want to charge more than a single battery at once, you'd need a separate single channel charger for each of them"

Unless you are seeing something I'm not, they ONLY recommend the 1C charge. They only present other options, but they are not recommended. Certain charge rate rages are specifically NOT recommended actually.

 

[VidPro]

sanyo has it figured, Just use our charger OK :nana: i dont even HAVE one of thiers, and i got like 8 now.

Though it is possible to charge an eneloop battery in a "Quick Charger", it is not recommended. We recommend charging eneloop batteries in a NiMh charger that is 2 hours or more. Charging eneloop batteries in a "Quick Charger" can reduce the overall life of the battery. It is strongly recommended to use eneloop, GE/Sanyo or Sanyo NiMh battery chargers. We only warrant eneloop if used with an eneloop, GE/Sanyo or Sanyo NiMh battery charger.

http://us.sanyo.com/Batteries/FAQs

their charge things they say in the faqs are 2-4hours and the other charger 4-7 hours

 

[Mr Happy]

It is true that modern NiMH cells are designed to withstand a 0.1C charge without damage when already fully charged. Therefore a 0.1C standard charge is in theory fine to use. Where this comes unstuck is that the charger cannot know what 0.1C actually is -- it relies on a human to enter that data into the charger. And most ordinary consumers and users of NiMH batteries do not know what the capacity is and do not want to enter it into the charger like they might have to do in the "complicated" MH-C9000.

So to build a slow timed charger, what charge rate should it use? Maybe 270 mA for 2700 mAh cells? Then it will overcharge and damage eneloops. OK, so maybe 200 mA for eneloops. But that will overcharge and damage 1600 mAh cells. Ah, then 160 mA. But that will overcharge and damage the new Sanyo Harmolattice cells (1000 mAh).

Therefore, to play it safe the universal timed charger should use a charging current of 100 mA, and should have a timer for 160% of 2700 mAh giving 43 hours. That's the design of the perfect consumer charger! Consumers will flock to buy it! Everyone will love to wait nearly two days for their batteries to charge

 

[travelinman]

Every time I hear the phrase "smart charger", I'm reminded of some movie or other that has the line....."strong like bull, smart like tractor" in it.

 

[VidPro]

i want mine to charge at a pulse 2amps 50-50 then drop to 1amp and .5amp and .1amp till full, and finish with very short pulses of .05 for final topping for an hour, all based on temperature or voltage.

ray-o-vac didnt care what you do they will work with anything , any charger you got, they dont care.

Here is energysers OLD speel http://data.energizer.com/PDFs/nickelmetalhydride_appman.pdf

Determining when overcharge has occurred is critical to charging schemes that
minimize the amount of time spent at high charge rates in overcharge. In turn, these
efficient charging techniques are a key to maximizing cell life, as will be discussed later.
Primary charge control schemes typically depend on sensing either the dramatic rise in cell temperature
or the peak in voltage​

Charge control based on temperature sensing is the most reliable approach to
determining appropriate amounts of charge for the nickel-metal hydride cell.
Temperature-based techniques are thus recommended over voltage-sensing control
techniques for the primary charge control mechanism.
Recommended Charging Rates
Today's trend to faster charge times requires higher charge rates than the 0.1 to 0.3C
rates often recommended for many nickel-cadmium charging systems. Both Figures 18
and 19 indicate that fast-charge rates serve to accentuate the slope changes used to
trigger both the temperature and voltage-related charge terminations. A charge rate of
1C is recommended for restoring a discharge cell to full capacity. For charging schemes
that then rely on a timed "topping' charge to ensure complete charge, a rate of 0.1C
appears to balance adequate charge input with minimum adverse effects in overcharge.
Finally a maintenance (or trickle) charge rate of 0.025C (C/40) is adequate to counter
self-discharge and maintain cell capacity.​

. . . there is lots more right under that too.

I wonder when they are going to tell the guy that wrote it , that none of the energyser chargers (cept the 15min) even followed any of it ​

 

[travelinman]

That's what happens when the head office is in Florida, the engineering is done in Romania, marketing is in London and manufacturing is in Taiwan!

 

[VidPro]

then 160 mA. But that will overcharge and damage the new Sanyo Harmolattice cells (1000 mAh).

Therefore, to play it safe the universal timed charger should use a charging current of 100 mA, and should have a timer for 160% of 2700 mAh giving 43 hours. That's the design of the perfect consumer charger! Consumers will flock to buy it! Everyone will love to wait nearly two days for their batteries to charge

43 hours , that sounds like fun. the harmolattice and other very robust lower capacity cells like even the eneloop cope with the "overcharge" rates (relative to capacity) better than the high capacity.

but that all makes BONES slow charger just about right.

one thing i dont understand is why people call the 12-14 hour chargers "Travel" chargers, when i got mabey 6 hours in a motel before the maid is turning the bed out on me. Checkout time :wave:, and at home i got all day.