Appropriate Charge Rate for NiMH Batteries

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
As far as I know such a level simply doesn't exist. You either have a higher risk of overcharging or higher risk of premature termination. With the usual level there is probably non-negligible risk of both. Except the false positives or negatives the sensitivity doesn't change the moment of termination in any important way. The -dV overcharges the cells - if you apply the trickle charge afterwards they will be even more overcharged.

... and I'm suggesting it doesn't necessarily overcharge, because they use a conservative termination with the idea that yes, it's a little undercharged, will finish charging with the trickle.

dT/dt is not based on the temperature value but on the detection of the temperature rise rate. Please read the descriptions of appropriate algorithms if you are interested in comparing them.

... but this is even worse, in aforementioned example the 90% charged battery temp rise is much faster than the other. I am still of the opinion that dT is not so good unless only comparing two cells going through the majority of a charge cycle, and with similar impedance else you again have different temp rise.

You may be right but the documents I read claim that it would be more costly to implement it correctly than to have a simple -dV. I tend to believe they are right here.

I can believe it is more costly to implement "correctly" because it would need a microprocessor to factor for several things in addition to just temperature change, like the amount of time and rate vs cell size. This is one of the reasons I feel Delta -V is at present still the reasonable method to use, but the higher the charge rate, the more important the charger have a safety temperature cutoff circuit, emergency type not just typical end-of-charge detection type. Often a circuit like this is quite simple, simply a resettable thermal breaker in series with the charging contact and of course as close to the battery tube as reasonably possible.

I understand your position and sometimes it is more appropriate also for me. Nevertheless I am interested in finding out what are the results of using various available charging methods. Especially since there seems to be a popular impression that there exist a single best method that can be recommended to everyone and I think it is simply not true.

I think the best method is the most convenient one, that will vary per the situation.
 

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
... but this is even worse, in aforementioned example the 90% charged battery temp rise is much faster than the other. I am still of the opinion that dT is not so good unless only comparing two cells going through the majority of a charge cycle, and with similar impedance else you again have different temp rise.
Actually, not so. The rate of temperature increase provides a very certain and predictable indicator of when the cell is fully charged. When the cell has capacity remaining most of the supplied energy is absorbed by the cell and stored, so the temperature rise is moderate. But as soon as the cell is fully charged and can store no more energy the whole of the applied power is converted to heat inside the cell leading to a rapid temperature rise. This temperature rise provides a certain signal that charging is complete and will be just as evident whether the cell starts out 0% full or 90% full.

You can see how it works in this test with an eneloop:

eneloop2rf2.png


You can see the really sharp uptick in temperature at the end. This is what chargers look for when they are doing dT/dt control. If I had not stopped charging when I did the temperature would rapidly have gone off the scale.

The eneloop is a good quality cell. Here is a similar test with a "crap" cell:

voltagecurvemz8.png


Due to the poor cell performance the rate of temperature increase is higher during the normal charging phase here, but there is still an uptick when the cell is fully charged. As with the eneloop or any other cell, if I had not stopped charging at this point the temperature would have risen steeply off the scale and a charger would be able to detect this.

(Here is the original thread containing those pictures for reference.)
 

wapkil

Enlightened
Joined
Jan 17, 2009
Messages
739
Thanks for the graphs Mr Happy :)

I'll try to write a longer paragraph describing how I understand some things related to charging. I'll need to find some time for it so probably later this week.

I think your graph with a "crap" cell demonstrates why I don't believe -dV or maxV can be universally relied upon. Here is mine:



The cell was charged with ~0.7C in Maha C9000 charge mode. At ~130min. it was already fully charged. The voltage at this point was ~1.44V. Apparently Maha measures when its pulsing is off, so it takes the lower "boundary" of the green voltage values. 1.44V is lower than Maha's 1.47V maxV so it happily continued charging. The problem was that -dV hasn't appeared (or was small enough to stay undetected).

When the voltage finally reached the 1.47V maxV, the cell was seriously overcharged. I'd have to check but I think by ~15%-20%. You can see how the temperature skyrocketed. Maha stopped the fast charge but continued topping off an seriously overcharged cell. I let it continue for 5 minutes and stopped the torture.
 
Last edited:

wapkil

Enlightened
Joined
Jan 17, 2009
Messages
739
Interesting plot and result, wapkil. I'm curious, was the cell an AAA?

No - an old AA. It already lost some of its capacity but I think it could still live and work many more cycles. Unless of course it is charged in the way depicted above that simply kills it.
 

45/70

Flashlight Enthusiast
Joined
Oct 9, 2005
Messages
2,800
Location
Rural Ohio
OK, thanks wapkil. I asked because I've noted that the C-9000 seems to work somewhat differently with AAA's. I'm not so sure it's as good a solution for them, as it is for AA's.

Dave
 

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
No - an old AA. It already lost some of its capacity but I think it could still live and work many more cycles. Unless of course it is charged in the way depicted above that simply kills it.
Your cell reached 49°C. Mine went over 51°C. I think this is simply what happens when -dV/dt charge termination is applied. It could be argued that it's not a good system in terms of stress on the cells. Probably a lower charge rate and 0 dV/dt is kinder to the cells, but is also more expensive to implement.

To do a really good job of charging NiMH cells I think you need microprocessor control and some very clever programming.
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
I recommend you look up what "Standard Charge" actually means. It does NOT mean that is the recommend charge scheme. That is the method by which the capacity is determined, NOT the typical recommended method.

If you look on the same data sheet, it recommend that the typical charge is 1C...

No, it does not recommend, "recommend" is the word you apply rather than recognizing all they have done is suggested the maximum safe fast charge current.

That doesn't make it "standard" nor recommended moreso than any other, they just needed to list a maximum range they don't want people to exceed.

Again, this is something many component datasheets do, it does not in any way mean there is a problem with something other than the absolute max.

What you are suggesting is like seeing that an automobile tire is rated for 120MPH and thinking that everyone should only drive at 120MPH or else something bad will happen.
 

TorchBoy

Flashlight Enthusiast
Joined
Jan 15, 2007
Messages
4,486
Location
New Zealand
"Recommend" is not just a word that has been applied by the users, even if it doesn't actually appear on the datasheet. The Twicell manual uses the word.
When charging the Twicell, we recommend that you ensure precise charging control, and use constant current charging (quick charging) with a constant current of 0.5 It to 1 It,*** taking into consideration the various characteristics of the battery. Depending on the model of battery, charging at a constant current equivalent to 0.1 It is also possible.
...
... Set the charging current at a value between 0.5It and 1It,*** during quick charging.
If charging is performed at a current less than 0.5 It, the voltage change and temperature rise that occurs when the battery becomes fully charged will not be sufficiently pronounced, making charging control difficult.
...
***···Maximum charge current for HR-4/3AU and HR-4/3FAU is 3.0A
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
"Recommend" is not just a word that has been applied by the users, even if it doesn't actually appear on the datasheet. The Twicell manual uses the word.

Ok, to that extent I am wrong if you really dig up some generic old document you do find the word recommend. However,

1) That manual is ancient in terms of NiMH and charger tech. Presumably back then, even they didn't have chargers that could cope which fell into consumer price-points but now they clearly do sell chargers at different rates.

2) A paragraph, maybe two or three before and after it mentions that cells can be charged at different rates and to contact a rep for details (meaning it can be done or else they'd just write "no").

3) This isn't even for the cells we're talking about it's for the old 4/3 A size used in multi-cell packs as the *** indicates.

Do you realize the significance of that? Back then, the volumetric (energy) density of cells was lower, it means that the temperature rise wasn't as high as it is with today's more energy dense cells. For example, if today's battery were exactly twice as energy dense w/same voltage, then it halves the "recommended" charge rate to 0.25 to 0.5C instead of their old recommendation.

4) They also use the word recommend liberally, stating things like that they don't recommend discharge above 1C but people do that too.

It's pretty clear cut, Sanyo sells several chargers WITH the batteries we've been talking about that do not charge at 0.5 to 1C. To think that is not a recommendation to use the charger with those batteries is a bit beyond bizarre. They even go so far as to state they don't warrant them if you don't use their charger on pages specific to the Eneloops *some* of us have been talking about.

In another topic we even have people advising about which of these chargers to buy... not to avoid all of them.
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
Actually, not so. The rate of temperature increase provides a very certain and predictable indicator of when the cell is fully charged. When the cell has capacity remaining most of the supplied energy is absorbed by the cell and stored, so the temperature rise is moderate. But as soon as the cell is fully charged and can store no more energy the whole of the applied power is converted to heat inside the cell leading to a rapid temperature rise. This temperature rise provides a certain signal that charging is complete and will be just as evident whether the cell starts out 0% full or 90% full.

You can see how it works in this test with an eneloop:

eneloop2rf2.png


You can see the really sharp uptick in temperature at the end. This is what chargers look for when they are doing dT/dt control. If I had not stopped charging when I did the temperature would rapidly have gone off the scale.

The chart illustrates my point. If only considering the graph of fully recharging a cell, not comparing against a cell (same cell...) that has a large % of charge already before being placed in a charger, it is not contrasting the other scenario.

When a cell starts with a large % of charge already, upon starting it is at room temp still, while the graph clearly shows the other cell has elevated temp when reaching the same amount of charge.

When you have two cells at the same charge level, supposing they have roughly the same waste energy being produced by charging, but one at a *pre-heated* state already because it's been sitting in the charger recharging for time as the graph shows, the pre-heated one has a higher difference between it and ambient temp compared to the cell which already had most of a charge before being placed in a charger.

The already charging cell's rate of temperature rise is then lower than that of the partially charged cell that just started to recharge, yet it's absolute temperature will remain higher until some point, possibly after the safe cutoff level from a rapid charge rate to avoid overcharging. You cannot then use rate of temperature change to terminate charge on both cells and end up with them both at the same charge (or overcharge) level, unless some elaborate calculations and data are added and used by the charger, probably requiring manual user input as well if we want to drift down the river of precision at any cost or effort. On the other hand, I am not certain using Delta -V would be accurate either but it is certainly simpler to implement *properly*.

If the already partially charged cell had a smaller difference in charge between it and the drained cell when both started to charge, then their final temps and rate of change would be approaching each other but this is contrary to the typical desires of users to be able to top off a cell at any point.
 
Last edited:

Marduke

Flashaholic
Joined
Jun 19, 2007
Messages
10,110
Location
Huntsville, AL
I suggest you re-read both the Duracell, Rayovac, and Energizer docs above. Both RECOMMEND 1C, and do NOT state it as a maximum.

Also, the temperature termination being discussed is dT/dt, NOT dT/To as you are describing.
 
Last edited:

travelinman

Enlightened
Joined
Feb 15, 2009
Messages
298
Location
Western Canada
The more I'm reading here, the more I'm tempted to go out and get a handful of LM317 variable voltage regulators, set one up with a scrap 12v wall wart, at a fixed 1.47v setting and feed it into a gutted AA charger body from the thrift store. (Maybe do all 4 bays in separate circuits)

I'll then put a 3v, 200 ma incan bulb in series with the circuit to limit current to 200 ma and plug the whole thing into a standard wall timer set for 16 hours. Whole thing should be available for under $5. Simple, failsafe, cheap.

That'll charge my AA eneloops safely and just the way I want it to.
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
I suggest you re-read both the Duracell, Rayovac, and Energizer docs above. Both RECOMMEND 1C, and do NOT state it as a maximum.

Also, the temperature termination being discussed is dT/dt, NOT dT/To as you are describing.

You don't accept that the slope is steeper if the cell starts out significantly cooler at the same charge level?

You still haven't addressed that in fact, recomend does not mean "because it's not possible for another rate to work fine", when plenty of chargers, their designers, and battery manufacturers themselves seem confident enough in what they sell.

Picket these manufacturers to pull their chargers, THAT would at least be aligned with your supposed goal of keeping people doing what you assume to be the only correct way to charge NiMH.

<yawn>, I'm fine with us disagreeing since I get fine performance for my needs with chargers that don't do 0.5-1C. That pretty much makes more of this topic a waste of my time no matter how long someone else disagrees about it. The proof is in doing. Since I have chargers that DO it ok, text on the internet against it is a bit of fiction to me.

Maybe I need to rebadge these magical chargers and sell 'em for $100 a pop as they must perform miracles, but till then I think this is pretty much...:dedhorse:
 
Last edited:

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
You don't accept that the slope is steeper if the cell starts out significantly cooler at the same charge level?
It may be that it is, but even so it is a benefit rather than a problem. What the charger needs is a positive end of charge signal -- and if the slope is steeper the signal is more pronounced, not less. Since chargers traditionally have a problem detecting end of charge when faced with partially charged cells this more pronounced signal is a good thing.
 

Marduke

Flashaholic
Joined
Jun 19, 2007
Messages
10,110
Location
Huntsville, AL
Again, YOU missed the entire point of his thread. It is not about what "will work" but what is BEST. As you yourself admitted several times, 1C is "best".

No one here (other than you) claimed we were saying there is only ONE way to do it, we were all discussing the BEST, and RECOMMENDED method.

Do whatever you want to your cells, but when someone asks for the "best" method of doing something you don't tell them to use inferior methods that offer ZERO benefits.
 

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
Again, YOU missed the entire point of his thread. It is not about what "will work" but what is BEST. As you yourself admitted several times, 1C is "best".

No one here (other than you) claimed we were saying there is only ONE way to do it, we were all discussing the BEST, and RECOMMENDED method.

BUT we totally disagree on why. I think it's best because it takes less time and is more energy efficient, and chargers aren't *necessarily* excessively expensive yet at that rate. Then again, I'd just as soon charge at 2C even if it got me only 80% capacity and 50% lifespan, IF chargers were reasonably priced. I'm never in such a hurry that I need a 15 minute charger though and wouldn't want to limit whether I can charge some of the older cells lying around at 4C.

Do whatever you want to your cells, but when someone asks for the "best" method of doing something you don't tell them to use inferior methods that offer ZERO benefits.
People often ask a generic question with the word "best" in it, and then people answer what is best to them, because if there was only one "best" in all ways, there would only be one charger.

That wasn't what was asked though, what was asked is what is an "appropriate" rate, to which the answer is there is a range that will work, it need not be 0.5C to 1C and if someone is not in a rush to charge or overly concerned about battery life, they are likely to put a little less wear on their battery at a little lower rate, or likewise to put a little less wear on their battery if they pick any rate fast enough their particular charger is able to sense a sufficiently low termination threshold if they were going to use an intermediate rate just a bit above a trickle but weren't likey to remember to manually terminate it.

The short answer is easy, but seldom enough or we wouldn't keep seeing discussions about charging.

... I haven't even mentioned yet that higher current through a charger will tend to wear it out faster, if not melt it, all else being equal.
 
Last edited:

J_C

Enlightened
Joined
Oct 14, 2007
Messages
309
It may be that it is, but even so it is a benefit rather than a problem. What the charger needs is a positive end of charge signal -- and if the slope is steeper the signal is more pronounced, not less. Since chargers traditionally have a problem detecting end of charge when faced with partially charged cells this more pronounced signal is a good thing.

If the charger is designed properly to terminate a cell charged from a fully depleted state (as it should be), a steeper slope could cause premature termination, depending on how the logic onboard is implemented. It typically seems an either/or/both, never a weighted combination of the two so temperature rate change, nor absolute temp alone isn't necessarily, unlike what a full charge graph suggests, a universal solution.
 

VidPro

Flashlight Enthusiast
Joined
Apr 7, 2004
Messages
5,441
Location
Lost In Space
The reason the .5C-1C rate is chosen has little to do with the speed of charge.
It comes from INSURING charge termination, on a device that seeks a V-drop termination method. on a "smart" computer, that is so stupid it will fail to terminate properly everytime with every cell at a lower rate.

Its all about the charger itself, and it's termination method. Just because the 9000 has ADDED in a (sort of) secondary method (voltage peak), it is still a V-drop termination charger.

The ONLY other possible "wise Recommendation" is to be slow enough that it is below the specs listed for an overcharge rate, and only if the charger can do that slow rate.

WHEN your trying to get a stupid computer to try and terminate AND V-drop is it's method for doing so. a SLOW ramping down of the voltage drop on a battery that is being overcharged , over spec, can sometimes be missed because microcontrollers are not humans.
Computers have specific timings and specific sensitivites to the voltage. they dont Look back in hindsight and stop on thier own, till any timer function is reached.
It is a simple Program code set that loops moronically sencing things, it doesnt have 50KB of ram stored that it back compares sencor data to.
Although it does hold a few hundred bytes of data in storage, for the user to glean over, its not that much different than other V-drop seeking chargers.
Round and round they go when they stop the v-drop knows.

The advice for recommended was Solid, there are other choices, but there is a heck of a good reason that the rates selected were recommended, in the first of the thread. Specifically applied to THAT machine the 9000. Other rates that the charger has would usually work most of the time, but could still rarely cause a cell to continue to Overcharge, beyond overcharge specs, damaging it.
 
Last edited:

TakeTheActive

Enlightened
Joined
Dec 16, 2008
Messages
830
Location
Central NJ, USA
C9000 Impedance Check Voltage on *CRAP* Cells...

...Here is a similar test with a "crap" cell:

...Due to the poor cell performance the rate of temperature increase is higher during the normal charging phase here, but there is still an uptick when the cell is fully charged...

No - an old AA. It already lost some of its capacity but I think it could still live and work many more cycles...

What is the C9000 Impedance Check Voltage on these *crap* cells?

...
  • DISCHARGE them @ 100mA on the C9000.
  • Post the voltage from an IMPEDANCE CHECK on the C9000:
    Individually, in Slot #4 (with all other slots empty):

Thanks! :)
 
Top