Charging efficiency of Ni-Mh at 0.5C?
- Thread starter malow
- Start date
Mr Happy
Flashlight Enthusiast
There are two measures of charging efficiency, energy efficiency and coulombic efficiency.
The first of these is given by,
Code:
energy out
efficiency = --------------
energy in
Code:
charge out
efficiency = -------------
charge inBoth of these will vary with both the rate of charging and discharging, and also with the extent of charging and discharging. In other words, if you only charge to 80% before discharging you will get a higher efficiency than if you attempt to charge to 100%.
Mostly I tend to think less energy is wasted when charging and discharging at lower rates, therefore higher efficiencies would be obtainable at lower rates than at higher rates.
But this whole question is a "how long is a piece of string" question, so you can basically obtain any answer you wish depending on how you pose the question. In other words, don't get too concerned about it.
TakeTheActive
Enlightened
I don't know the answer to your questions, but, I did find the LENGTHY articles (LINKed to in my Sig Line LINK) by jtr1962 (2004) and carbonlife (2007) referencing Rayovac I-C³ 15-Minute Battery Chargers and Charge Rate quite interesting.
An excerpt from jtr1962's post:
Emphasis / "Excessive Formatting" mine. ;)
For example, based on jtr1962's post, for the RARE occasions when I'm in a hurry, I have no qualms with pumping 2000mA (via my C9000) or 1800mA (via my BC-900) into an 800mAh, or less, AAA while monitoring the cell temperature (with my finger, or my RadioShack Indoor/Outdoor Thermometer).
The most critical 'Ingredient' to me, regarding Charge Rate, is the ability of the charger to terminate properly.
I have 4 seemingly healthy "Low ICV / Low Internal Resistance" Sanyo 1700mAh non-LSD NiMH cells (Bought: 12/17/02) that overheated in my BC-900 during a 700/350 REFRESH. Was it the cells or the charger? :confused:
Code:
07/24/10 C9000 Impedance Check: | 1.71 1.68 1.72 1.69 VDC
07/24/10 C9000 Break-In: 1500 | 1305 1325 1253 1242 mAh#
07/22/10 C9000 Discharge: 300 | 1249 1259 1216 1132 mAh
07/21/10 BC900 Refresh: 700/ 350| 1443 1486 1368 1372 mAh [CANCELLED -> OVERHEATED!]
07/21/10 BC900 Discharge: 100 | 23 21 23 26 mAh
07/21/10 C9000 Impedance Check: | 1.60 1.59 1.67 1.59 VDC
07/21/10 C9000 Discharge: 300 | 1305 1345 1238 1228 mAh
07/20/10 C9000 Break-In: 1500 | 1360 1405 1275 1259 mAh#
07/18/10 C9000 Impedance Check: | 1.72 1.73 1.78 1.69 VDC(IIRC, you have 3 Sony Chargers attached to a board such that you can prop them up vertically.)
45/70
Flashlight Enthusiast
Are you sure about this Mr H? I think this is probably true as applies to discharging, but I would think the reverse would be true during charging.
In a perfect world where no resistance is involved I could see it working the same both ways. In fact however, the battery or cell offers resistance to the circuit. At lower charging rates this resistance consumes a larger percentage of the charging current than at higher rates of charge (Edit: or, maybe it doesn't). This would seem to make higher charge rates more efficient, no? Maybe not!
It seems from experience that charging NiMH cells at 0.5 C typically inputs 110-120 % or so of the charged cell capacity. On the other hand a 0.1C forming charge inputs ~160% of the charged cell capacity. Of course a forming charge does overcharge to a significant degree, as part of the process, but still, it seems that the losses would be higher at slower rates.
Thinking about this while typing though, you're probably correct.:) The resistance at higher charge rates would still add up to the same percentage. Then again, that would seem to support the idea that all charge rates would actually result in the same efficiency. Help Mr H!:hairpull:
Dave
Mr Happy
Flashlight Enthusiast
Like I was saying above, it's complicated. If anyone says to you "the charging efficiency of NiMH cells is X%", without giving a whole lot of qualifying details about the circumstances under which that efficiency is measured, then the number is not to be trusted.
What I might do is charge an Eneloop from empty on my C9000 at 200 mA, let it rest for an hour off the charger (to eliminate the top-off charge), and then discharge it at the same rate. I will report back what I find. My guess is that it will put about 1900 mAh in and take about 1700 mAh out, for a charge efficiency of about 90%. But as I say that is an estimate from prior experiments and I will do it afresh this weekend and record the numbers.
TakeTheActive
Enlightened
Mr Happy
Flashlight Enthusiast
While waiting for my experiment to complete, we can do a theoretical analysis based on resistance and energy. This won't tell us anything about charge storage, but it will tell us about energy losses.
Suppose a cell has an internal resistance of 0.05 ohms and a capacity of 2000 mAh (like an Eneloop, say). To apply 2000 mAh of charge at the 1C rate the current will be 2 A for 1 h. The power loss due to internal resistance is given by W = I²R, in this case 2 x 2 x 0.05 = 0.2 W (200 mW). Over 1 hour the energy lost will be 200 mWh.
Now consider charging at a 0.1C rate. The current will be 0.2 A for 10 h. The power loss will be 0.2 x 0.2 x 0.05 = 0.002 W (2 mW). Over 10 hours the energy lost will be 20 mWh.
We can see that purely in terms of energy losses, charging at a lower rate is more efficient.
We can also compare the energy lost with the energy storage capacity. For our 2000 mAh NiMH cell the total storage is about 1.2 x 2000 = 2400 mWh. So by charging at a 1C rate we have lost at least 8% of that to resistive heat dissipation. If the cell had a 0.1 ohm internal resistance it would be 15% or so.
Another interesting angle is to look directly at the voltages. When charging our Eneloop the voltage at the plateau during much of the charging period is around 1.4 V. If we are charging at 1C, the power being supplied is 1.4 V x 2 A = 2.8 W. When discharging, the plateau voltage is somewhere around 1.2 V, hence the power output at the 1C rate is 1.2 V x 2 A = 2.4 W. This gives an instantaneous estimate of efficiency as 2.4 / 2.8 = 86%. This considers only power, and does not consider the need to put more charge (amp-hours) in that you can get out. So the total energy efficiency will be less than this instantaneous estimate.
This all suggests it is going to be hard to get more than 80% of the energy out of an NiMH cell than you put into it.
wow... tnks you guys.. so much info ;)
my initial question was because my accucell-6 arrived (after months, damn Brazilian customs) and as it have the indication of "how much" amps it sent to the battery, this give an "idea" of how much its correct, to make sure the charger is reaching the full-charge on time, and also to set the capacity cut-off correctly.
so im comparing them with my other sony chargers, to see which one charge "the best", not cooking them, not overcharging ;)
a few initial tests show that the average value is about right (after more digging i found some values):
1C = 1.1
0.5C = 1.2
0.2C = 1.4
0.1C = 1.6
im doing some "preparations" to do 5s2p ni-mh battery packs
my initial question was because my accucell-6 arrived (after months, damn Brazilian customs) and as it have the indication of "how much" amps it sent to the battery, this give an "idea" of how much its correct, to make sure the charger is reaching the full-charge on time, and also to set the capacity cut-off correctly.
so im comparing them with my other sony chargers, to see which one charge "the best", not cooking them, not overcharging ;)
a few initial tests show that the average value is about right (after more digging i found some values):
1C = 1.1
0.5C = 1.2
0.2C = 1.4
0.1C = 1.6
im doing some "preparations" to do 5s2p ni-mh battery packs
Last edited:
Mr Happy
Flashlight Enthusiast
If you are looking at how much charge is fed into the batteries, then a rule of thumb is no more than 10-20% above the capacity. Therefore if you are charging Eneloops the charge supplied should ideally be 2200 mAh if charging at 1000 mA and really no more than 2400 mAh when charging at 500 mA. Numbers higher than this will involve some stress and overheating of the batteries. If you use the Accucel-6 to charge several batteries in series (which is how you would normally use it), you should use a charge rate of 1000 - 2000 mA and set a hard stop at 2200 mAh. Try to make sure the pack is evenly discharged before charging.
You can also use a balancing charge of 0.1 C for 14-16 hours, in which case it is best to use a timer and disable the other charge termination features so that all the batteries in the pack are fully saturated with charge.
What you are asking about here is not charging efficiency, but rather safe charging protocols to avoid damaging batteries. These are quite different things.
You can also use a balancing charge of 0.1 C for 14-16 hours, in which case it is best to use a timer and disable the other charge termination features so that all the batteries in the pack are fully saturated with charge.
What you are asking about here is not charging efficiency, but rather safe charging protocols to avoid damaging batteries. These are quite different things.
Last edited:
ok
done that, using the sony chargers. discharged all til 0.9v
the charger is doing that right now ;)
:twothumbs correct again. will wait the "balance charge" to finish, discharge (and test individual for "evenness") and recharge at 1C to see "what happens" ;) :popcorn:
Mr Happy
Flashlight Enthusiast
So here are the results of my charge/discharge experiment from post #5.
I discharged an Eneloop at 200 mA on the C9000 to make sure it was empty, then I charged it at 200 mA charge rate up to about 90% of maximum capacity. Then I discharged it at 200 mA to empty and compared the two numbers.
Charge supplied: 1707 mAh
Charge extracted: 1683 mAh
Coulombic efficiency: 1683 / 1707 = 98.6%
Conclusions:
In this test, the charge efficiency was very close to 100%. We might therefore predict that to obtain the maximum efficiency from an NiMH cell we should use a low charge/discharge rate, and we should de-rate the cell to about 90% of its maximum capacity.
I will follow up this test by repeating it at a 1000 mA charge/discharge rate to see how the result compares.
I discharged an Eneloop at 200 mA on the C9000 to make sure it was empty, then I charged it at 200 mA charge rate up to about 90% of maximum capacity. Then I discharged it at 200 mA to empty and compared the two numbers.
Charge supplied: 1707 mAh
Charge extracted: 1683 mAh
Coulombic efficiency: 1683 / 1707 = 98.6%
Conclusions:
In this test, the charge efficiency was very close to 100%. We might therefore predict that to obtain the maximum efficiency from an NiMH cell we should use a low charge/discharge rate, and we should de-rate the cell to about 90% of its maximum capacity.
I will follow up this test by repeating it at a 1000 mA charge/discharge rate to see how the result compares.
Last edited:
45/70
Flashlight Enthusiast
Very good, Mr Happy. The only thing I'm wondering is, how did you determine that you charged the cell to about 90% capacity? I'm guessing you stopped the charge at 0.9 x 2200mAh, where 2200mAh equals 110%, which is the presumed amount of input charge to achieve full charge. This would be better than charging with the C9000 until "DONE" appeared, as the C9000 terminates charge with good eneloops at 1.47 Volt, and then applies the "topping charge". At Higher charge rates, this would not allow as much charge to go into the cell, due to the fact that the peak voltage of the charging circuit would be higher at high charge rates vs. low charge rates and the 1.47 volt limiting "feature" of the C9000 will not allow this to happen.
I suppose the proper way to compare rates would be to use a power supply. Discharge the cell, as you did, and then apply a charge with a PS calculating the amount of mAh put in, say 2200mAh, and then do the discharge. It'd be a little more complicated than that though, considering voltage differences etc. during charging, and right now anyway, I don't have the time to undertake such a venture. I certainly don't expect you to either!
Carry on sir!
Take, if you're referring to why the cells you charged overheated, the only thing I can think of, is that your cells were around 80% of the original capacity, ie. due for replacement. This as you know, can cause heating of the cells when charging. That and the fact that the La Crosse chargers have a record for heating up cells when charging, anyway. In addition to other possible shortcomings, such as the algorithm used, I think this is mostly due to the close spacing of the cells in this line of chargers.
Dave
Mr Happy
Flashlight Enthusiast
This is very simple. I charged the cell until the C9000 indicated that 1707 mAh had been fed into the cell, and then I stopped the charge. I have not measured the actual capacity of this cell, but if I assume it is 1900 mAh I would have 1707/1900 = 90%. A cell is like a tank: I can try to fill it to the brim, or I can choose to fill it only 3/4 full or so. NiMH cells are much more efficient if you don't try to use that last 10%.
Mr Happy
Flashlight Enthusiast
Here is the result of that test, charging and discharging the cell at 1000 mA:
Charge supplied: 1707 mAh
Charge extracted: 1667 mAh
Coulombic efficiency: 97.7%
Recall that when charging and discharging at 200 mA the results were:
Charge supplied: 1707 mAh
Charge extracted: 1683 mAh
Coulombic efficiency: 1683 / 1707 = 98.6%
So charging is slightly more efficient at lower currents (as we might expect), but there is not a lot in it.
Hi Mr. Happy,
there might be a problem with this logic. If I recall correctly, Maha charges/discharges in bursts of 1A, so in both scenarios we have 1A fed in/got out; the only difference is the time each burst applied per second. More accurate results would be obtained if used a constant current charger at 200mAh and 1 Ah IMHO.
there might be a problem with this logic. If I recall correctly, Maha charges/discharges in bursts of 1A, so in both scenarios we have 1A fed in/got out; the only difference is the time each burst applied per second. More accurate results would be obtained if used a constant current charger at 200mAh and 1 Ah IMHO.
Last edited:
Mr Happy
Flashlight Enthusiast
It is true the C9000 does not use a continuous current of 200 mA, however over a charge/discharge period of several hours a one second pulsed cycle is so small on that time scale that it should be almost indistinguishable from a constant current when averaged out.
Mikl1984
Newly Enlightened
I have another results, more visible on high Ri cells
Comparison between Maha and iCharger(quasi-constant current) discharge results shows big difference on small currents (100-200 mA)
Almost the same results on 1A
45/70
Flashlight Enthusiast
97.7 vs. 98.6%. I'll go along with that. If we were to come up with a margin of error calculation however, due to all the possible variables, I'd say that the efficiency is about the same, for all practical intents and purposes. Your finite results do give slow rates the edge though, I'll give you that.:)
Thanks for going to the trouble of performing this test, Mr H. It is much appreciated. As is normal, learning is a never ending experience.
Dave
A good analogy would be a tank with holes on the top 30%, with the holes getting ever larger as you move towards the top. Filling the tank to 70% is perfectly efficient, more or less, but you waste more and more water the closer you get to the top.
I do pretty much the same. I've noticed that Eneloops have such a low internal resistance (only 15 minute cells like IC3 are less) that I have no problems dumping 2000 mA into a AAA, and letting my C9000 terminate on its own. I only do this if I'm around to make sure the cells terminate properly, but so far the C9000 hasn't missed with the Eneloops. Given the performance of the AAAs, I wouldn't hesitate to put 5 amps into an Eneloop AA. 7 years ago when I first wrote that post I wouldn't have dreamed of doing that unless the cell was specially designed for it.
