Calculating Loss of NiMH Charge Over Time at a Given Temperature

[Artric]

Wikipedia says that a standard NiMH battery would self-discharge about half of its capacity at a constant 68 degrees F over the course of a year, but that data seems quite wrong from experience.

But elsewhere, I found that NiMH cells are generally quoted to lose between .5% - 1% of their rated capacity per day. At around 68F, one person stated that they and others normally see around .7% loss per day (although I believe this is for smaller cells).

Could anyone chime in with some in-depth knowledge on what is to be expected with the Insight packs? I know they are essentially a large number of D-cell NiMH packs welded together, so I would expect them on average to lose the same percentage. But then there is also the variable of temperature. How would that affect discharge?

The reason why I ask is that I would like to be able to roughly calculate what would be appropriate timing for those in the cold weather to charge their batteries when storing their cars for the Winter (such as in a garage with an average temperature of 20 degrees, for instance). That way, they can set it on a timer to either charge for a certain amount of time daily or weekly.

With the help of Peter's amazing BCM gauge (I can't imagine how I ever lived without this device), I found that it takes about 11:20 to charge 1%, per the BCM's readings at 350ma in my case.

That seems to mean that via the information above, running it for 15 minutes per day would be plenty - but would it? How would the temperature factor in? And is there any reason that charging weekly would be more beneficial than charging daily (assuming someone has a daily, not a weekly timer)?

Any insight in this area would be greatly appreciated. I'm helping a forum member calculate the amount of time they should charge while their car is in storage. If anyone has any hard numbers, I'd love to hear them!

Unfortunately, I drive my Insight too much to be able to get a meaningful measurement on self-discharge, and I also am not seeing the cold, cold temperatures in FL that you all must put up with in the North!

Thanks in advance for anyone able to contribute to the community by helping with this data! If it already has been done, and I failed to find it, please point me in the right direction.

[uhtrinity]

As far as cold weather charging, I have left mine plugged in for upto 3 days at a time. But I use a lower voltage on the charger, ~171 volts. You could theoretically leave one plugged in all winter and at most you would only draw 5 - 10 watts plus fans if you choose to run them.

Mine sits at 15 watts in cold weather once charged for two packs.

[aisbell]

Originally Posted by Artric

The reason why I ask is that I would like to be able to roughly calculate what would be appropriate timing for those in the cold weather to charge their batteries when storing their cars for the Winter (such as in a garage with an average temperature of 20 degrees, for instance). That way, they can set it on a timer to either charge for a certain amount of time daily or weekly.

I don't think there's any advantage to charging periodically during the winter (please, someone chime in if there is a good reason to do so). In fact, recharging only to lose much of this charge through self-discharge would just add to the number of charge-discharge cycles, each of which degrades the battery a bit.

The rate of self-discharge is temperature-dependent. Storing an Insight at a low temperature would reduce the self-discharge rate considerably.

I recharge/rebalance only just before driving my Insight after it's been stored all winter. So far, this hasn't caused any problems.

[IamIan]

The Self Discharge rate is related to both the temperature and the SoC of a given cell.

Higher SoC has a faster self discharge rate than a lower SoC.

Higher temperature has a faster self discharge rate than lower temperature.
( keep in mind lower temperature also has lower total capacity )

The self discharge rate for the cells will change over time as they age , and are used, heated , etc...

The only NiMH that go as low as ~15% SoC loss per year are the Low-Self Discharge NiMH ... usually labeled as 'ready to use' NiMH ... or LSD-NiMH.

With regards to Gen-I Insight battery pack Self Discharge rates in particular.

Attached bellow is a ~1 year series of tests I ran... showing the daily mAh of self discharge vs the number of days ... the graph is logarithmic and levels off close to ~10 mAh a day once the remaining SoC in the cells drops low enough... the loss of usable energy is more than just the loss of mAh , as the cell's nominal voltage also gets lower as the SoC gets lower.

[Artric]

Originally Posted by uhtrinity

As far as cold weather charging, I have left mine plugged in for upto 3 days at a time. But I use a lower voltage on the charger, ~171 volts. You could theoretically leave one plugged in all winter and at most you would only draw 5 - 10 watts plus fans if you choose to run them.

Mine sits at 15 watts in cold weather once charged for two packs.

I was thinking the same thing - pick the voltage and set it/forget it. I actually use that trick on my booster charger, so I get a fast charge up to x volts, then it slowly tapers off, until the 350ma charger totally takes over nd brings it up to 100%.

The person I'm helping skipped hooking up the fan for now, as she recognizes the wastefulness of running a fan at 20 degrees! I should have thought of setting the external voltage and wiring up a rheostat on her unit to allow for easy adjustment. We'll see how it works out.

I'm not as educated with all electrical components as I'd like to be, but since I use a molex connector, couldn't I make a little harness addition that sandwiches between the charger and the connection to the battery that uses a resistor to limit the current max voltage to a lower value, like the ~171 you mention? I usually see resistors that lower voltage to a much larger degree - are there ones available that would limit it only around 3-5%? And would it just need to be connected on the positive lead?

Obviously, I would dive much deeper and make sure everything is tested 100% before passing onto another member, but any info that can be given to kick-start my process would be super. There's a really intelligent guy at my local Radio Shack that I might start a chat with.

Thanks!

[Artric]

Another quick thought: Rheostats/Potentiometers are generally meant for low voltage, so it wouldn't be appropriate to put it in line with the high voltage in this way, correct? Would there be an indirect way to use it that wouldn't be too incredibly complicated?

By the way, it's not just a lucky coincidence, I used molex connectors on my chargers because of the potential for adding piggyback functionality, etc in the future (though I didn't really know what at the time, aside from controlling fan speed).

[Artric]

Thanks so much for the data! Can I ask at what kind of temperatures this was collected at (or rather the approximate date when you started/location)?

Also, you mentioned voltage drop. I assume if you took the time to collect that data, you probably charted voltage as well? I would love to see that if available and you're willing to share.

I'm especially curious to know what the end voltage was for the experiment. Do I understand correctly that you let this battery set, uncharged for that period of time?

If so, this is really valuable information for me, as I'm trying to find a low-cost battery for going parallel, and will probably end up with something that's been sitting for awhile.

Thanks again!

Originally Posted by IamIan

The Self Discharge rate is related to both the temperature and the SoC of a given cell.

Higher SoC has a faster self discharge rate than a lower SoC.

Higher temperature has a faster self discharge rate than lower temperature.
( keep in mind lower temperature also has lower total capacity )

The self discharge rate for the cells will change over time as they age , and are used, heated , etc...

The only NiMH that go as low as ~15% SoC loss per year are the Low-Self Discharge NiMH ... usually labeled as 'ready to use' NiMH ... or LSD-NiMH.

With regards to Gen-I Insight battery pack Self Discharge rates in particular.

Attached bellow is a ~1 year series of tests I ran... showing the daily mAh of self discharge vs the number of days ... the graph is logarithmic and levels off close to ~10 mAh a day once the remaining SoC in the cells drops low enough... the loss of usable energy is more than just the loss of mAh , as the cell's nominal voltage also gets lower as the SoC gets lower.

[3-Wheeler]

Artic,

My Insight sat for 4 months over the summer while working on the tail of the car.

If I recall correctly, battery voltage was topped off at the end of May at 166 or so, and in October the voltage was measured at 148 volts DC or so.

Temperature varied from 50 at night to over 95°F in the day, and I think even hotter in the garage where the heat build-up occurred in the daytime.

Jim.

[IamIan]

Originally Posted by Artric

Thanks so much for the data! Can I ask at what kind of temperatures this was collected at (or rather the approximate date when you started/location)?

Also, you mentioned voltage drop. I assume if you took the time to collect that data, you probably charted voltage as well? I would love to see that if available and you're willing to share.

I'm especially curious to know what the end voltage was for the experiment. Do I understand correctly that you let this battery set, uncharged for that period of time?

If so, this is really valuable information for me, as I'm trying to find a low-cost battery for going parallel, and will probably end up with something that's been sitting for awhile.

Thanks again!

You're welcome.

If you know if it is going to sit for a long period of time LSD-NiMH would be worth considering ... although they are more expensive ... and they are not recommended for more than 2C charge or discharge rates.

All the tests were all done and stored at room temperature ... which even with heating in the winter and AC in the summer will still have temperature fluctuations.

I do not have a temperature controlled box to run these kind of tests in ... at least not yet

Terminal voltage can be very misleading when it comes to NiMH ... remember they can suffer from the condition known as 'voltage depression' ... I did not use the method some people do , by just taking measurements of the terminal voltage , and then assuming some SoC% based on that terminal voltage... I did not feel I would get the accuracy level I wanted with that method... however I do have that the initial voltage graph prior to each discharge ( attached bellow ) ... I had hoped to eventually correlate some kind of adjustment to narrow the SoC % determination on NiMH based on terminal Voltage... but that is just one other project on hold for more time and money.

The ~308 days of testing started on 3-25-2009

- - - - - -

My methodology was as follows:

#1>
I took the 40 sub pack 6 cell sticks that I currently have separate from the 20 in the car.

#2>
Determine the capacity for each one ... Fully charge it than discharge it ... repeat process until I have reasonably consistent results for a given 6 cell subpack stick... I wanted to know the usable discharge mAh ... the number of mAh charged into the battery is not the same as what it will give up ... In Hind sight now, mWh would have been a better thing to measure , as it would have accounted for the fluctuations in Nominal Voltage.

#3>
Charge up the sticks... noting the date.

#4>
Discharge sticks after some period of days ... initially started at 1 day ... then 2 days ... etc.

#5>
The subpack sticks used to be discharged in the first couple days were then recharged and allowed to sit again for a slightly longer period of time , while the other subpack sticks continued to sit for longer periods of time ... this gave me more data sets / data points for discharged mAh in the initial couple of days .... which allowed me to plot the curve reasonably well to the average.

#6>
Comparing the discharged mAh against the expected mAh from Step #2 for each individual stick... this gives a change in expected discharge mAh over time in days.

#7>
By comparing the results across multiple tests during the series over the 308 days I ran this series of tests I was then able to plot the 'best fit' curve shown in the previously attached picture.

- - - - - - -

Eventually one of my many projects will include a temperature controlled box... with spare time and spare money that is .... hahahaha I crack myself up ... anyway ... eventually when I get around to it ... I would like to re run this series of tests with more sensor logs and controls ... including mWh instead of mAh... most likely that will not happen any time soon.

[Artric]

Thanks so much for the responses!

IamIan - great scientific work! No worries about the temperature controlled atmosphere - it's not like the Insight is going to see a constant temp year round anyway. Maybe storing them in a simulated Insight outside would actually give the most accurate results (at least for your region)?