problem with measuring Li-ion battery capacity

Go To Last Post
27 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi All!

I am trying to measure capacity of a Li-ion battery(APR18650M1). For this goal, I am using a DAQ, a Mosfet key, a high power resistance with Matlab software. With all of these components and with creating PWM to on/off the Mosfet key, I discharged a full charged battery to completely full discharged with constant current.
Meanwhile, the number I am finding for battery's capacity is about .65 Ah while it's nominal capacity is about 1.1 Ah(cells are new and I bought them from an official retailer for A123 systems company).
For instance Once I discharged one of those cells with 1.04 A current,number of times that PWM worked was 2321(and each one took 1 sec), so according to these numbers capacity is about 2424 mAh, however whole time to run the program was about 3821.41 seconds which if I multiply it to 1.04A, the result is about 3927mAh.
The second number is close to battery's nominal capacity, but I think it is not acceptable since the time that I had current in the circuit was about 2321 seconds.

What do you think? if you have any suggestion.
Thanks in advance for your help and attention.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think you should pose these questions to the manufacturer. There is probably standard methods used.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

battery capacity has a lot of marketing gotcha's. What you care about is under a constant or varying load, how long until voltage under load is < x.

li-ion voltage curve under load is quite different than other types. It tends to be straight-line then decline very rapidly - say, in 5% of batter life it's declining.

ye ole lead acid battery decline almost linearly over time.

a good vendor should have honest curves of voltage vs. load and time.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

stevech wrote:
a good vendor should have honest curves of voltage vs. load and time.
:lol: two difficult to achieve KPIs

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A new battery needs a few cycles to develop full energy. Complete discharge is detrimental. For long life, the charge, discharge is kept in 60 to 40 percent range.

It all starts with a mental vision.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KitCarlson wrote:
A new battery needs a few cycles to develop full energy. Complete discharge is detrimental. For long life, the charge, discharge is kept in 60 to 40 percent range.

I think you must be thinking of a different chemistry. Typically after being manufactured li-ion cells are run through some special conditioning cycles (not a normal cycle, but I'm not sure of the difference) and are then shipped fully ready to use. We test a lot of cells at work and I've never seen one that had an increase in capacity after the first cycle.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KitCarlson wrote:
A new battery needs a few cycles to develop full energy. Complete discharge is detrimental. For long life, the charge, discharge is kept in 60 to 40 percent range.
But what KIND of battery?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks all for your responses. As I said, I am discharging under constant current(Using a constant load , DAQ and a MOS key).After discharging the current summation doesn't equal to battery's capacity(measuring intervals are 1 seconds). Any IDEA!?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Meanwhile, the number I am finding for battery's capacity is about .65 Ah
...
For instance Once I discharged one of those cells with 1.04 A current,number of times that PWM worked was 2321(and each one took 1 sec), so according to these numbers capacity is about 2424 mAh
??

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have no idea where the 0.65 amp-hour figure comes from, but it seems the "constant current" discharge is done using PWM for one second, followed by a rest period during which the voltage is sampled and fed to Matlab. That is really a pulsed discharge, which would tend to a slight increase in the actual capacity.

But I wonder how long it takes for the current to stabilize at one amp during each one second pulse. A slow ramp up from zero would result in a much larger apparent capacity.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ereihani wrote:
Hi All!

I am trying to measure capacity of a Li-ion battery(APR18650M1). For this goal, I am using a DAQ, a Mosfet key, a high power resistance with Matlab software. With all of these components and with creating PWM to on/off the Mosfet key, I discharged a full charged battery to completely full discharged with constant current.

I believe using PWM will not give reliable results.

For almost all batteries, you will find that capacity is higher at low discharge current and vice versa. Discharging with PWM is pulsed high current, you should discharge with constant current.

You can check some results here: http://lygte-info.dk/info/Batter...

Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The datasheet shows full capacity even with 10A discharge.
http://www.flinkenberg.fi/batter...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

I believe using PWM will not give reliable results.
Peter

I am trying to build the tester circuit from scratch. Do you have any reliable circuit in mind so I can get accurate results?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KitCarlson wrote:
A new battery needs a few cycles to develop full energy. Complete discharge is detrimental. For long life, the charge, discharge is kept in 60 to 40 percent range.

what is the time length of battery rest?

nleahcim wrote:

We test a lot of cells at work and I've never seen one that had an increase in capacity after the first cycle

Do you have any documented procedure for testing batteries?

dak664 wrote:

But I wonder how long it takes for the current to stabilize at one amp during each one second pulse.

Do you have any idea how I can get the stabilization period for a battery? What is this period called in technical documents?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

1.
Load battery.

2.
Connect a bulb 4V/0.5A to the battery. (or more bulbs in parallel)
Measure the current through the bulb and start counting time. After the battery voltage drops to 2V, stop discharging.
Capacity is current[mA] * time[hour].

3. If the cell is new, do this more times and watch if capacity grows.

I suspect you do not charge the cell fully.
How do you charge it?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

your cell is 1.1Ah
what happens if you discharge with 110mA instead of 1.1A?

and what happens if you discharge with only 11mA?

how close to the 1.1Ah figure do you get then?

note that the 1.1Ah figure gives an abs max number. At that point the cell is competely empty.

we work with CR2032s a lot they are specced to be 220mAh but we only get 150mAh out of them as the load we use is rather high, so try out these different loads and see were you end.(effectively you make a rudemenatry load graph by doing so 100Hrs(expected 11mA load on time) = 4 days so start on friday that gives it 2 days in the weekend without attendance

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

meslop:
Look at the link in my post from Dec,10 6:17 pm.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ereihani wrote:

Do you have any idea how I can get the stabilization period for a battery? What is this period called in technical documents?

Not so much the stabilization period for battery current, although in a pulsed discharge the voltage can drop rapidly as the available ions are depleted and new ones take longer to migrate out of the plates and across the electrolyte.

The PWM control loop will take a finite time to settle down to 1 amp and the above voltage change could also affect the settling time. On an oscilloscope, does the voltage across the sense resistor appear as a sharp square wave with exactly 1 second duration, or is there considerable overshoot, undershoot, or ringing?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Q. What do you measure the capacity of a Li-ion in?
A. Wi-ildebeests.

Well, it is almost cracker pulling time.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Q. What do you measure the capacity of a Li-ion in?
A. Wi-ildebeests.


Wouldn't that be wildebeests per fortnight?

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
Quote:

Q. What do you measure the capacity of a Li-ion in?
A. Wi-ildebeests.


Wouldn't that be wildebeests per fortnight?

I guess it depends on how you define capacity.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

theusch wrote:
Quote:

Q. What do you measure the capacity of a Li-ion in?
A. Wi-ildebeests.

Wouldn't that be wildebeests per fortnight?

I guess it depends on how you define capacity.


The analogy is quite obvious, isn't it? If you measure it in "wildebeests" then it is analogous to "amps" maximum draw of the battery. If you measure it in "wildebeests+fortnight", then it is analogous to "amp-hours".

Right? I will now retire to my drive home in sub-zero (F) temperatures.

And I still don't know why OP's numbers are so far off. We do extensive testing in certain battery apps, and the end result is that the capacity generally matches the battery specs. But normally we'd have a steady draw.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
The analogy is quite obvious, isn't it?

No.
If you talk about the capacity of your car engine(or indeed the luggage carrying space of your car), or a beer mug, it is strictly a volume metric. There is no time element to capacity in in many usages of the term. However, the Li-ion/Wi-ildebeest thing was a cheap joke, and will not survive stringent analysis.

I have to retire now, as I will have to get out early to scrape the ice from my car in sub(celsius) zero temperatures.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm gonna open a battery store and advertise the capacities fudged up about 10%. I'm gonna call it the "Li-ion King".

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bobgardner wrote:
I'm gonna open a battery store and advertise the capacities fudged up about 10%. I'm gonna call it the "Li-ion King".

I like it. I'm going to open up a rival store over the way called "The Lying Li-Ion King". I'm going to bump up the capacities by 20%.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Google "ultrafire 18650" - with these you are lucky to get a battery that actually works!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
Google "ultrafire 18650" - with these you are lucky to get a battery that actually works!

Yes, a client of mine drew this to my attention when I was suggesting using an 18650. He ended up buying from a reputable source, but the difference in price between the reputable and the unknown is vast.

Four legs good, two legs bad, three legs stable.