Internal impedance/resistance of batteries

60 posts / 0 new
Last post

Pages

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark
Internal impedance/resistance of batteries

Internal impedance/resistance of batteries

One parameter that is sometimes used when talking about batteries is the internal impedance/resistance. The difference between impedance and resistance is that impedance is for AC and resistance is for DC. Not all people are aware of this difference and might use the wrong term.
In this article I have split impedance and resistance into different chapters, but first something about measuring small resistances/impedances.

DSC_2204

I will use two LiIon batteries for examples, but the technique and math can be used for any type of battery.
The green one (NCR18650A) is fairly new and has a low resistance/impedance.
The red one (AW IMR16340) is old and has a high resistance/impedance.



Measuring basic

Internal impedance is in the milliohm range, i.e. between 0.001 ohm and 1 ohm. This is a low value to measure and requires special technique. As a comparison a good quality DMM probe has 20 to 30 mohm (milliohm).
The two examples I show below is only for one battery pole, it is necessary to connect both battery poles!

DSC_2202

The best way to measure these low impedances, is to split current and voltage into separate cables and contact points. I.e. use one set of wires for the voltage measurements and another set of wires for the load current.
The problem is that no people has four hands and it is very difficult to connect with only two hands. Professional equipment usual has a rig that can handle this. For some purposes it is also possible to use kelvin clips (That is a special alligator clip, where the two parts of the clip each has a wire connected).
This connection will eliminate both contact resistance and any voltage drop in the probes and cables.

DSC_2203

A more practical way to measure, is to use one contact point, with two wires connected. In the above picture I have done it with test leads. This will give a small error due to contact resistance and voltage drop in the probe tip. To minimize that I have the current wire closest to the battery.
This connection can add a few milliohms to the measurement, depending on how it is made.



Resistance (DC)

Resistance is a DC measurement, this is usual measured by one of two methods, that both requires the same setup.
The resistor is selected to give a current in the actual working range for this battery, it will probably have to be a power resistor.

ResistanceMeasurement

When measuring with load connected, do not wait for a stable value, that might not occur before the battery is empty!


First method

In this method the battery voltage is first measured without the resistor connected, then the resistor is connected and the voltage is measured again, the internal resistance is calculated with this formula:

Resistance = (volt_no_load - volt_at_load) * load_resistance / volt_at_load
or
load_current = voltage_at_load / load_resistance -> load_current
Resistance = (volt_no_load-volt_at_load) / load_current



Second method

With this method both measurements are made with a load on the battery, this makes it possible to measure the resistance closer to the actual working point.

The formula to calculate resistance is this.
Note: measurement 1 is the high resistance, i.e. low current measurement.

current1 = voltage1 / resistance1
current2 = voltage2 / resistance2
Resistance = (voltage1 - voltage2) / (current2 - current1)



Examples DC

DSC_2193

At no load: 4.1399

DSC_2194

At 10 ohm load: 3.9987 volt -> 3.9987/10 -> 0.39987 ampere

DSC_2195

At 5 ohm load: 3.8661 volt -> 3.8661/5 -> 0.77322 ampere

Internal resistance no load to 10 ohm: (4.1399-3.9987)/0.39987 -> 0.353 ohm
Internal resistance no load to 5 ohm: (4.1399-3.8651)/0.77322 -> 0.355 ohm
Internal resistance 10 ohm to 5 ohm: (3.9987-3.8661)/(0.77322-0.39987) -> 0.358 ohm


DSC_2196

At no load: 4.1940

DSC_2199

At 10 ohm load: 4.1466 volt -> 0.41466 ampere

DSC_2200

At 5 ohm load: 4.0969 volt -> 0.81938 ampere

DSC_2201

At 2 ohm load: 3.9831 volt -> 1.99155 ampere

Internal resistance no load to 10 ohm: (4.1940-4.1466)/0.41466 -> 0.114 ohm
Internal resistance no load to 5 ohm: (4.1940-4.0969)/0.81938 -> 0.118 ohm
Internal resistance no load to 2 ohm: (4.1940-3.9831)/1.99155 -> 0.106 ohm
Internal resistance 10 ohm to 5 ohm: (4.1466-4.0969)/(0.81938-0.41466) -> 0.123 ohm
Internal resistance 5 ohm to 2 ohm: (4.0969-3.9831)/(1.99155-0.81938) -> 0.097 ohm
Internal resistance 10 ohm to 2 ohm: (4.1466-3.9831)/(1.99155-0.41466) -> 0.104 ohm



More about internal resistance

ResistanceCurve1

The internal resistance show how the voltage will drop when the battery is loaded. The slope of the yellow line is the internal resistance of the battery in the above chart. I.e. the single number, that is the internal resistance, can show the same as the above chart.

ResistanceCurve2

Lets try another battery, here I have added 3 yellow lines to show 3 different value of the internal resistance in the same battery, i.e. it changes with load, especially when the load is too high for the battery.

The internal resistance will change with temperature (decreases with increasing temperature), battery age (Increases with age), but also with charge on the battery (Increases when the battery is nearly empty).



What can it be used for

Keeping track of internal resistance on a battery will make it possible to see when the battery is getting too old, because the internal resistance will increase.

When designing with batteries, the internal resistance is also useful to estimate voltage drop from the battery, depending on load.



Impedance (AC)

Impedance is a AC measurements and the custom is to use 1000 Hz (1 kHz) for batteries. This makes it impossible to measure with an ordinary DMM.
The impedance is considerable lower than the resistance as can be seen in these measurements.



Lab setup

In a electronic lab there will usual be the necessary equipment to measure the impedance.

ImpedanceMeasurement

Because the impedance is measured at 1000 Hz a generator is needed.
Because I do not want a DC load on the battery or DC into my generator, I have added a capacitor in series with the generator, the actual value is not important, somewhere in the 5uF to 1000uF range will be fine.
I do also need a resistor in series with the generator, this resistor is used for comparison to the internal impedance of the battery, I have select 0.5 ohm.
Then I need one or two mV AC DMM's, one is enough, because it can be moved between the two measurement points, two DMM's is for the lazy people. This DMM must be able to measure 1000 Hz AC!

To calculate the impedance:

Impedance = voltage_battery / voltage_resistor * resistor

DSC_2191

Here is the actual setup.



Examples AC

DSC_2189

AC across battery: 4.566 millivolt
AC across 0.5 ohm resistor: 15.100 millivolt

Impedance is: 4.566/15.100*0.5 -> 0.151 ohm


DSC_2185

AC across battery: 1.513 millivolt
AC across 0.5 ohm resistor: 15.123 millivolt

Impedance is: 1.513 / 15.123 * 0.5 -> 0.050

Impedance meter

For people that need/want to measure impedance frequently, it is possible to buy a meter that replaces all of the above setup.

DSC_2175

A new IMR16340 battery is about 31 milliohm (Measured on a Efest IMR16340).

DSC_2176





What can it be used for

The AC impedance has mostly the same usage as the DC resistance, but with a impedance meter it is much easier to measure, even in a circuit (Like on a mounted car battery). Measuring in circuit will give an error, but as long as the battery load is light, the error will be small.

The usages are:
Keeping track of internal impedance on a battery will make it possible to see when the battery is getting too old, because the internal impedance will increase.



Equipment

DSC_2205

This is a homemade resistance box, it is very useful for many load tests. Inside is some power resistors:

DSC_2206

Using the power resistor stand alone would limit the power and make it more difficult to connect to the resistor.
I do use 50 watt resistors in the box, but 20 minutes with 25 watt will heat the box to 80°C.

DSC_2209

My usual DMM's that can easily measure mV AC at 1000 Hz.

DSC_2208

A capacitor with easy connection.

DSC_2210

SM8124 battery impedance meter. This is a cheap meter at about $40.
The precision is acceptable, notice it uses two wires to each probe, but the input protection is very bad according to the instruction sheet (Do not swap + and - probes).

DSC_2213

A function generator, that can do just about any type of curve. In this case here I only need a sinus at 1 kHz.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

Edited by: sb56637 on 04/26/2013 - 08:07
Milan
Offline
Last seen: 10 hours 21 min ago
Joined: 05/08/2012 - 12:13
Posts: 310
Location: Czech Republic

Where can you get the SM8124 for 25 USD?

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

Milan wrote:
Where can you get the SM8124 for 25 USD?

That is a typo, I forgot that I am using £ in my ebay setup. I will fix it.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

Milan
Offline
Last seen: 10 hours 21 min ago
Joined: 05/08/2012 - 12:13
Posts: 310
Location: Czech Republic

HKJ wrote:
notice it uses two wires to each probe, but the input protection is very bad according to the instruction sheet (Do not swap + and – probes)

Well, I use KT-98 and it measures IR in both directions, might be a better alternative if that ebay metter is supposed to burn the first time you switch poles.
HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

Milan wrote:
HKJ wrote:
notice it uses two wires to each probe, but the input protection is very bad according to the instruction sheet (Do not swap + and - probes)
Well, I use KT-98 and it measures IR in both directions, might be a better alternative if that ebay metter is supposed to burn the first time you switch poles.

It does not have a 4-wire connection, if you can use the rel button on the range, it might be useable.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

ImA4Wheelr
Offline
Last seen: 1 week 4 days ago
Joined: 02/03/2013 - 14:51
Posts: 7935
Location: SC

This is a nice write up. It should be stickied or something for easy future reference.

Pulsar13
Pulsar13's picture
Offline
Last seen: 10 months 1 week ago
Joined: 02/05/2012 - 10:23
Posts: 1909
Location: Malaysia

Nice one HKJ, thanks. Very useful info to measure battery internal resistance with stuff we already have.

sb56637
sb56637's picture
Online
Last seen: 11 min 31 sec ago
Joined: 01/08/2010 - 09:29
Posts: 7070
Location: The Light

Very nice article! You answered a question I had. Thanks very much for the details. Sticky’d.

Budget Light Forum ...where Frugal meets with Flashlight!

ohaya
Offline
Last seen: 3 years 5 months ago
Joined: 03/16/2013 - 19:01
Posts: 5337
Location: US

Accucel 6 has a way to measure IR supposedly. I have an Accucel, but haven’t done it, because it apparently involves using some additional cabling which I don’t think that I have. If I did do that, do you all think it would be an accurate IR measurement?

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

ohaya wrote:
Accucel 6 has a way to measure IR supposedly. I have an Accucel, but haven't done it, because it apparently involves using some additional cabling which I don't think that I have. If I did do that, do you all think it would be an accurate IR measurement?

It will probably do a DC measurement at a unknown current. If it uses four wires and you connect them correctly to the battery, I would expect it to do a accurate measurement.

But if it only uses two wires, it will not be very accurate.

What is most interesting is the repeatability, i.e. does it give the same reading (or very close) each time. If it does, you can use it to follow the ageing of the batteries.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

ohaya
Offline
Last seen: 3 years 5 months ago
Joined: 03/16/2013 - 19:01
Posts: 5337
Location: US
HKJ wrote:

ohaya wrote:
Accucel 6 has a way to measure IR supposedly. I have an Accucel, but haven’t done it, because it apparently involves using some additional cabling which I don’t think that I have. If I did do that, do you all think it would be an accurate IR measurement?

It will probably do a DC measurement at a unknown current. If it uses four wires and you connect them correctly to the battery, I would expect it to do a accurate measurement.

But if it only uses two wires, it will not be very accurate.

What is most interesting is the repeatability, i.e. does it give the same reading (or very close) each time. If it does, you can use it to follow the ageing of the batteries.

Hi,

I didn’t look into it much, because, like I said, I got info that you had to use the side-port and I think cable using something like a balanced charging type configuration, and I only have the default cables that came with the Accucel. But, they do make a deal about it, with even a small yellow flyer that says it’s like the most important thing to measure (but no instructions how to cable) :).

Thanks. If I ever get around to doing it, I’ll post about it.

Jim

Hikelite
Offline
Last seen: 6 months 4 days ago
Joined: 07/13/2011 - 16:18
Posts: 3888
Location: Honeyland

Thanks for this great info HKJ!

DB Custom
DB Custom's picture
Offline
Last seen: 6 days 17 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20726
Location: Heart of Texas

Thanks HKJ for a very interesting read! Of course, I’ll have to read it about 10 more times for 25% of it to sink in….

MRsDNF
MRsDNF's picture
Offline
Last seen: 1 month 1 week ago
Joined: 12/22/2011 - 21:18
Posts: 13473
Location: A light beam away from the missus in the land of Aus.
DBCstm wrote:
Thanks HKJ for a very interesting read! Of course, I’ll have to read it about 10 more times for 25% of it to sink in….

+1 except I’d be lying if I understood 2%.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

DENGOH
DENGOH's picture
Offline
Last seen: 1 year 10 months ago
Joined: 10/06/2012 - 05:26
Posts: 2157

As usual, sure I can learn new thing by reading HKJ post. Thanks for sharing, HKJ.
I had long wanted to have easy way to keep track on my battery condition, but recently lazy to keep track on them.

cool i'll see you when you get there

photon1k
photon1k's picture
Offline
Last seen: 4 years 5 months ago
Joined: 02/11/2011 - 23:29
Posts: 901
Location: Lotus Land, Canada

I think I remember seeing an iCharger 208B in one of your many informative posts. Have you looked at its IR measurement function, and if so, what do you think of it?

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

photon1k wrote:

I think I remember seeing an iCharger 208B in one of your many informative posts. Have you looked at its IR measurement function, and if so, what do you think of it?

Yes, I have a iCharger 208B or two. They can measure IR, but I have not really used it or analysed it. It uses two wire and will give an error due to that.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

photon1k
photon1k's picture
Offline
Last seen: 4 years 5 months ago
Joined: 02/11/2011 - 23:29
Posts: 901
Location: Lotus Land, Canada

HKJ wrote:

photon1k wrote:

I think I remember seeing an iCharger 208B in one of your many informative posts. Have you looked at its IR measurement function, and if so, what do you think of it?

Yes, I have a iCharger 208B or two. They can measure IR, but I have not really used it or analysed it. It uses two wire and will give an error due to that.

Thanks for the quick responseSmile. I will have to study the OP more closely to learn about what the error may be.Innocent

snakebite
snakebite's picture
Offline
Last seen: 3 weeks 6 days ago
Joined: 11/20/2013 - 20:21
Posts: 2054
Location: dayton oh

meant for capacitors but does ok for batteries.once you get the feel for good esr readings for a certain cell go/nogo tests are fast.like when grading laptop pack salvage cells.
http://www.amazon.com/Anatek-Corporation-blueesrkit-Blue-ESR/dp/B005NB6CKS
i have one of the original kits from 15+ years ago.great tool.

nml
Offline
Last seen: 7 years 3 months ago
Joined: 06/24/2014 - 15:19
Posts: 1

I’m new to all of this but would like to replicate this set up. Would you be able to explain how the resistance/capacitor boxes were built.

I bought a Reed R5001 LCR meter for measuring impedance. Can I use that with this set up?

Thanks!

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

nml wrote:
I'm new to all of this but would like to replicate this set up. Would you be able to explain how the resistance/capacitor boxes were built.

That is just resistors/capacitors connectors to the binding posts. When doing a lot of different test setups this is much easier than soldering the wires each time.

Even though I have a bunch of electronic loads, I uses that resistor box all the time.

nml wrote:
I bought a Reed R5001 LCR meter for measuring impedance. Can I use that with this set up? Thanks!

Probably not, the problem is the battery voltage. When measuring the internal resistance you are not going to work on a voltage free circuit, due to the battery it will have about 4 volt.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

USSR
Offline
Last seen: 5 years 5 months ago
Joined: 07/12/2012 - 15:13
Posts: 74
Location: Pakistan

 @HKJ

1: while measuring internal resistance , do we need to be concerned with the wattage rating of a resistor ?

2: how can we conclude some usefull information from internl resistance figure ? like how good/bad is it ? how much int r is good how is bad ? bad would mean it cant supply great amperagess  ?

3:Battery is pretty much dc source so how does ac comes into play ?

pardon me for those bunch of questions as it has been long time i studied electronics/circuits.

 

 

 

 

 

 

 

 

 

 

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

USSR wrote:
1: while measuring internal resistance , do we need to be concerned with the wattage rating of a resistor ?

Yes, but if you are fast, you might get away with some overloading of the resistor.

 

USSR wrote:
2: how can we conclude some usefull information from internl resistance figure ? like how good/bad is it ? how much int r is good how is bad ? bad would mean it cant supply great amperagess  ?

Yes, but I do not have a table with good/bad values for different batteries.

USSR wrote:
3:Battery is pretty much dc source so how does ac comes into play ?

Probably because it is easy to make a device that can measure impedance in-circuit, i.e. you can check the battery without disconnecting it. AC and DC gives different values, but in both cases the battery will be bad when the value has increased significantly from the initial value.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

USSR
Offline
Last seen: 5 years 5 months ago
Joined: 07/12/2012 - 15:13
Posts: 74
Location: Pakistan

how much should be the battery charged (ideal no load voltage ) to do this test .

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark

USSR wrote:

how much should be the battery charged (ideal no load voltage ) to do this test .

There is no rule about that, but you want the same charge each time to get the best comparison.

One charge level that is easy to get each time is fully charged.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

Milan
Offline
Last seen: 10 hours 21 min ago
Joined: 05/08/2012 - 12:13
Posts: 310
Location: Czech Republic

Only lead acid batteries needs to be tested fully charged (and rested ideally).
IR of li-ion batteries does not change with state of charge (only very slightly) and can be neglected for practical reasons.

jhalb
jhalb's picture
Offline
Last seen: 4 years 5 months ago
Joined: 03/23/2015 - 21:24
Posts: 1350
Location: California

HKJ, I always love checking out your posts. You have some of the coolest toys. Do you have a stereo microscope yet? I just bought one… Glasses

HKJ
HKJ's picture
Offline
Last seen: 5 days 12 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7425
Location: Copenhagen, Denmark
jhalb wrote:
HKJ, I always love checking out your posts. You have some of the coolest toys. Do you have a stereo microscope yet? I just bought one… Glasses

No stereo, only a flat view for me.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

snakebite
snakebite's picture
Offline
Last seen: 3 weeks 6 days ago
Joined: 11/20/2013 - 20:21
Posts: 2054
Location: dayton oh

that chart while interesting is depicting a lousy battery.if the ohmic value were preceded by a decimal it would be close.
in my experienc at room temp the rise is not as bad as depicted.probably old data.

gauss163 wrote:
Milan wrote:
IR of li-ion batteries does not change with state of charge (only very slightly) and can be neglected for practical reasons.

No, IR does depend on SOC, especially for low SOC and low temperature, e.g. see below


Graph excerpted from p. 1-4 of Challenges and Solutions in Battery Fuel Gauging, by Yevgen Barsukov.

The Miller
The Miller's picture
Offline
Last seen: 1 year 10 months ago
Joined: 12/14/2015 - 12:08
Posts: 9908
Location: Charente France

Nice to see this topic
I have a few questions.
My Liitokala gives a value in mR and I think this is the internal resistance of the cell

When I receive a cell I charge it, do discharge mAh test and connect/disconnect the cell in the charger to see what the lowest mR is (how a cell is placed affect the mR sometimes and I want to rule that out)
I then write the mAh and mR on a small piece of paper and stick it on the cell with tape.

I use my KeepPower 2200 measured 2300mAh for most testing and they got used the most of my cells.

They all three read 53mR when first tested.
Now the two most used (a lot of charge cycles went into the thorfire tests) now read 75mR and the third 61mR.
The two I always use together, if not in series then in a SRK never alone.

I have cheap rubbish cells that measure 125mR from the start and not getting used, and a 26650 Ultrafire that gives 35mR and 1646mAh that I use the first run of single 26650 flashlights since it is the only 26650 I currently own (2 good ones in transit)
Non of the cells become warm when charging or discharged.

When should I do a new mAh test on the KeepPowers? ( though at 100mR)

When used in serie I always put cells in the Liitokala and only use cells with same brand and mR even if the mAh is little (max 25) off, so if I have 4 Panasonic’s
1 3345mAh 61mR
2 3345mAh 70mR
3 3296mAh 65mR
4 3327mAh 61mR
I use cell 1 and 4 in series.
Not 1 and 2. Since mR is easy to measure and mAh takes a lot of time)
Is this good practice?

At what value is a cell considered written off?

snakebite
snakebite's picture
Offline
Last seen: 3 weeks 6 days ago
Joined: 11/20/2013 - 20:21
Posts: 2054
Location: dayton oh

the repeatability of the Liitokala test is poor.
anything over 150mohm is poor.decent laptop cells average 100.
i would keep the originally paired cells in service as they have in the past.its normal to see i.r rise with cycle count.1 and 4 are decently matched.

Pages