How does Internal Resistance affect Performance?

With the move from analog to digital, new demands are placed on the battery. Unlike analog portable devices that draw a steady current, the digital equipment loads the battery with short, heavy current spikes. 

One of the urgent requirements of a battery for digital applications is low internal resistance. Measured in milliohms, the internal resistance is the gatekeeper that, to a large extent, determines the runtime. The lower the resistance, the less restriction the battery encounters in delivering the needed power spikes. A high mW reading can trigger an early 'low battery' indication on a seemingly good battery because the available energy cannot be delivered in the required manner and remains in the battery

Figure 1 demonstrates the voltage signature and corresponding runtime of a battery with low, medium and high internal resistance when connected to a digital load. Similar to a soft ball that easily deforms when squeezed, the voltage of a battery with high internal resistance modulates the supply voltage and leaves dips, reflecting the load pulses. These pulses push the voltage towards the end-of-discharge line, resulting in a premature cut-off. As seen in the chart, the internal resistance governs much of the runtime.

 


 
  Figure 1: Discharge curve on a pulsed load with diverse internal resistance. This chart demonstrates the runtime of 3 batteries with same capacities but different internal resistance levels.

Talk-time as a function of internal resistance 

As part of ongoing research to measure the runtime of batteries with various internal resistance levels, Cadex Electronics examined several cell phone batteries that had been in service for a while. All batteries were similar in size and generated good capacity readings when checked with a battery analyzer under a steady discharge load. The nickel-cadmium pack produced a capacity of 113%, nickel-metal-hydride checked in at 107% and the lithium-ion provided 94%. The internal resistance varied widely and measured a low 155 mOhm for nickel-cadmium, a high 778 mOhm for nickel-metal-hydride and a moderate 320 mOhm for lithium-ion. These internal resistance readings are typical of aging batteries with these chemistries.

Let's now check how the test batteries perform on a cell phone. The maximum pulse current of a GSM (Global System for Mobile Communications) cell phones is 2.5 amperes. This represents a large current from a relatively small battery of about 800 milliampere (mAh) hours. A current pulse of 2.4 amperes from an 800 mAh battery, for example, correspond to a C-rate of 3C. This is three times the current rating of the battery. Such high current pulses can only be delivered if the internal battery resistance is low.

Figures 2, 3 and 4 reveal the talk time of the three batteries under a simulated GSM current of 1C, 2C and 3C. One can see a direct relationship between the battery's internal resistance and the talk time. nickel-cadmium performed best under the circumstances and provided a talk time of 120 minutes at a 3C discharge (orange line). nickel-metal-hydride performed only at 1C (blue line) and failed at 3C. lithium-ion allowed a moderate 50 minutes talk time at 3C. 


 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2: Discharge and resulting talk-time of nickel-cadmium at 1C, 2C and 3C under the GSM load schedule. The battery tested has a capacity of 113%, the internal resistance is a low 155 mOhm.

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 3: Discharge and resulting talk-time of nickel-metal-hydride at 1C, 2C and 3C under the GSM load schedule. The battery tested has a capacity of 107%, the internal resistance is a high 778 mOhm.
Figure 4: Discharge and resulting talk-time of a lithium-ion battery at 1C, 2C and 3C under the GSM load schedule. The battery tested has a capacity of 94%, the internal resistance is 320 mOhm. 

Internal resistance as a function of state-of-charge

The internal resistance varies with the state-of-charge of the battery. The largest changes are noticeable on nickel-based batteries. In Figure 5, we observe the internal resistance of nickel-metal-hydride when empty, during charge, at full charge and after a 4-hour rest period.
The resistance levels are highest at low state-of-charge and immediately after charging. Contrary to popular belief, the best battery performance is not achieved immediately after a full charge but following a rest period of a few hours. During discharge, the internal battery resistance decreases, reaches the lowest point at half charge and starts creeping up again (dotted line). 
 
  Figure 5: Internal resistance in nickel-metal-hydride. Note the higher readings immediately after a full discharge and full charge. Resting a battery before use produces the best results. 
References: Shukla et al. 1998. Rodrigues et al. 1999.


The internal resistance of lithium-ion is fairly flat from empty to full charge. The battery decreases asymptotically from 270 mW at 0% to 250 mW at 70% state-of-charge. The largest changes occur between 0% and 30% SoC. 

The resistance of lead acid goes up with discharge. This change is caused by the decrease of the specific gravity, a depletion of the electrolyte as it becomes more watery. The resistance increase is almost linear with the decrease of the specific gravity. A rest of a few hours will partially restore the battery as the sulphate ions can replenish themselves. The resistance change between full charge and discharge is about 40%. Cold temperature increases the internal resistance on all batteries and adds about 50% between +30°C and -18°C to lead acid batteries. Figure 6 reveals the increase of the internal resistance of a gelled lead acid battery used for wheelchairs. 

 


 
Figure 6: Typical internal resistance readings of a lead acid wheelchair battery. The battery was discharged from full charge to 10.50V. The readings were taken at open circuit voltage (OCV).
Cadex battery laboratories.

Comments

On December 3, 2010 at 4:16pm
Antonio Sarasua wrote:

Hello, where can I find the Rerences: Shukla et al. 1998. Rodrigues et al. 1999?
Tahnks a lot in advance

On January 28, 2011 at 4:07pm
M.Nouman Ansari wrote:

Wonderfull Web I never saw before ,with full information Thanks

On February 25, 2011 at 9:58am
Ingo Kitzmann wrote:

Can you please provide me with a state-of-charge curve for a typical lithium-ion manganese cell, maybe discharged at 0.2C, 1C and 2C, with the voltage both measured under these discharge conditions, and as open terminal voltage.
If there is such a curve available: Is the open terminal voltage a reliable measure for the state of charge?

On June 10, 2011 at 2:28pm
Antonio Sarasua wrote:

Hello, is anybody here? please i need to know the full reference of Shukla et al. 1998. Rodrigues et al. 1999. THANKS!

On October 9, 2011 at 12:36pm
Roberlanio Melo wrote:

Gentlemen, what was the method used to measure the internal resistance of cell phone batteries?

On November 6, 2012 at 5:50am
Claudius Mudefi wrote:

hi guys, mine is not a comment but honestly i want to know if there is any advantage of a battery having a higher value of internal resistance.

On December 17, 2012 at 1:37am
gaurav wrote:

how internal impedence calculation is better than self dicharge voltage ?

On February 25, 2014 at 6:03am
Thegraj Govindasamy wrote:

Hi All

Please assist with the following related to flooded lead acid batteries;

1) Can an internal resistance test detemine state of charge of the battery.
2) Is it still necessary to do a Specific Gravity reading if the above confirms this?
3) What’s the relation between internal resistance tests and specific gravity tests.

REGARDS
THEGRAJ GOVINDASAMY

On March 23, 2014 at 10:48pm
Rob David wrote:

I am flying my 450 helicopter with a set of 6 2.2A 3 cell 11.1V 30C Lipo Battery Packs which I monitor as carefully as possible. Post every flight I measure output voltage, IR of each cell, internal temperature and then I measure these same parameters pre and post charge.

This possibly seems a bit of an overkill but I am able to carefully see if any trends are occurring in any of my battery packs and I believe I have saved my helicopter exactly for this reason as one of my pack failed as I was spooling up the motor - as expected according to my recorded parameters.

I was expecting this failure because the one pack’s IR different between the 3 cells was getting progressively further apart and progressively more.  IR for the 3 new cells started of at 4,4,1 failed at 18,15,12. Knowing perilously little about battery electronics, I saw this as a sign of pending failure and I was, possibly by coincidence, correct.

But here is the real question.
If this parameter is as important as I think it is, How much difference between the 3 cells is too much? I am currently getting concerned when the max and minimum cell are 50% of each other. Is this too much or too little or am I barking up the totally incorrect tree here?

On August 20, 2014 at 9:36pm
YeonWoo Kim wrote:

hello.

I’m interested in Battery.
I have currious that how to look for State of Charge(SOC).
Could you please let me know a mathematical formula of SOC.

Thank you.

On November 23, 2014 at 12:06pm
dave backer wrote:

well, i think that there are many other things who affect performance, and they should be taken under consideration, green card lottery .