Health of the Battery

[sumiii]

Hello:
some one can help me to give information(formula/algorithm used in controller) about a battery monitoring system to measure the Battery present health
Thanks

 

[klystron]

To measure battery capacity the best method is to measure energy in and out.

 

[sumiii]

To measure battery capacity the best method is to measure energy in and out.

will you please elaborate your answer...
Thank you

 

[klystron]

You monitor the input and output battery voltage and current (input current = charge current) (output current = load current).
Assuming a mathematical model for the battery ito efficiency and the effects of temperature ; you can measure the energy that is going into the battery during charging ; and the reverse if the battery is supplying current to a load. At any stage you will know the size remaining energy in the battery

 

[dr pepper]

And also you need to take into account of run down of the battery, the capacity will reduce as the battery ages.
You can get ic's for the purpose, laptops and high end battery powered power tools use them.

 

[sumiii]

I have designed the algorithm to detect the battery current health please correct me if i am wrong or any modification required.

Let suppose my battery is 12V and 100amp battery.i have connected the DC-Ct which detect the charging/discharging of battery via +ve/-ve current

If (Battery_status =charging ) //(first charging current would be high and decreases as battery reaches to its maximum capacity)
if( Battery_Voltage =12 V) //(last max voltage notable)
if( charging_current = milliamp) //( detecting battery is near to its full capacity)
// note the value of voltages (present voltage V1 & voltage after 2 min V2 )
if (V1 = V2 ) //thats mean charging current is not increasing battery capacity}
{
Battery_Health= V1 * 100amp(Rated current) * time taken to charge the battery

 

[wizpic]

You will need to use the Pukert's factor, Because if your taking more than the 1APH out if the run time will reduce. have a look here this explains it in depth
SmartGauge Electronics - Peukert's Equation - what it means to you

Hope this helps a lillte

 

[sumiii]

Thanks to your reply but you did n't understand my problem , i am well aware of Pukert's factor. I am not concerned about the Discharging capacity .i want to know whats the present health of my battery(i.e maximum charge holding capacity which will be less from the rated capacity as the day progess)?

Thanks Again
best Regards

 

[andre_luis]

I don´t know if is that what you are looking for; but recomended charging current of lead-acid battery may not exceed 6% from it capatity in order to don´t reduce it life time.

+++

 

[sumiii]

I think i am unable to explain my problem clearly.
Lemme try one more time
I am designing a system using micro controller which can tell you about difference of Present health with health at the time of purchaisng battery

let suppose i purchased a battery of capacity (original Battery Capacity) = 5500mAh as the time progress its capacity(current Capacity) reduces to 4986mAh.

Please tell me how can i calculate current capacity of battery.
the algorithm i wrote above is right or is there need any modification

Thanks

 

[andre_luis]

hi sumii

Now I got the problem.
You want a metric to determine if battery capability discharging behaviour is according specified amount capacity...wright ?

Well, coincidentally I´m facing to the same problem.
There are a lot o UPS installed in field, but I don´t know if there are in time to replace its batteryes.

The only parameters we got are : I_bat / V_bat /Temperature.
However, like you, I don´t how to estimate it health status.

The only way sugested to me uses a impedance measurement approach :
www.inl.gov/technicalpublications/Documents/4027516.pdf

But it is not applyeable to my case where hardware is already closed.

+++

 

[BradtheRad]

Let suppose my battery is 12V and 100amp battery.i have connected the DC-Ct which detect the charging/discharging of battery via +ve/-ve current

If (Battery_status =charging ) //(first charging current would be high and decreases as battery reaches to its maximum capacity)
if( Battery_Voltage =12 V) //(last max voltage notable)
if( charging_current = milliamp) //( detecting battery is near to its full capacity)
// note the value of voltages (present voltage V1 & voltage after 2 min V2 )
if (V1 = V2 ) //thats mean charging current is not increasing battery capacity}
{
Battery_Health= V1 * 100amp(Rated current) * time taken to charge the battery

Yes this is the general idea, but a volt reading is not sufficient.

I had six years of experience with my battery bank. Several lead-acid batteries. Each was 200 A-Hrs. I purchased them new.

Generally speaking...
Newly-charged lead-acid battery reads above 13.5 V.
Full battery at rest 12.8V.
Discharged battery 11V or less.

Charging at low current, voltage rises slowly.
Charging at high current, voltage rises quickly.

There comes a point where the voltage starts leveling off. Then the batteries are nearly full.
However volt reading is a rough guide. The rate of charging can make a difference of a volt or more.

If the battery was charged at high current, it may read 15.5 V.
If it was charged at low current, it may read 13.5 V.
We cannot find out the state of charge simply by taking a volt reading.

After you stop charging, the battery V stays high for a while.
Finally it settles around 12.8 V. This is the normal stable 'full' reading. It will go down a percent or so every day.

There is a graph for voltage of lead-acid batteries as it discharges. Half-discharged may read 12.4 V.

It also depends on temperature. Voltage may rise or lower by a tenth of a volt or so.

An old or ailing battery may be unable to maintain a charge. It may read 14 V when charging ends. It may settle to a stable figure at the normal 12.8 V. But it will quickly drop if you attach a load. This is a way you detect a battery's health, by attaching a load.

My batteries declined in capacity over the years. Only by connecting a load was I able to notice that they didn't last as long as when they were new. I could tell because the voltage went down more quickly. It still took a few hours of watching volt readings.

Once the batteries are down to 10.5 V there's little use taking any more current out of them. They're discharged.

What's a simple test to find out how much current a battery can provide? Attach a very heavy load briefly. The voltage reading will tell you a lot. You'll see if the battery has high internal resistance. Nevertheless you still need to know its previous behavior in order to know how healthy it is today. You need to know how much makes a heavy load for that size battery.

--------------

Post #2 said the best way is to measure energy going in versus energy going out.

The most useful meter to do this is the Amp-Hour meter. It measures amps and it measure hours. It multiplies them.

There are smart battery chargers that will tell you a battery's capacity. It charges the battery fully, then discharges it fully. The reading tells you how many Amp-Hours the battery produces.

Then you compare the figure with the capacity printed on the battery. That way you know how healthy the battery is.

 

[andre_luis]

...The most useful meter to do this is the Amp-Hour meter. It measures amps and it measure hours. It multiplies them...Then you compare the figure with the capacity printed on the battery. That way you know how healthy the battery is.

Brad

I´m interested to know how it´s possible to deal with temperature change ?
At datasheets the capability curves assumes under a constant isotermic behaviour.

Other point : How to classify the measured Ahxt ( good / medium / bad ) ?
What are the threshold ranges wich define the quality of battery ?

+++

 

[dpj]

study below datasheet. This will give you fair idea how level of charge in the battery is find out.

www.ic72.com/pdf/download.html?id=22852

 

[BradtheRad]

I´m interested to know how it´s possible to deal with temperature change ?
At datasheets the capability curves assumes under a constant isotermic behaviour.

Yes. Temperature affects all battery readings.

There is a graph telling how much to add or subtract (depending on temperature) when reading a calibrated state-of-charge floating hydrometer. These cost several dollars however they don't give an accurate reading above or below a certain temperature.

If you charge a lead-acid battery at C/5 it will heat up noticeably. It will fool you by showing a very high state-of-charge reading. And its voltage will read high. However using a taper charge toward the end of charging will reduce the error in readings.

A battery may be ailing and not be able to hold the charge. It too will give deceptive high readings when newly charged.

Low temperature causes any battery to give under-readings even if it is healthy.
At low temperature a battery performs less well.

There are graphs for:
* voltage vs temperature
* voltage vs charge rate
* voltage vs state of charge
Etc.

However I don't remember seeing a graph combining 3 specs at once. With some you see several lines which you apply to a different spec.

Other point : How to classify the measured Ahxt ( good / medium / bad ) ?

Good/med/bad will depend on the capacity of the battery.

The amp-hour meter informs you regarding the most vital spec of a battery. Amp-hours.

It's very important to stop charging once it is full. With plain lead-acid cells, overcharging may cause bubbling and a drop in acid level.
If you overcharge gel cells it can ruin them or reduce their useful life.

That's why an amp-hour meter is invaluable. The advantage is that you can keep track how much you've taken out. And you can track how much charge you've put in.

What are good threshold points? Experience with a battery tells what level is good/medium/bad.

A simple volt reading may be sufficient once you've watched a few normal charge/discharge cycles. However watching a meter is inconvenient.

AN amp-hour meter retains the updated value and you can read it anytime. You can also reset to zero anytime.

What are the threshold ranges wich define the quality of battery ?

Thresholds are arbitrary as with most any battery.

Normally a battery loses capacity gradually. Often it develops higher internal resistance. Often the plates become porous.

Sulfation is a very common problem. There are ways to combat this.

The time to replace the battery is when the price of replacement equals the price in lost performance.

Or when operating conditions cause the battery to underperform. Example, to power a refrigerator which keeps vital medication cold.

One cut-off point is the voltage below which an inverter will not run. A typical figure is 10.5 V.

 

[andre_luis]

hi Brad


I appreciate very much the above explanation. Thanks a lot.

However it is not clear yet, how we can develop a closed procedure to evaluate discharge capability, and thus estimate the battery health considering all those paramenters. Datasheets usually presents them separatelly and is not clear how all those variables interacts toghether.
It is well known that enviroment temperature may change significantly battery discharge capacity but seems to be hard to define how to manage calculus while enviroment temperature is changing.

+++

 

[RCinFLA]

Battery health degradation is a bit different for various types of battery chemistries.

For a lead acid battery, two of the most common degradations are capacity loss due to sulfations, and positive grid corrosion causing increase in battery internal resistance. There are other things that can degrade performance but these two are most common.

For LiIon cellphone batteries it is usually battery series resistance that limits useful life. For every cycle there is some degradation is Rs of battery. After 200-300 cycles the Rs can rise by a factor of two over its initial new value. This is usually where phones start having problems with battery voltage slump during heavy current loads.

Rs test is probably the best single test that can be done but you need the history to know the relative degradation.

 

[BradtheRad]

However it is not clear yet, how we can develop a closed procedure to evaluate discharge capability, and thus estimate the battery health considering all those paramenters. Datasheets usually presents them separatelly and is not clear how all those variables interacts toghether.
It is well known that enviroment temperature may change significantly battery discharge capacity but seems to be hard to define how to manage calculus while enviroment temperature is changing.

Yes. Just having automatic temperature correction would be a big advance. Voltage could be used as a more direct indicator of state of charge.

This could be done by gluing a temp sensor against the battery case. Attach it to a computerized voltmeter module which contains the graph of voltage vs temperature in memory. Readout would be in volts corrected for temperature.

Such a setup would need calibrating of course. Not too practical for commercial marketing.

So for the time being a human must act as the module that consults the temp vs voltage graph. And factors in the other variables.

 

[wizpic]

I came acrros this web site. It may help you or give you some idea's
If you view the source code(HTML) I think the calculations are in there only had a quick look
**broken link removed**

 

[BradtheRad]

I came acrros this web site. It may help you or give you some idea's
If you view the source code(HTML) I think the calculations are in there only had a quick look
**broken link removed**

Very ingenious. It makes a good point that Watt-Hrs are a more valid way to count battery charge and discharge.

I tried figures for an automobile battery.

It needs to have internal resistance of .02 ohm in order to crank a car engine at 200A.

It shows voltage dropping to 9.5 which is common when starting a car.

But it only does it for a second. Total draw one A-Hr.