Battery state of health determination with impedance measurement

So , as you said about the Z (of capacitor alone direct subtracting from the Z of the series )
if i get a negative value ? (i can take absolute one )
Not a problem??
The result of getting a negative value is because using a series capacitor whose ESR is much larger than the battery impedance means that you are trying to get a small value (battery impedance of something like .0226 ohms) by subtracting one large number (capacitor Z) from another large number (battery + capacitor Z). This is a well known numerical problem that sometimes occurs when making physical measurements.

This is why you will not get accurate results when using a series capacitor of only 1000 uF. If you can, use the 4700 uF capacitor; it should have a smaller ESR. And, of course, the very best solution is to avoid using a series capacitor at all--in other words, use the method with the external bias offset applied to the LCR meter and no capacitor.

If you cannot convince your tutor that the external bias method is better, be sure to note in your report that the accuracy limitations imposed by using a series capacitor are significant and unavoidable with the series capacitor method.

You should probably consult with your tutor as to the question about the introduction.
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
Ok i shall try tomorrow using a larger capacitor value that will help me get accurate results , i shall also mention in my documentation that the series capacitor value , led to some accuracy in the measurements and explain it correctly.

Will post the results tomorrow and let you how it goes with using another capacitor value...



Thank you
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
after using the capacitor 2200µF in series with the meter and battery , it is showing fluctuations , i mean the reading does not remain constant what can be the reason ?
I checked the capacitor before combining it with the battery , it was fine with C and ESR about 0.039 at 100Hz , but as soon as i placed it in series with the meter n battery the readings started going up down on the meter .........
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
i guess i cannot go beyond some capacitor values , as i also tried with the cap 2200µF in the combination it is still showing me the fluctuations.
 
Make sure you have the polarity correct. Try another of the same capacitor. Try a 4700 uF capacitor. Wiggle all the connections to make sure you don't have a bad connection somewhere.
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
No i checked it twice all the connections are perfectly fine............
what do you think can be the reason ?
because with the 1000micro one all was okay ....
 
Did you try my other suggestions? Try a new 2200 uF capacitor. Try a 4700 uF capacitor. Try a different 1000 uF capacitor.

I see in the picture of the display on the LCR meter that the applied voltage is 1.000 volt. Try pressing the MENU button, or read the user's manual and change the applied voltage to .500 volt. A lower applied voltage will make sure that not too much reverse voltage is applied to the capacitor.
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
what if we take Xc ? of the capacitor at each frequency ? can we use it as a equation to get the impedance of battery only ?
As in Ztotal =Xc (capacitor)+ R(resistive part of battery impedance ) ie the real part of the battery and capacitor combination ?

I am really confused .......

I calculated the method that you suggested for the real and imag part subtraction from the real and imag part of battery + capacitor
Only at 100Hz i found the real value -ve and imag +ve .rest all was okay
 

Attachments

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
Did you try my other suggestions? Try a new 2200 uF capacitor. Try a 4700 uF capacitor. Try a different 1000 uF capacitor.

I see in the picture of the display on the LCR meter that the applied voltage is 1.000 volt. Try pressing the MENU button, or read the user's manual and change the applied voltage to .500 volt. A lower applied voltage will make sure that not too much reverse voltage is applied to the capacitor.




Ya i will try changing the voltages and getting other capacitor values ......
 
At each frequency do this:

Measure Z/theta of the capacitor you are using. Calculate Rcap = Z*cos(theta) and Xcap = Z*sin(theta)
Measure Z/theta of the capacitor + battery. Calculate Rtotal = Z*cos(theta) and Xtotal = Z*sin(theta)

Calculate Z of battery like this: Rbatt = Rtotal - Rcap and Xbatt = Xtotal - Xcap

It's important that you measure the capacitor at each frequency at the same time as you measure the capacitor + battery.
 
Another possible source of unstable readings occurs to me. The electrolytic capacitors you're using could take a long time to charge to the battery voltage when the charging current has to flow through the LCR meter.

So, try this. Before you make a measurement, connect the capacitor directly to the battery for a few seconds, being careful to connect with proper polarity. That will charge up the capacitor to the same voltage as the battery voltage. Now that the capacitor is charged, you can connect the series combination of battery + capacitor to the LCR meter, again being careful of polarity.
 
The capacitor value should not change much after reducing the applied voltage to .5 volts.

Have you downloaded the user manual: http://www.gwinstek.com/en-global/Download/DownloadFile/DownloadFile/download#_#08_LCR#_#82CR-81900MK1.pdf

On page 24 and 25 there is a procedure for the open and short circuit calibration of the LCR meter. Have you done this procedure?

In the picture of your setup I see that you have the capacitor plugged into a breadboard. I think it would be wise to remove the capacitor from the breadboard and connect it directly to the Kelvin clips when you are measuring the capacitor.
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
No sir there is some issue i guess , i took a new 2200µF cap then changed the voltage supply to 0.5V
then too the readings of Cap is showing me not so good measurements ,
My battery terminal voltage was 10.20V just after few minutes of this series combination i checked it again it shows me now 9.20V
i shall try with the 1000µF cap what i used yesterday
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
Ya i did the open and short circuit test

It shows on the display PASS for both the test and while only taking capacitor i directly attatched it to the Kelvin probes , but while doing it with the battery i used the breadboard
 
No sir there is some issue i guess , i took a new 2200µF cap then changed the voltage supply to 0.5V
then too the readings of Cap is showing me not so good measurements ,
My battery terminal voltage was 10.20V just after few minutes of this series combination i checked it again it shows me now 9.20V
i shall try with the 1000µF cap what i used yesterday
A battery voltage of 10.2 volts is a nearly discharged battery. I'm not surprised that its voltage is continuing to decrease. Do you have a fully charged battery you can use for these experiments?

About the LCR meter. The measurement you get with a particular capacitor shouldn't be different with a 1.000 or .500 applied voltage. If it is different that would indicate that something is wrong with the LCR meter.

Get a low value resistor such as 10 ohms, or even lower and measure it with the LCR meter, both with 1.000 volt applied voltage and also with .500 volt applied voltage. The readings should be exactly the same to about 4 or 5 digits. This will verify that the LCR meter is working properly.
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
it is showing similar results.......i guess there is no problem with the meter , mean time i used 1000µF capacitor in series with the battery and meter and checked the results , the results are too high the impedance is in ohms upto 9
 

Attachments

it is showing similar results.......i guess there is no problem with the meter , mean time i used 1000µF capacitor in series with the battery and meter and checked the results , the results are too high the impedance is in ohms upto 9
This chart shows that you are using a battery with 10.2 volts. Such a battery is almost completely discharged, and I would expect a high impedance. What result do you get if you use a fully charged battery?
 

Thread Starter

Bhargav Jani

Joined May 20, 2016
114
i have put it to charge , unfortunately the technician has to leave so tomorrow i shall look for the results with a completely charged battery..


The stupid rules of not using the lab without supervisor
:(
 
Top