How do I test a charger

When testing chargers I want to measure and record both current and voltage simultaneous for the full charge and know exactly when the charger says the battery is full. I have done this small write-up to show how I solved the above measurement task.

Recording
The measurement of voltage and current can be done with two DMM and to record the data I uses a computer connection from each meter. This was supposed to be easy enough, but the Fluke computer program could only handle one meter at a time, i.e. I had to do my own recording program.
The computer interfaces for the meters uses an encapsulated IR link, i.e. the DMM has absolutely no electrical connection to the computer or to the other DMM. The encapsulation secures that I can use multiple meters close together without interference.





Measuring current

I have shown how to do this before:



But this is a simple way to do it, and not the most precise way. A check shows that with my Fluke 289, probes and tinfoil I has a voltage drop of 160 mV when the current is 1 A, i.e. 160 mohm resistance. This is much too high for me. I could have used shorter probe leds and multiple layers of tinfoil to reduce the resistance, but I selected another solution.



A piece of double sided PCB with the wires soldered to it. This is 1.6 mm thick and can be used in most chargers, if I uses short batteries, maybe even unprotected batteries. Used together with my Fluke 289 I get the voltage drop reduced to 39 mV at 1 A, i.e. 39 mohm resistance and this will be the same each time I use it. This is good, but I can make it even better.



Placing the series resistance on the PCB, removes any resistance from wires. With 10 resistors of 0.1 ohm 1% I have 10 mohm (Measurements says 10.6, the copper adds some). The meter will not affect the resistance or voltage drop. A difference of less than 11 mV at 1 ampere is small enough to be ignored.
There is one disadvantage with this small resistance, it is difficult to read small currents. I have to accept a tolerance of about ±1 mA. At 1 ampere the measurement with this adapter and measured directly with a DMM matches within 1% (Assuming 10 mV/A).



Measuring voltage

This is no problem, a piece of tinfoil at the other end of the battery and the voltmeter connected between the current adapter from above and the tinfoil. I connected the meter to the charges side of the current adapter, i.e. the meter will show the output voltage of the charger, not the battery voltage, but the difference will be very small.



When does the charger say "battery full"

Charging a battery takes hours and I did not want to sit looking for when the lamp on the charger changed from red to green to say "battery full". With some chargers it is easy enough to see on the recorded data, because they turn the current off, but not all chargers does this!
The solution was a video camera on time lapse setting, taking a picture each second. Instead of using the video time, I placed a clock besides my meters, i.e. the video records the charger lamp, both meters and the clock



Putting everything together



The video is on a tripod to the left, looking down on everything.



Examples

The finally result looks this way:





The red line is the voltage and on both examples the voltage drops when the current is turned off and this also matched when the video recording showed that the charger reported that the battery was full.
The green line is the current, the scale on changes with batteries and chargers. The WF-139 charger turns the current off at regular intervals to check the battery voltage, this is too fast for the meter to record a zero current, and only a drop in current is shown.
The blue line is calculated as a summation of the current.
The yellow line is used to mark where the charger says "battery full".



Notes

I have ignored one resistance in the above explanation, the contact resistance. Each connection will have a small resistance and by adding the adapter and the tinfoil I have added two extra connections.

For testing I am using older batteries, that have lost some of their capacity, this is very obvious with the IMR16340-550.