[Methods] Charger Tests Version 2.0


I've decided to write up a short post describing my methods for performing certain tests. This will allow me to link to this in my reviews and be more transparent about how I do things.
Today, I'm going to go through how I test a charger to graph the charge voltage and current.


This is the most important section of this post - I am not an electrical expert. What I'm showing here has plenty of flaws I am aware of, and may have some that I have not considered. This is not meant as a guide or a description of the proper way to perform a test. This is only a write up of how I currently test a charger with the equipment I have.

Equipment Used:

I use two Uni-T UT61C multimeters connected to the cell - one for current measurements, and the other for voltage. When the charger is powered via USB, I use the ZKE EDB-USB+ to monitor the voltage and current at the power supply.
The only other equipment I used are some double ended stackable banana leads, some tinfoil (or copper foil), some paper, and a few alligator clips.
Finally, I also use a cheap old webcam I found in a drawer.



I cut one of the double-ended leads in half and stripped some of the insulation off the end. That's all the customization that had to go into this.


In use, the negative voltage lead is measured via a small piece of foil in between the negative end of the cell and the negative charge contact.
At the positive end, two pieces of foil are separated with an insulator (a sticky note, in this case). These go to the positive and negative terminals of the current measurement DMM. Another lead is used to connect the positive side of the current measurement to the positive terminal on the voltage DMM, so that the reading accurately reflects what the charger is seeing.
Finally, the USB meter is connected between the power supply and the charger.


While Testing:

I use UltraDMM to record the information from the two DMMs over serial port. I think this software is actually deprecated, and Sigrok is probably a better option - but I find this to be easier to set up. These record the voltage of the cell and the current flowing into or out of it.
I use the included software to record measurements from the USB meter. This works well, but a major caveat is that the output is stored in relative time measurements instead of with a timestamp, so some note taking is required to make sure the measurements line up.
The webcam is used in conjunction with Chronolapse, set to take an image at intervals. Using this, I can see when the charger indicates that charging is terminated, or when the screen shows some other information I'd like to note.


Output from everything above includes:

  • An array of voltage and current measurements from the two DMMs
  • An array of voltage and current measurements from the USB meter
  • A manual timestamp for when the USB meter started recording
  • Webcam captures, which let me determine the time charging started and the time termination was indicated
I use a script in python to plot all the data, and export as an image for my reviews.


The biggest disadvantage with this setup is the resistance added to the cell for the current measurement. This can lead to a voltage drop over the DMM, especially when measuring higher currents. This also means that chargers that automatically choose a charge current based on the internal resistance of the cell may choose a lower rate.
The manual nature of recording the times is a bit labor intensive - this shouldn't impact the accuracy, but it's annoying.

Possible Upgrades:

The biggest upgrades at this point are in software. If I can dedicate some time to getting sigrok setup, all the measurements will happen in one file and there will be a bit less data manipulation needed. Likewise, a script could be used to automatically determine when the status LED changes on the charger, so I wouldn't have to manually search for those times. A bit of smart scripting could also improve the sampling rate of the camera, so it takes more photos at the beginning of a charge and when the cell is getting close to fully charged, but I don't do enough of these tests for that to be worthwhile at this time.
The pesky voltage drop over the current sensing multimeter could be solved by building a current shunt and measuring the voltage drop. This is what HKJ does, and it's definitely a better method.
Finally, adding another DMM and setting it up to measure the temperature of the cell would be a great data point to add to the plots, especially for NiMH charging.