GreenLED's right. The only way to safely match cells is to determine the charge or capacity (mAh) of the cells, which is current flow for a given time. Voltage is a different attribute of electricity. While the cheap battery meters that use voltage may given you an indication of a single cell's state, it may not because it's measuring the wrong thing. For that reason, they simply can't be used to match cells, which is what being safe requires. It's kind of like a river's flow. Current is like knowing how much total water is flowing past a point for a given amount of time, while voltage simply tells you how deep & wide the river is. Say we have 2 rivers that are the same width. Well, one could be a shallow mountain river that has far more total flow than a deep but slow river because the shallow mountain river is flowing faster. In other words, the deeper river has the potential for more flow, but if the water is moving slowly enough, the faster and shallower river can easily have more actual flow. You can have a cell with almost no charge left that still has an apparently good voltage. If you put that cell in a light with a cell that has the same voltage but a lot more charge, the low charge cell will be "used up" quicker than the good cell. When that happens, the low charge cell will start absorbing electricity from the good cell. Then two things can start to happen -a cell that was not designed to be re-charged is re-charged, and its polarity may/will be reversed by that re-charging. That's when bad things happen. To re-use the river analogy, having mismatched cells (unequal charge remaining but similar voltages) is like having 2 rivers of the same depth and width, with one having a faster current that is running into a one with a slower current. Where's all the extra water from the faster current going to go? Out of the banks and you have a flood. The same thing happens with mismatched cells. What we want is for the current to flow smoothly and that requires similar currents. ZTS tests using a pulse load method for testing. That's the most efficient way to determine remaining capacity with any useful accuracy. Yeah, it's a bit of money. If you want to be safe, though, it's really the only way. ZTS isn't the only such company making testers, but they've proven to be reliable. Plus, when you think about it, the piece of mind they give you is worth a lot more than the testers cost, IMO. Hope this helps!