I bought one of these 3 weeks ago and tested for over a week using 4 different volt metres and a mains power meter. I sent it back as to overcharged about 25 % of the time.
I like the features of this charger so much I tried very hard to give it every chance to show that it worked safely and due to the timescales of charging and discharging lithium batteries this has taken some time. It have repeatedly discharged and charged 4 types of battery, all in good condition (5 X Panasonic CGR186559CH 2150 mAh, 6 X Sony US18650 VTC4 2100 mAh, 2 X Panasonic NCR18650B 3400 mAh & 2 X Samsaung ICR19650 28F SDI 2800 mAh) at 1 Amp, and a couple of times with the VTC4 & NCR18650B at 2 Amp.
With the Samsaung: channel 2 failed to stop charging the first time, I took it out when the charger had been on 4.2 V for 2 hours and it was at 4.53 V, and the other battery in channel 3 charged to 4.268 V. Both these batteries were charged above the safety limit of 4.25 V.
Charging 4 batteries at the same time at 1 A produced mixed results sometimes with one or two batteries under the safety limit, but only just and I measured 4.26 V a lot. I did this test 5 times, it can take two days to discharge the batteries and find a 3 hour time slot to monitor the charger.
The charge status screen is also quite misleading: the statuses of each battery can vary even when they all start at the same charge level (within a percent) and finish charging within a minute of each other. Looking at the power drawn from the mains I suspect that this screen is independent form the charging circuits, which gave me false hope for a while.
I also questioned the accuracy of my multimeter, and took additional readings with a Tank O’meter that were much more encouraging. This left me with more uncertainty so I purchased 2 more volt meters (this took a few extra days) and did a statistical analysis on 4 meters to see if my original meter was in error and the result was it was probably not. It as agreed with two new meters rated a 1% accuracy and a differences were a mere 0.0606% and 0.0098% (between 4 and 5 V). The maximum error in favour of the charger was over –reading by 0.018 V or 0.421% (ignoring that 3 meters within 0.0606% and taking too much account of the one that was 1.4% under-reading) still showed this charger overcharges batteries.