That doesn't mean it isn't isolated.
USB 0V goes to BAT 0V
USB +5V is passing a Si-Diode and one transistor/regulator, then going to BAT +3V
USB D- and USB D+ could be traced nowhere and look unsoldered
USB Shield reads 18kOhm against DMM GND and COM input
USB Shield reads 10MOhm against V and A input
So, if I would be mad and hook it up against the live wire on COM, there would be a chance to trip the RCD by touching the USB port :-)
Is it really for charging? I don't see why these meters would need to be rechargeable.
With two ok-ish NiMH cells (2.67V combined):
Rotary in OFF position, USB connected, I read a charging current of 2.8mA on the batteries.
Rotary in V position, USB disconnected, I read a current draw of 1.6mA on the batteries.
Rotary in V position, USB connected, I read a charging current of 1.2mA on the batteries.
It even works without batteries only powered via USB.
On an empty pack (2V combined) it provides a charging current of about 200mA.
It tried to charge empty alkaline batteries as well. :-)
No rechargeable batteries there. I'm guessing it's more likely to be a comms port.
I hoped so somehow, might be interesting to play around with. But it's a standard chinese suizide charging port.
Such an interesting device... Removed branding, standard SMD DMM Chip, no COB... Chip has it's labeling removed obviously via laser or CNC cutting, very precise.
What an unusual thing.
My guess, looks like someone designed it to make use of existing AN8001/8002 production line equipment with a new PCB, but then the customer refused it, so they dumped the series 0 production on the marked or something. Standard SMD DMM-Chip with high-effort removal of the labeling seems very unusual for this kind of a garbage device.
PCB has on the back the label
"WM16A/B/C
2019-OCT-21
FR4 REV:00"
Also the accuracy of the readings in V and Ohms is interestingly extremely good compared with the BM789 and the MM12, but I don't know if this is normal for ANENG or whoever makes these things.