AVR Freaks

Wow! Really didn't expect any answers or comments this quickly, thanks to all!

Boards I've got is from here (KeeYees):

https://www.amazon.ca/gp/product/B07WPCLF8Y/ref=ppx_yo_dt_b_asin_title_o05_s00?ie=UTF8&psc=1

So it is, I'd say, SparkFun pro micro copy. Exactly the same except board is recognized as Leonardo and it runs on Leonardo bootloader.

Some of the comments were about the value of time. First, have to say, that KeeYees send a replacement in three days. Mean time I've ordered original Arduino micro, and additional two clones from China that will arrive in about few months :). 

Going back to time ... it all depends. I do not make any money with AVR, long time ago played a bit with PIC, got Arduino kit for a present and it sat there few years until I've felt I need it. Then I've transferred to micro as it has ability to control keyboard and mouse.

But I see value in trying to understand why this happened as it may happen again. And to have a way to tell - take a hammer and smash it, get another one.

I think I did all that was suggested so far here, with Studio and a Programmer. Before I've got programmer, I've tried using UNO as a programmer, using AVRDUDE but the problem was always COM port.

With a programmer I've found that I had to do things in certain order: connect programmer to USB, start Studio, select Programmer, click Apply, connect target, Read target voltage (external power to target, both battery set to Raw, and 5V wall adapter to raw), then read "Device Signature". Sometimes in the process Studio loses connection to the programmer and I have to restart it (studio).

Reading signature would not be always successful. Sometimes works with default 125 kHz, sometimes had to go to low ISP clock. Sometimes device signature is wrong, sometimes is right (ox 1E9587). 

Sometimes I can erase chip, sometimes not. Sometimes can copy .hex file (bootloader + empty program, from a good board - exported from Arduino IDE). Few times I've managed to check all on the board, signature, Device information, Memories, Fuses, Lock bits. But then as soon as it is connected via USB to a system, goes back to it's crazy state. 

Sometimes board would read Fuse bit slightly different that other good board, but still usable. Ie Extended 0xF3 (isntead of 0xCB ), High 0xD8 (OK ), Low 0xFF (OK).

At this time, board looks dead. Only power LED is on. System does not see it (it would before), and cannot read anything with a programmer.

Almost like processor is dead internally (can it be dead externally :)) and flash would not hold up. Are those guys that sensitive even with a proper use? 

So unless another idea from here pops up I'll consider it is dead.

thanks again 

The 'thing' here is that I have another two working pro micro clone boards. When I test them, there are no issues at all (yet :)). True is that not all USB cables are good. I've tested five that I had around and only three were 'good' or were not making any issues. Also I've mentioned that I've tested bad board using Wall wart (used for router so I'd say is good) and using set of batteries as well.

One of the first things when I've got programmer is to copy bootloader and empty program binary. I've Used Arduino IDE, selected Leonardo and option 'export compiled binary'. IDE exports bootloader only and bootloader+sketch. I have to assume that exported hex file is good. I did not try reading from good board in Studio using programmer and copying back to bad one, but I may (asuming that Arduino export is not good).

In regards to USB I am using ... it is extension from back of PC and I know that it will not be good enough to run external USB HDD (like passport) but with low power devices (assume AVRISP is one-did not check) has no issues. But I think I've tested using USB connection without extension, direct on PC but may try doing it again. I've seen somewhere that some claim USB hub is better?

So ... now we are saying that with normal use board cannot be broken (except for maybe bootloader but that is just a flash) and some suggesting, trash it?