Reapairing a broken Cooler Master Rapid-I (help needed)

Myoth · 07 Jul 2018, 12:36

So I bought this keyboard recently, I knew it was broken but since I am in vacation, I don't have a lot to do and I thought this would be a fun project to work on.

Now about the thing, and what's broken about it : I actually don't know, I just know that it fell and that it didn't work anymore after that.

So my conclusion was that some bits of solder must have been broken by the fall and that I only had to find out what and where did this happen.

Except that I didn't see anything broken, meaning that this probably isn't it. Which led me to think about something else, the adaptor "pcb to micro usb" might be broken, since when I plug it in nothing happens, which would be also what would happen if I didn't plug it in, meaning that somewhere the cable is maybe broken, and since I've tested with another cable I know that it's after the cable.

My dad being an electronician, he poked around the pcb and didn't seem to see anything broken or whatever (his conclusion was that the keyboard was not flashed (which is more than very unlikely)).

So there you have it, my diagnosis on the situatioon, now the problem is that I have yet to test the "pcb to micro usb" part of the keyboard, I plan on testing it but I don't really know how since the pins of the micro usb are much smaller than the ones on a DIN5 (god dammit, why did I have to buy such a modern keyboard ??

)

Anyway here are some pics of the pcb in case someone notices something.

I know this blob of solder seems suspicious but from what I've seen from the poorly pictured pcbs of other Rapid-I this seems to be normal, though as I'm not sure, I'd still prefer a picture of a member that confirms it.

what I think is "le coupable" :

what do you guys think ?

vvp · 07 Jul 2018, 13:50

Well at least you bought something with Cortex M3 (HT32F1755) and with available data sheet.
Hopefully it was very cheap since it is broken.

Connect it to the computer while it is open and check MCU power pins with voltmeter.
If voltages are not correct then trace the power delivery circuit.
If they are fine then check the reset pin. Typically it needs to be at HIGH so that it is not to resetting MCU all the time. Electrical details in the data sheet.
If reset is fine then find the debug pins on the PCB. They may be pulled out to something like CN3. CN3 may be even it but it is 1 pin short for a standard SWD debug header. When you find the pins then connect the PCB to a debugger and find out whether the MCU itself seems ok. You can go on and read/backup the original flash and flash it with your test programs to verify the rest.

That is the hard way. My recommendation is:

Check all the traces whether something is not visibly broken or shorten to something unexpected. If yes then fix it and try whether it works.
Check the power delivery. It is easy to do without special tools and if power circuit is wrong you probably can fix it easily).
Check the reset (still easy to fix if it is shorted to ground by accident or something similar).
Pull out keycaps and switches and throw the rest to a garbage bin. There is no point wasting your time with a crap without the circuit/pcb documentation and without firmware source code. Build your own DIY ergonomic keyboard with column stagger and thumb clusters. Your hands will thank you for it.

Myoth · 07 Jul 2018, 14:37

maartenwut wrote: Was a usb cable in the port when it fell? Btw, the blob of solder is normal.

I asked the seller, I'm waiting for a response

vvp wrote: Well at least you bought something with Cortex M3 (HT32F1755) and with available data sheet.
Hopefully it was very cheap since it is broken.

Connect it to the computer while it is open and check MCU power pins with voltmeter.
If voltages are not correct then trace the power delivery circuit.
If they are fine then check the reset pin. Typically it needs to be at HIGH so that it is not to resetting MCU all the time. Electrical details in the data sheet.
If reset is fine then find the debug pins on the PCB. They may be pulled out to something like CN3. CN3 may be even it but it is 1 pin short for a standard SWD debug header. When you find the pins then connect the PCB to a debugger and find out whether the MCU itself seems ok. You can go on and read/backup the original flash and flash it with your test programs to verify the rest.

That is the hard way. My recommendation is:

Check all the traces whether something is not visibly broken or shorten to something unexpected. If yes then fix it and try whether it works.

Check the power delivery. It is easy to do without special tools and if power circuit is wrong you probably can fix it easily).

Check the reset (still easy to fix if it is shorted to ground by accident or something similar).

Pull out keycaps and switches and throw the rest to a garbage bin. There is no point wasting your time with a crap without the circuit/pcb documentation and without firmware source code. Build your own DIY ergonomic keyboard with column stagger and thumb clusters. Your hands will thank you for it.

It was sub 30€ which is honestly more than fairr for this kind oof keyboard, I'm going to do all of that tomorrow as I've got people over right now, I shall test everything you said and will report back.