Dead Board Troubleshooting: Help Needed

[Aogu]

So a few days ago I opened up my Leopold 750R after a minor spill (it was still operating without any issues, but I thought it'd be good to let it air out and avoid use).
Gave it an interior clean (isopropyl alcohol + swabs), let it dry for a bit.

Now when I plug it in there is no sign off life at all (nothing from the pc, no leds lighting up on the board, nada)

I've checked known good cables and usb ports. Unplugged and replugged the internal power cable multiple times.

I got out a multimeter(but don't really know what I'm doing). I've been testing various bits with the connection mode. All the keys seem fine. I think the internal micro usb power cable is fine too (I get a connection from the case connection to the back of the board connection on one of the pins).

I'm now baffled as to what/where to check next? Could it be the multicontroller? The chip is seemingly "CY7C63413-PVXC". How would I confirm that's the problem/fix it?

Thanks for reading

[yui]

do you have anywhere on the board, more particularly near the microcontroler and usb area(s), small white, green or orange-ish brown crystals? sugar holds to water very well and could have stayed behind an IC or connector and caused havoc with oxidation over time.
one thing you could start to check with the controller's datasheet (https://www.cypress.com/file/125846/download) is if you get power (VCC) and ground (VSS) (you should get 5V when plugged in) on its pins and if the D- and D+ lines are connected to the USB port, next would be the xtal connections to the quartz if there is one, if all of those check good then you controller is likely dead, or it could be one of the data pins in the USB connector.
although looking only at the datasheet it should not be impossible to retrofit a pro micro in its place, as it is a programmable controller except if Leopold is willing to sell you a programmed chip it will not be possible to replace it by an other CY7C63413, need to retrace the matrix and solder wires to the SMD pads, but seems doable. Mind you i have no experience with that board and can only make guesses based on the very little information i have, and i have never rewired a board to a pro micro just saw quite a few peoples do it on this forum.
if you need help providing photos of the pcb would help tremendously too especially if it is not mat black and we can see the traces.
hope this makes a bit of sense to you, i like helping but i am far from being to most talented fixer on this forum

[Aogu]

Hey there, thanks for the thoughts!

do you have anywhere on the board, more particularly near the microcontroler and usb area(s), small white, green or orange-ish brown crystals?

No crystals as far I can see, there is what looks like some water damage to me though?

if you get power (VCC) and ground (VSS) (you should get 5V when plugged in) on its pins and if the D- and D+ lines are connected to the USB port,

Thanks for the datasheet and info, I think I see what pins that is, gonna test it shortly

next would be the xtal connections to the quartz if there is one

The pins I think are the xtal connections from the MC don't seem to lead anywhere, so I think there isn't one.

photos of the pcb would help tremendously too especially if it is not mat black and we can see the traces.

Ah sure thing, here are some images.

General shot of the business end of the PCB:


Most in focus shot I could get of the MC + a spot of water damage (?) nearby:


More of the suspicious discolouration:


MC from above, I think with pin 1 in the top left, and Vcc in the top right:

[Aogu]

Okay, I think I really screwed the pooch now.

I plugged the board in, and put the pins of the multimeter in a few places. I tried what I thought what the main line for power in; checking the connection between the "S" pin on the microusb input and the VCC pin of the microcontroller. Immediately there was a little puff of grey smoke. The kind that was just straight away an "oh ****".

In panic I thought I try Vcc to Vss, but its too close for my multimeter pins and they just sparked when near it (which Im also sure is not helpful)...

[yui]

so yes you have a quartz, it is the metal can next to the controler, but by now you have bigger problems, what seem strange is having something burn up on usb, it is so current limited that it should not really be able to, at least if the pc is respecting spec, same for the sparks...
and for next time, there are 2 vss on the chip, one is much further away and anyway it was dead you are trying to fix it, can't really get worse
as long as your pc isn't dead that is.
so where was the smoke coming from? if it was from the controller then it is pretty much guaranteed it is dead, although any usb compliant device should support a short from dataline to vbus (D-/+ to VCC or VSS). otherwise i would guess it is a resistor or diode somewhere else that did not like that, still going to need replacement but may not even prevent the board from working depending on the failure, and is cheap and easy to find.
i forgot you said you did not know how to use a multimeter so i did not warn about anything... sorry.
-in voltage mode (V) you can probe across anything (low voltage if you have a cheap meter, or low experience, and care for your life) the meter will not short anything
-in current mode (A) if you put the meter across 2 connection it will act as a short, and measure the current going through itself
-in resistance mode (Ω) the meter will act as a power supply and will measure the current across what you are measuring, it will also act as a short in some ways, this mode should not be used on powered electronics, while the other 2 are useless without power.
as per your water damage i was trying to describe a very bad case of oxidation on a pcb, the parts at the bottom should only affect the switches if anything, in the other hand it seems like the water damage near the chip is only on the ground plane (huge copper areas) and should not affect anything but is awfully close to the D+ pin, and it that pin is too oxidized it would prevent communication with the pc.
hopefully your board is reasonably fixable. good luck

[Aogu]

so yes you have a quartz, it is the metal can next to the controler,

Ah okay I thought that was a capacitor of some sort.

as long as your pc isn't dead that is.
so where was the smoke coming from?

Pc is alive and well! The little buff of smoke seemed to come from directly under the MC though 

i forgot you said you did not know how to use a multimeter so i did not warn about anything... sorry.

Don't worry about it, as you mentioned I have to give it up for dead otherwise and its my own foolhardiness either way.

in voltage mode (V) you can probe across anything (low voltage if you have a cheap meter, or low experience, and care for your life) the meter will not short anything

I think the issue was probably that I used in the higher voltage mode then, it has 2 sockets for the red probe and I used the max 10A one rather than the max 400mA one (I didnt really realize there was a difference until reading your post just now...)

D+ pin, and it that pin is too oxidized it would prevent communication with the pc.

I think this might have been the initial problem then.

What I'm thinking I'll do now is give the are around the D+ pin a clean with isopropyl alcohol. Let that dry, then see if I get anything on the low voltage mode the multimeter. Does that sound sane?
Thanks for taking the time!

[Aogu]

There may be life!?!

So now in the correct mode on the multi meter I can confirm 5v across V & G, and V & S on the microusb connector. But also 5v exactly between the D & V pins of the the microcontroller.

So I moved on to trying to check out the quartz connection but wasnt really sure how. First I touched the connectors exposed at either end of the the quartz (got 0v), then I tried and XTAL to the to far side of the quartz.

When I did this I got a reading briefly, but also the sound of a usb device connecting and disconnecting from the PC (along with "device not recognized")
I havent managed to replicate, but I also had pushed the microusb connector a cable a bit when I did the probe. So I'm wondering if maybe theres power but one of the data lines inside the microusb connector cable is damaged

[yui]

so yes you have a quartz, it is the metal can next to the controler,

Ah okay I thought that was a capacitor of some sort.

it kinda is, just using a piece of quartz that resonate at a precise frequency

in voltage mode (V) you can probe across anything (low voltage if you have a cheap meter, or low experience, and care for your life) the meter will not short anything

I think the issue was probably that I used in the higher voltage mode then, it has 2 sockets for the red probe and I used the max 10A one rather than the max 400mA one (I didnt really realize there was a difference until reading your post just now...)

most mutlimeter on the 10A socket do not have switching and are always a short, so yeah it will cause issues

What I'm thinking I'll do now is give the are around the D+ pin a clean with isopropyl alcohol. Let that dry, then see if I get anything on the low voltage mode the multimeter. Does that sound sane?
Thanks for taking the time!

in V mode the voltage indicated is only the maximum the meter can measure, you can measure 5V in 200V mode and it will only make your reading less precise, in common electronic repair use you should extremely rarely need A mode and on cheap meters you should likely never use the 10A socket as it is not fused and sometimes they do explode before reaching the 10A.

There may be life!?!

So now in the correct mode on the multi meter I can confirm 5v across V & G, and V & S on the microusb connector. But also 5v exactly between the D & V pins of the the microcontroller.

So I moved on to trying to check out the quartz connection but wasnt really sure how. First I touched the connectors exposed at either end of the the quartz (got 0v), then I tried and XTAL to the to far side of the quartz.

When I did this I got a reading briefly, but also the sound of a usb device connecting and disconnecting from the PC (along with "device not recognized")
I havent managed to replicate, but I also had pushed the microusb connector a cable a bit when I did the probe. So I'm wondering if maybe theres power but one of the data lines inside the microusb connector cable is damaged

so the controller got 5V, it needs that to start but also need the quartz and the data lines to communicate and be detected by the pc, if the PC recognized it briefly that mean that there is a bad contact, and the quartz is pretty much guaranteed to be connected to the controller, it could also disconnect because something heats up and disconnect because of thermal expansion. if anything gets hot it is likely part of your problem, a modern keyboard is a very low power device (well without LED backlight at least) and so anything that is warm or hot is abnormal.
and yeah the theory that the micro usb connector is damaged do work also with your symptoms, you could try with your meter in ohm mode to probe the 2 data contacts in the computer side USB connector of your cable and probe the usb pins in the keyboard, if you get less than 10 ohms it is fine but if you get in the kilo or Mega ohms then your problem may be in there (i tell you to use the lead as the contacts in the USB A are much larger and so easier to reach, although if you have the same probes as me you will short on the shield, so discard any readings on the shield pins on the keyboard side, you should be able to reach the center pins from the holes on the top of the USB connector).

[Aogu]

Interesting stuff, thanks for the detail.

I had an explore of the two things you mentioned. In terms of temperature I plugged the board in left it for 5 min, then ran an infrared thermometer over it. In a room 18.2 Celsius, the coldest part of the board registered 18.4 Celsius, the warmest part 19.7 Celsius. The warmest part was around the microprocessor and the quartz, but this difference seems like a very normal amount to me. I tried again after another 5 minutes and got a low of 18.3 and a high of 19.5. So I think no problem there.

On probing the data connections of the usb cable (plugged into the board, but not into the pc of course) I got 6 ohms. I could only really do this on the pc side of the usb cable. So that also seems pretty normal.

I was thinking I might try and order a new internal micro usb cable (the one that goes from the port to the board). But if the data and power seems to be carrying is that a waste of time?

[yui]

when i said warm i meant to the touch, so 19C is plenty fine.
sadly if you have only a few ohms between the pc and controller i do not thing the problem is anywhere in there but is actually the controller, or something that i always forget to ask, if this board has special drivers, have you tried reinstalling them?

[Aogu]

No special drivers, so no go there.

I got in touch with CandyKeys (who I got the original board from) and they've very kindly offered to ship me a B stock.

So, fingers crossed that should give me an easy repair via the pcb from the second one! 

[yui]

nice, so hopefully you will soon have a working board.