Can anybody help me identify my problem (first build)

[murermader]

Hello,

So after finally building my first keyboard, I am facing some problems, and I am not sure what could be the cause. Maybe some of you experienced the same problems, and could point me in the right direction.

Short description of all the things I did:

1. Check if PCB registers clicks (I didn't check for "correct amount" of clicks, if that is even possible)
2. Lube switches (stem, spring, front and back) with Krytox 205g0
3. Lubed, clipeed and bandaid modded the stabilizers
4. Soldered at 380C° with SN60PB40


So there are 2 weird things going on:

1) They keyboard registers clicks, without clicking the switches. Demonstration:

2) Keys get stuck on repeat, until I press another key. Demonstration:

A few guesses I have, as to what could be wrong:

1. Is it possible, that I got some lube (or dialectric grease) on the PCB, and that somehow ****s things up
2. Bad soldering? First time soldering, I think it looks fine, but maybe you can see something from this picture: https://imgur.com/a/D1RDBqO
3. Maybe I just did a sloppy job on lubing the switches, and i got some lube on the contact plate part, and now that is causing problems?

Thanks in advance for any answers!

[suicidal_orange]

Krytox is non conductive, you didn't say exactly which grease you used or where but dialectric grease usually isn't conductive either, so getting a bit of that on the PCB shouldn't be a problem.  Your soldering looks clean enough, it's not the best pic for assessing cold joints especially with the flux residue still on the board.  Short version - shiny soldering is good soldering.  Some flux can be conductive so clean it off, PCB cleaner spray or Isopropyl alcohol will do nicely.

I see purple stems and get flashbacks to the time I did a switch swap for R1 Zealios - it was not fun as strange keypresses were seen, though not as bad as this.


Have you tried shorting the pins since soldering the switches?  If the problem that keeps repeating the press is in the switch it shouldn't happen when shorting pins, if it's PCB related (damaged controller/diodes or shorts) it will still happen.

[murermader]

Thanks for the quick answer suicidal_orange.

I did use dialectric grease, so that should be fine then. Most joints look nice, but there are a some things I am not sure about: https://imgur.com/a/O4Wmow9

- Picture 1&3, you can see that the soldering is leaking to next round brass plate, but for the same switch. That shouldn't be a problem, right?

- Picture 2 is a closeup of a solder near the USB C port, where the flux is is coming dangerously close the the pins of the USB connector. Is that maybe the cause of the problem? I really didn't think of flux being conductive.

I will definetly get some isopropyl alcohol and clean off all excess flux.

[suicidal_orange]

Never mind the flux the ground pad of that USB port is dangerously close to the switch pad - from the angle the pic is taken it could even be a short.  That would definitely do bad things...

The two position switches have both/all the pads connected anyway so no worries about a little solder running over.

[murermader]

I took another picture (https://imgur.com/a/JjhC1LR) of the soldering next to the USB, and I think the solder is not directly touching the USB-C ground pad. Should be OK, if I clean off the solder.

I find it really strange, that something that could potentionally short the PCB is so close to a switch soldering pad. The PCB is the DZ60 REV3 from Kbdfans. Not very beginner friendly

Thanks for all the input, really appreciate it!

[murermader]

Ok, so after getting in touch with ZealPC about what could be wrong with my switches, I got a reply:

If the soldering iron was set too hot or placed on the pad for too long, there's a chance that the switch leaves were overheated, which caused it to shift slightly allowing the cross-point to actuate prematurely. Our general advice is to heat the pad for no longer than 2-3 seconds, and to solder at 340C with leaded solder (our recommendation is Kester 44, 63/37).

So it seems that my soldering iron was set too high, and that damaged the switches. Just thought I post an update, in case someone stumbles upon this thread in the future.