Ask Your Question
1

Troubleshooting Bad Switch on NightFox

asked 2018-05-29 23:26:03 -0500

Kaiede gravatar image

updated 2018-05-29 23:26:33 -0500

I got two NightFox keyboards, but wasn't as much a fan of the Hako Trues as I thought I'd be. So I swapped in Hako Violets into one keyboard, which went well and I was happy with the results, so I do the second keyboard.

Unfortunately, I must have screwed something up with Switch 64 (Left Alt by default). Doesn't work anymore, but did before I swapped the switches. I did what I could with my multimeter, and this is what I was able to confirm:

  1. Continuity is correct across the two solder joints. Open normally, closes when the switch is depressed.
  2. Diode check works from the switch's "positive" side through the diode (D64) and gives me the usual 0.6V drop when the switch is depressed and open normally. So the trace between the switch and the diode seems to be good.
  3. Also did a bit of voltage probing while plugged in, but because of the nature of the multimeter (it tends to give me averages if frequencies are high enough), and the keyboard matrix, I didn't expect to find anything interesting. Almost all the switches showed a ~4.9V drop compared to +5V, while S64 showed a 0V drop. Compared to VSS, it still shows a 0V drop, while the good switches show a measurable voltage difference.

Number 3 may just be a red herring, but I'm thinking either I managed to disconnect a via by accident that disconnected the switch, or break a trace. But I'm not sure where to go next trying to track it down.

My EE is rusty, so any help pointing me in the right direction from here, or correcting my thinking, would be appreciated.

edit retag flag offensive close merge delete

1 Answer

Sort by » oldest newest most voted
1

answered 2018-05-31 00:35:07 -0500

HaaTa gravatar image
input.club

It sounds like either a bad solder joint or possibly a lifted trace. Try tracing continuity between the diode and the switch pin. As well as the other switch pin to another switch on the same trace.

Since it's only one switch the MCU is fine, and it sounds like the diode is also ok.

edit flag offensive delete publish link more

Comments

Already checked the joint (and swapped the switch once) and traced continuity for the switch/diode as a whole (See #2).

I’ll double check continuity with the rest of the column. I assume you can check continuity with the row by checking the negative side of the diodes in the row?

Kaiede gravatar imageKaiede ( 2018-05-31 13:42:59 -0500 )edit

It might still be the switch if the switch is super bouncy. Sometimes taking too long to solder a switch can deform the plastic a bit.

HaaTa gravatar imageHaaTa ( 2018-05-31 13:51:10 -0500 )edit

I've confirmed it is either the pad or trace on the positive side. Not sure how I would have lifted the trace (linger time is less than 3 seconds). The pad still looks clean after desoldering to confirm that the trace itself is also bad.

I can bodge, but is there is a better bodge point than S55?

Kaiede gravatar imageKaiede ( 2018-05-31 20:03:35 -0500 )edit

Nah, one of the switches (such as S55) is the easiest to bodge to. It might be easier to locate other switches to do the rework to using the schematic https://github.com/kiibohd/pcb/blob/m...

HaaTa gravatar imageHaaTa ( 2018-05-31 20:33:23 -0500 )edit

Sounds good. I was using the schematic already. I just blanked and forgot the switches in the column should have continuity. I did say my EE was rusty. :|

Everything is working now, just need to get some sound dampening material put in the case like the first keyboard. Thanks.

Kaiede gravatar imageKaiede ( 2018-05-31 21:06:16 -0500 )edit
Login/Signup to Answer

Question Tools

1 follower

Stats

Asked: 2018-05-29 23:26:03 -0500

Seen: 26 times

Last updated: May 31