Postage-Board
Postage-Board copied to clipboard
LifeIsOnTheWire
Grounding the keyboard via the board causes resetting.
Hi, I have built a steel case keyboard using Postage Board. However, I had problem with electrical shocks so I connected steel casing to the Postage Board's GND. That solved the problem with electrical shock damaging the Postage Board, but now each time I get shocked board gets restarted, I am guessing due to the reset using GND which gets put on very high voltage for a very short time.
Any advice how to approach this? Can I somehow configure reset to work differently, maybe in such way that reset has to be shorted for longer time?
Hi Martinsos, can you elaborate on the electric shock problem? What was conducting the shocks? The keyboard case?
Unfortunately there isn't a way to adjust the behaviour of the reset function, it is just how the Atmega chip works.
Hi, well yes shocks happen in contact with the keyboard case, which is steel in my case. I reached out to guys on geekhack and learned that it is common, and that I should ground the keyboard. I grounded the steel case to the ground on the postage board (I checked before where is ground on postage board), which should ensure electricity goes through the ground, not through the chip (and damage it), but now keyboard resets on any static shock. Therefore I assumed that voltage on ground is interpreted as reset.
Hm so no way to change it? I understand this is out of scope of postage board then, but do you have any other suggestion how to handle this situation? Thanks!
On Tue, Jul 2, 2019 at 1:48 PM LifeIsOnTheWire [email protected] wrote:
Unfortunately there isn't a way to adjust the behaviour of the reset function, it is just how the Atmega chip works.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/LifeIsOnTheWire/Postage-Board/issues/3?email_source=notifications&email_token=AALXFB3WA453LGXXFBVC2JTP5M6ATA5CNFSM4HJFX4JKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZBACIA#issuecomment-507642144, or mute the thread https://github.com/notifications/unsubscribe-auth/AALXFB67KW44SN6NLU4GGOLP5M6ATANCNFSM4HJFX4JA .
Hmm, I've built several steel-plate style keyboards, and never had any static shock problems. I do static-shock the keyboard itself all the time, but I suppose my static-shock isn't arcing onto the PCB.
If I had that problem, I would troubleshoot it by finding what point of the PCB is nearest to the steel plate, and try to insulate that area, to prevent the static shock from arcing from the steel to the PCB. I think I would probably just cut strips of electrical tape, and cover the PCB in tape, because I'm an overkill kind of person.
Lucky you! I wonder why is it arcing onto the PCB hm. Maybe you have your case grounded? Correct me if I am wrong, but even if I additionally insulated the PCB from the case to prevent static shock arcing, it can easily arc to any of the wires (since keyboard is handwired, and while wires are mostly insulated, there is always some copper standing out somewhere) and therefore enter the PCB through the wire, right? So I would really need to isolate all wires and the PCB from the case.
Anyway, thanks for the help, I understand this is not your concern any more. I was hoping that there would be a way to avoid shock to GND from triggering reset, but if there is none then I will have to look for other options. Thanks!
On Sat, Jul 6, 2019 at 3:45 PM LifeIsOnTheWire [email protected] wrote:
Hmm, I've built several steel-plate style keyboards, and never had any static shock problems. I do static-shock the keyboard itself all the time, but I suppose my static-shock isn't arcing onto the PCB.
If I had that problem, I would troubleshoot it by finding what point of the PCB is nearest to the steel plate, and try to insulate that area, to prevent the static shock from arcing from the steel to the PCB. I think I would probably just cut strips of electrical tape, and cover the PCB in tape, because I'm an overkill kind of person.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/LifeIsOnTheWire/Postage-Board/issues/3?email_source=notifications&email_token=AALXFB7YFNOV3YUPHTF3AHDP6COYXA5CNFSM4HJFX4JKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZKZ5ZA#issuecomment-508927716, or mute the thread https://github.com/notifications/unsubscribe-auth/AALXFB5CE5HYPBGRSLZKEXLP6COYXANCNFSM4HJFX4JA .
Yeah good point, all the wires will still conduct the shock.
The actual trigger for the Atmega chip resetting is the act of shorting the Reset pin to the GND. So if it is causing a reset, I'm inclined to think that the shock must be jumping between the two pins designated for Reset. One of those pins is connected to the GND plane, and the other runs to the reset pin on the Atmega chip.
Maybe if you insulate both of the reset pinouts the shock wouldn't be able to arc between those pinouts?
Thanks! I will search more about this problem regarding Atmega, I hope I will find something. Feel free to close the issue, I will post later if/when I find out smth!