alps64: Fix for delay time for matrix scan
Delay less than 20us can cause false key detection in some situations. With week internal pull-up takes time to charge stray capacitance of trace between ground fill(and fingers), perhaps? In particular, when testing Alps64 PCB without diodes tweezer is used to close a key and this makes trace longer, more capacitance in result.
This commit is contained in:
parent
45e45691e0
commit
ba4f28a484
1 changed files with 1 additions and 1 deletions
|
|
@ -72,7 +72,7 @@ uint8_t matrix_scan(void)
|
||||||
{
|
{
|
||||||
for (uint8_t i = 0; i < MATRIX_ROWS; i++) {
|
for (uint8_t i = 0; i < MATRIX_ROWS; i++) {
|
||||||
select_row(i);
|
select_row(i);
|
||||||
_delay_us(1); // delay for settling
|
_delay_us(30); // delay for settling
|
||||||
matrix_row_t cols = read_cols();
|
matrix_row_t cols = read_cols();
|
||||||
if (matrix_debouncing[i] != cols) {
|
if (matrix_debouncing[i] != cols) {
|
||||||
if (debouncing) {
|
if (debouncing) {
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue