PB, request for code review and merge back(2)

PhracturedBlue wrote: Sorry, I should have said duty-cycle, not pulsewidth. All documentation I can find says that max-volume occurs at 50% duty-cycle, and reduces as the duty-cycle is reduced (or increased).

Maybe it will work if add resistor between FET and piezo speaker. Piezo speaker is really a capacitor that charged and discharged by FET and its internal diode.

RC (time constant) = T / 4

R = T / (4 * C)

where:
R - added resistor (Ohm)
C - capacitance of piezo speaker (F)
T - average period of sound impulses (sec)

But it requires the hand work.

suvsuv wrote: PB, My yesterday's codes has a bug causing volume control not work。 I got it fixed and verify that volume control is working in my devo10(still need to use ej189's sound.ini to achieve louder volume), please have it verified in your devo8.

The vibrating motor feature is also checked in, it works in my devo10 as well. It seems all features are there in the devo10 now.

Sounds good. I'll look everything over and try to get it synced up sometime today.

So here are some buzzer findings:

The connection is actually:
5V->buzzer->17ohmR->NPN (8505)->GND

I hooked up my scope, and see that the time-constant is around 50us. This is well below the duty-cycle we're currently using, which explains why there is no volume control, but also indicates we're already at maximum volume. So the only way to get additional volume will likely be to get closer to the resonant-frequency.

I'll try to use a pulse-width nearer to the time-constant, and see if I can get any volume control.

Suvsuv,
I think I've merged everything from your tree.
I've removed all the quickpage key-config stuff, so you can sync that stuff from my repo.

I made several changes to how the vibration/music code works, but it all seems to work fine for me on both devo8 and devo10

Volume control now works, but it sounds tinny at low volume. I don't think I can do anything about that.

If you think I missed anything, let me know.

PhracturedBlue wrote: So here are some buzzer findings:

The connection is actually:
5V->buzzer->17ohmR->NPN (8505)->GND

I hooked up my scope, and see that the time-constant is around 50us. This is well below the duty-cycle we're currently using, which explains why there is no volume control, but also indicates we're already at maximum volume. So the only way to get additional volume will likely be to get closer to the resonant-frequency.

Is it a piezo speaker? At that case the piezo speaker (capacitor) charged by NPN transistor (BJT), but is not discharged by anithing (only by leakage current). No change in voltage = no sound out of the peizo. Maybe adding any resistor (500ohm) at parallel to piezo speaker will increase the volume?

+5V->--buzzer-->17ohmR->NPN (8505)->GND
+5V->-resistor-+

The code won't compile, we have run out of ROM:

FDR wrote: The code won't compile, we have run out of ROM:

+ Building 'devo8.elf'
c:/mingw/stm32/bin/../lib/gcc/arm-none-eabi/4.7.1/../../../../arm-none-eabi/bin/
ld.exe: devo8.elf section `.data' will not fit in region `rom'
c:/mingw/stm32/bin/../lib/gcc/arm-none-eabi/4.7.1/../../../../arm-none-eabi/bin/
ld.exe: region `rom' overflowed by 544 bytes
collect2.exe: error: ld returned 1 exit status
make: *** [devo8.elf] Error 1

Yeah, it was bound to happen.
Compile with TYPE=prd

sbstnp wrote:

suvsuv wrote: Some minor update:
3) Change timers' hotkeys to R/L , because the timers are easily false triggered or reset with EXT/Up assigned

I think timers should be reset with either:
1. Long-press down
2. Long press EXT

R/L are not good options because we basically can do 2 sets of quickkeys:

1. Long R/L
2. Short R/L - quick use, I would probably restrict these to one page bound to R and one bound to L, with any key reverting to main page.

We've had lot sof discussion about the quickkeys. For this release, they will stay as long up/down on the dveo10 (and short R/L on the Devo8).
I'll work on making them configurable in the future. When that happens, I'll actually make all main-page keys configurable so you can set them how you like. For now though, they won't be.

We can move the watchdog into the interrupt handler, or increase the timeout. I didn't want to do the 1st because it would defeat the purpose to some extent