OK, I know there isn’t a real ‘sleep’ as such with ProffieOS, but this is another odd one…
Normally, once all the accent LEDs shut down after the specified Idle Off Time duration, you just need to hit the MAIN button to light the blade and bring it all back to life.
But in this install, once the Idle Off Time has happened, the saber becomes completely unresponsive with either button. The only way to make it work again is to reboot using the kill key.
Config is here, but is nothing particularly unusual, and upload was done with Serial only (no mass storage or WebUSB):
OS 7.8 Master, downloaded from Github about a week ago.
BT630 Bluetooth module, wired from LED pad 6, capacitor fitted between 3.3 volt and GND pads.
Two conventional accent LEDs, one pixel crystal chamber, two pixel accents coded as separate sub-blades, shared power pins between pixel accents.
I’ve checked voltage at the switch pads and they’re around 3.27 volts both before and after the saber becomes unresponsive.
As always, any thoughts welcomed and gratefully received.
Use 7.8, not github. I don’t believe the current github Master is ‘done’, it had some 7.9 stuff that wasn’t working as expected if memory serves.
Thanks Fernando - will give it a go when I get home from work later today.
The reason I used the Master version was because of the pulsing thing that the Prof tweaked a week or so back:
Interesting. So it sounds like you’re seeing what must surely be the same problem, it just looks a little different on your setup. Compared to yours, my flicker is more a slow pulse, perhaps 1.5 hertz, and it doesn’t stop no matter how long I leave it.
I just re-downloaded 7.8 to make doubly sure it was the correct OS, though I’m pretty sure it was. Battery is brand new and fully charged too.
I also put a stopwatch on it to time exactly how long before the pulsing starts. It took 32 seconds from boot to pulse.
Once it was pulsing with the battery power on, I tried connecting the USB to my computer without powering down the battery, and the pulsing stopped immediately. I left it for a couple of minutes, then I unplugged the USB again, and after the same delay of 32 seconds the pulsing was back.
However there is one more interesting observation - selecting Serial or Serial + Mass Storage for upload has an effect. If I upload under Serial only, the pulsing is slightly faster than if I have Serial + Mass Storage selected. Comparison video below.
[Split HD 1080p]
Though I do recall there was a 7.9 that proved to be rather a misfire, so yes, that could indeed be a factor here.
Will try the standard 7.8 and report back…
7.9 has ultra low power mode that was hastily tested before Fredrik went to Japan. Several of us found that the board won’t wake once shut down due to this, so it was held until it can be diagnosed. There’s a thread about it from about three weeks ago.
Thanks Astro. Yeh, I’d forgotten that ultra low power was the root of the 7.9 misfire.
The good news is going to the main release 7.8 has sorted it.
Thanks as always.