@petrockblog

Hi,

I have the following usb devices connected
usb to 3.5mm jack converter
keyboard, mouse
xbox one controller

Though I've attempted the install without the xbox controller connected; with it connected, it appears to occupy /dev/input/js0, shifting the controlblock to js1+js2.

Kind regards

@Boum-01 Thanks! I will use this to further improve the driver.

@petrockblog said in Is Controlblock / powerblock tech support gone?:

@StuMcBill abi, I am sorry that I missed your posts. I messed up the notification settings. They should be corrected now 🤦‍♂️

No worries at all - glad everything is OK.

@CHfingertrap Hi, I will update the driver! See also https://retropie.org.uk/forum/post/299255.
Sorry for this awful delay.

@mitu Thank you very much for all your support!!!

@StuMcBill Hi, I will update the driver! See also https://retropie.org.uk/forum/post/299255.
Sorry for this awful delay.

Sorry, I have messed up my forum notification settings, but they should be good again now 🤦‍♂️
Sort for this delayed reply!

The ControlBlock driver was updated five months ago for some RPi 5 specifics, but if a new kernel version introduces changes to the device tree, the driver probably needs an update again. I will take a look at this! I will update the test script that comes with driver as well. It is not up to date anymore and not very helpful with the RPi 5.

@petrockblog said in Have I fried my Controlblock (or Pi5)?:

@StuMcBill If the Pi does not respond to a power button event, I would look for

The connection between the button and the ControlBlock: Do you see the same behavior with another button or, for example, just by using jumper wires like most simple buttons? The ControlBlock service installation: Does the ControlBlock LED stop pulsing and switch to a steady on state after booting? If not, the ControlBlock service is not running properly for some reason. The ControlBlock configuration: This is rare, but sometimes the power switch functionality is not enabled in the ControlBlock configuration in /etc/controlblockconfig.cfg.

Perhaps these items will give you a clue as to what is causing your observation?

Hi,

Thanks for the response.

I've not tried this, however, I have tried directly connected to GPIO of my Pi5 and my Pi4 - same behaviour - Controls work, but Power doesn't. I will check this I have tried toggling on and off in controlblock configuration - no change

I'll report back.

@papichulo I messed up my forum notification settings and I am sorry for this! I am replying just in case this is still an issue for you:

Maintained/latching switches for on my site with the ControlBlock rev. 3.4. Do you have any any special hardware setup? The ControlBlock switches from maintained/latching mode to momentary switch mode, if a logic low signal is observed within 2.5 seconds after pressing the switch for switching the Pi on. Do you observe the same behavior when you replace the button by simple jump wires to simulate a maintained button?

I have just updated the driver. It comes with better checks for GPIO tools now and should restore compatibility to previous RPi models.

@StuMcBill I am glad to hear that!

@Goss66 The fact that the LED is still fading in and out indicates that the communication between the RPi and the PowerBlock is not working for some reasons. It the the PowerBlock driver on the RPi that signal that the driver is loaded. In that moment the LED that is attached to the PowerBlock should change to a steady state.

Do you have other peripherals connected to the RPi? Or other device drivers that might interfere with the PowerBlock driver?

I know this is a pretty old post, but I am at least now, so that it can be of any help in the future:

The buttons should work with the PowerBlock! Also the PowerBlock already comes with a resistor on the LED signal line so that you should be able to directly connect the LED pins with the corresponding LED pins of the button.

Hi! Sorry that I missed your post. I messed up my watching settings of this thread.

It sounds as if the one PowerBlock somehow got broken, I am sorry for that!

Unfortunately, I only have the PowerBlocks already assembled. Too few people were asking for the bare ones so far. However, I am planning for a next revision that offers more variants when it comes to the connectors, e.g., no connectors, Molex connectors, etc.

I could offer to desolder the header(s) for a small allowance. Please get in contact with me if you are interested in that.

@gerb45 I did a little bit of testing with Raspberry Pi OS on a Pi Zero 2w and a PowerBlock. This should work for a RPi4. You can add the following to config.txt to emulate the driver. (This way you don't have to worry about startup scripts.)

dtoverlay=gpio-poweroff,gpiopin=17,active_low=1,inactive_delay_ms=0 dtoverlay=gpio-shutdown,gpio_pin=18,active_low=0,gpio_pull=down

It seems to work identical, except that it indicates boot finished before boot actually finished. (Can anyone confirm this please?)

I am glad to hear that!

@blacknell Hi! I have just replied to your email.

Generally, the default PowerBlock firmware does not return to the previous power state upon power loss. However, I did work on this feature some months ago and I built an updated firmware version for the PowerBlock that contains this feature: Upon switch on, the PowerBlock persists the ON state as target state. When in ON state and a power button switch off event is observed the OFF state is persisted as target state. When a power-reset event occurs the PowerBlock returns to the previously persisted target state.

If you have a programmer for the ATTiny85 microcontroller on the PowerBlock you could just download this other firmware.

I am happy to provide this firmware upon request. New batches of the PowerBlock might get this firmware per default, eventually.

@StuMcBill Glad to hear that!

Got it working without the GPIO expander,

I used GPIO19 for the status pin, which really needs a 'regular' GPIO, not an I2C pin.

Then I used a gpio= statement in config.txt to set GPIO3 (which is an I2C pin) to low on bootup. GPIO3's then used as the shutdown pin for the PowerBlock.

This arrangement works fine.

@petrockblog -- perhaps the documentation could be updated to make clear that GPIOs with hardwired pullup resistors (such as the I2C pins) cannot be used as the status pin?

Thanks,
Chris