Experience Needed

[algerdes]

For many years now, we have used pixels, pixel controllers, and players. Our players have changed from full blown computers (desktop and laptop) to credit card computers (Raspberry Pi, Beagle Bones, etc.), the main part of the display has several (player/remote). The controllers have been subjected to power being turned on and off, but the players have always been on full time (always on) circuits. We do have one set of props that the controller is a RPi with a Falcon Pi Hat. This unit is subjected to the nightly power on/off cycle. We set the sequence times for this unit at 00:01 till 23:59 and load an onboard RASclock so that when power comes on, it automatically starts its sequence. When the power goes off, well - it shuts off.

I say all of this as we are now going to add 3 larger matrices that will be in an area that does not have full time power. They will be subjected to power on (usually 30 minutes before opening) and power off (usually 15 minutes after closing) along with the occasional "bounce" during opening or close. (People who can't handle the large breaker sometimes slip and it comes right back on.) In the early days there was the possibility of corrupting data cards, but to be honest we never experienced it with the PiHat on the RPi.

I've purchased 3 Kulp K8-B controllers (one for each stand-a-lone matrix) and am building enclosures and the matrix panels themselves. I am very concerned as to the K8-B and how it handles power on/power off cycles. ANY information would be helpful. In particular, I am looking for input if anyone else is using their K8-B in a power on/power off scenario and if there are any problems associated with this setup.

Thanks in advance.

Al

 

[Notenoughlights]

I wouldn't imagine that within 1-3 seconds that FPP actually has begun booting from the card yet so shutting the power off by slipping on the breaker probably won't do any harm. Could a high current rated contactor not be used instead of using a breaker as an on-off switch?

 

[merryoncherry]

I don't know, but I'll be curious to hear how it goes for you. (Last year I left the controllers on 24/7 and the electric bill was fabulously high, so I'm keen to try shutting them down when not in use this year.)

 

[Katekate]

you can set up scripts in FPP to do a clean shutdown of the system on a schedule. get that running 15mins before you cut power and you should be golden.

 

[Grozzy]

Now just because I have powered off at the switch on multiple BBB based FPP capes and an rpi3 FPP player nightly during December for 3 years in a row with no issue doesn't make it right. My luck will eventually run out Safest would be as @Katekate suggested and run a shutdown script.

 

[algerdes]

I wouldn't imagine that within 1-3 seconds that FPP actually has begun booting from the card yet so shutting the power off by slipping on the breaker probably won't do any harm. Could a high current rated contactor not be used instead of using a breaker as an on-off switch?

There is no choice here. The whole side of the park is controlled by a 200 Amp primary circuit breaker which is used to kill power to about 1/3 of the display. (AC lights, mostly.)

 

[algerdes]

you can set up scripts in FPP to do a clean shutdown of the system on a schedule. get that running 15mins before you cut power and you should be golden.

Good suggestion. If I can get the folks working each night to wait, this would work. Unfortunately some of them are sitting on the switch the moment the gates are closed.

 

[algerdes]

Now just because I have powered off at the switch on multiple BBB based FPP capes and an rpi3 FPP player nightly during December for 3 years in a row with no issue doesn't make it right. My luck will eventually run out Safest would be as @Katekate suggested and run a shutdown script.

I know what you mean. Our RPi running a PiHat has run now for a few years without any problem. I'm just looking to see if a "bigger" controller (Kulp K8-B) would have any problems that others have spotted.

May you luck continue.

 

[KiwiPhil]

if you can SSH to the Pi and just run sudo halt that will shut the thing down and drop the SSH connection, you could automate it with scripts or create a one click action in something like home assistaint or FPP Script to do it based on user input or external button somewhere

 

[Megabazz]

Hi i use a KULP 16 A -B and use scripts to close down the FFP but also use a time switch to switch power off everything,i set the timer to come on 5 mins before show time and off 3 min after show finish, this keeps things safe during the day.Have had no problems with controllers .

 

[algerdes]

Thanks for that information. I'll be finishing the build of the first of three in the next couple of days. We will look into using the shutdown script and set it to go off as we are closing the front gates. It will take the greeters at least 10 minutes to close up our "greeter's hut" and then drive down to turn off the power to that side of the park. This should work. Thanks again everyone.

 

[Dreamin]

I've never bother shuting down my Pi, and never had an issue. Just make sure you have a backup in place once things are set up just in case. I just go out once show is over and flick the power off at the power point. I do the same with my 3D printers running Octoprint on a Pi. Never had an issue there either. Worse thing is maybe the SD card might corrupt but if you have a backup it's a quick fix anyway.

 

[Dreamin]

I don't know, but I'll be curious to hear how it goes for you. (Last year I left the controllers on 24/7 and the electric bill was fabulously high, so I'm keen to try shutting them down when not in use this year.)

When you leave your controllers on your actually powering all the pixels, they just don't receive data to turn on. Also would be a fire risk with them always being powered on and uses way more power than just running a show and turning it off. I hardly noticed a blip on my power bill with a 3-4 hour nightly show for December.

 

[merryoncherry]

Correct, I convinced myself that the pixels (especially one certain type) use a lot more power during the protracted time when they are off, compared to the brief time they are on, and that this is the only possible explanation for my bill being 250% of usual.

 

[algerdes]

Thanks for the information folks.