Hi Captain_Ron,

I'll look into this and get back to you.  Thanks for reporting the bug.

Bill

I am also experiencing this.  I built several of these maps a couple years ago for friends and am now going back through to update them to v3.1.  For the larger maps (120+ LEDs) they freeze in the middle of the rainbow animation upon the first startup until I login and re-run the startup script.  They then appear to function normally.  The two smaller maps (~60 LEDs) will make it through the initial rainbow animation about 2/3rds of the time, but also get stuck on occasion.  

Thank you for the information. This typically happens when the rpi goes out to the FAA for the data, and it takes longer to get the data back. In my case, the power to my neighborhood went out a few times and each time when the map rebooted, it rebooted quicker than my internet router did so it just hung at the rainbow transition.

I rewrote the routine that goes out to the internet to include a check for internet availability. If there is none, it will wait then retry. There is a print statement that will output to a monitor if one is hooked up that will say 'Internet Available' or 'Internet Not Available' as it checks. So if it hangs, you can check to see if its hanging at the internet routine.

If you feel comfortable replacing the original file with the updated one, navigate to /NeoSectional and copy the attached file to this directory. If you want to be ultra conservative, rename the original file before copying over this one.

sudo mv /NeoSectional/metar-v31.py /NeoSectional/metar-orig.py

You'll need to unzip the file first, then copy it over. Please supply us feedback on this issue. You will be 'beta' testing for us and we'll need to know if it works or not. If its working, then we'll include it in an updated image on the web site. Hope this helps. - Mark

@markyharris Thanks for the quick reply! I will give it a shot this week and let you know how it goes.  I’ll have to dust off my programming skills a bit and remember how to play with rpi files, but since I’m working from home why not? 😆 

I’ll report back my beta test results. Thanks again!

@markyharris  I was able to transfer the new metar-v31.py file over to a large map this morning but it is still getting stuck on the initial rainbow animation.  Is there something else I can try or any way that I can help diagnose where it is getting stuck?  Thanks again for all the help.  -Tyler

I plugged a monitor in and reviewed the initial startup.  It does display 'Internet Available' now with the new code so that's not my issue although I think that's a good addition.  Interestingly, it also launches into the six repeating cycles that it uses to update the LED's. Its just the LED's themselves that are stuck halfway through the rainbow sequence and never update until I login and re-run the startup script.  

Full disclosure, I am still using the old Version 1 diode method (albeit a much larger one since I had 120 LED's) to power the LED's since I had it all hardwired previously.  I'm thinking that could be the issue, but can't figure out why it would only occur during the initial startup. 🤨 I may tinker with re-wiring it to the level shifter to see if that fixes the issue.

I worked with Mark to figure out my issue. The combination of the RPI starting up and the 121 LED's doing the rainbow animation at full brightness was too much of a power draw for my single power supply and that was causing the strand of LED's to freeze in position.  One solution was to stop the rainbow animation entirely, another would be to add a second power supply, but the one I chose was to simply perform the initial rainbow animation at a lower brightness. In emails with Mark I believe the code will be changed in a future release so that the initial rainbow animation is performed at the maximum brightness that is set by the user in the GUI.

Thanks again to Mark for the numerous emails helping me troubleshoot that out. 

Hi - So I am running into the same issue above that I think the rainbow effect is drawing to many amps even with a seperate power supply. I was wondering if there is a way to edit the code to reduce the brightness on the rainbow effect as I would like to keep it. Thanks.

Yes, this will be fixed in an upcoming update. However, if you don't mind editing the script a bit you should be able to help this.

From the command line, enter;

sudo nano -c /NeoSectional/metar-v31.py

Around line 176, make the change shown here.

LED_BRIGHTNESS = bright_value #255 #starting brightness. It will be changed below.

This takes the value set in the web configuration for the Brightness and assigns it to the brightness for the Rainbow wipe. So if you reduce the brightness in the settings then there won't be as much load on the LED's at boot up. You may have to try a few different settings before you find one that works best.

Hope this helps. Let us know how it turns out. - Mark