I’ve been doing some testing with two Pixelblazes running in leader/follower mode, and I’ve noticed an odd startup behavior.
The follower Pixelblaze consistently boots before the leader. Because of that, it initially starts playing the rainbow font pattern that is saved as the first pattern in its local playlist. Once the leader finishes booting, the follower switches over to the leader’s animation as expected.
However, the first synchronized animation is always offset between the two boards. When the playlist advances to the second pattern, the boards immediately become perfectly synchronized and stay in sync from that point on.
Has anyone seen this behavior before or know what might be causing it?
I’d really like to solve this because I sell these LED garments. For the first 10 seconds after power up, the outfit appears to be malfunctioning until the boards finally synchronize which obviously isn’t ideal for customers.
I’ve attached two videos:
One showing the LEDs on the follower board starting up before the leader.
Another showing the status LEDs on the follower Pixelblaze booting before the leader Pixelblaze.
Edit: I was not able to upload the videos even after compressing the MP4 file. I had to upload screenshots. Let me know if there is a different way to upload the videos.
Sorry about the videos! The forum system is picky, sometimes it’s better to upload a video to YouTube or something.
So on first boot it does take a little while for WiFi to start up and connect and sync to get locked in. You can config them to start with a pattern that makes this less obvious, but can’t really remove it entirely. If leader and follower both start to the same pattern, then when sync starts it will be less of a difference. A static color, or something random enough not to be distracting, like blink fade or something.
Once sync initially connects it starts syncing animation time bases, so there might be a jump for the first sync, should be smooth from there. Also some patterns have their own running timers/state that doesn’t use the time based sync, so those only work when the pattern switches at the same time.
It’s also possible to set the first pattern in the playlist to run for a short time, so that when the follower does connect and sync, soon after you’d get a switch to something that brings both more in sync. Has to be long enough to let WiFi and follower connect.
Thank you for the response! I will try starting with a static color in order to give the controllers time to sync.
I am curious if you have any thoughts on why the follower is booting a couple seconds before the leader. They are both connected to the same power source in the circuit so they are receiving power at the exact same time but for whatever reason the follower board is booting faster than the leader. You can see this when you see the orange light turn on on the follower a second before the leader in the video. You can also see the sync behavior in the video
One final question regarding power cycling. I’m having an issue where my Pixelblaze system power
cycles about 10–15 seconds after startup. This only seems to happen when the system has been powered off for an extended period, usually about a day. If I unplug it and reconnect it shortly afterward it starts normally and does not power-cycle.
My power setup is:
12V USB-C PD trigger
Buck converter stepping 12V down to 5V
The Pixelblaze and LEDs share the same 5V power and ground output from the buck converter
This is happening consistently on two separate builds. Each build uses two Pixelblazes controlling approximately 1,500 LEDs total.
Does anyone have an idea why the system would power-cycle once, about 10–15 seconds after a cold startup, but not when restarted shortly afterward? I’ve included a video showing the issue.
This looks like your power supply/pack is resetting, the display flashes right when the LEDs go out. Maybe overload? Not sure why it would do that cold, maybe just on the margin normally. If you turn brightness down a tad, does it still happen?
Update: I believe I found the root cause. I replaced the INIU Cougar P63 25,000mAh 100W power bank with an Anker Power Bank (20K, 87W, Built-In USB-C Cable) and the problem completely disappeared.
The issue appears to have been related to the USB-C Power Delivery (PD) negotiation. After the INIU battery had been sitting idle for an extended period, it would initially negotiate 12V with my PD trigger, but then around 10–15 seconds later it appears to renegotiate or briefly reset the PD output. That momentary interruption caused the buck converter’s 5V output to drop, which rebooted both Pixelblazes. After restart, everything would run normally.
With the Anker power bank, the 12V PD output remains stable after startup and I haven’t experienced a single power cycle. Hopefully this helps anyone else running Pixelblazes from a USB-C PD trigger and buck converter.
