Lightsaber: an LED controller that Disney might sue me over

Hey CD,
I’m working on an addressable LED controller and wanted your feedback on the design/features I have included so far. Lightsaber will be a lot like the REV Blinkin, except it will take cues through CAN bus and the animations will be customizable through a desktop application. I’ve also included a microphone for sound-reactive animations, but i have yet to decide how to implement this in software. (this is all theoretical and very subject to change)

So, what are your thoughts?

7 Likes

Title seems a little clickbaity with the “probably going to get sued over”.

Are you supporting just one string of LED? You have the ports, can you support more?

Power distribution for all the LED?

And I know dark themes are the rage, but it’s hard to read the schematic.

1 Like

haha, I just figured that Disney might not like me calling it Lightsaber. I was thinking that only one continuous string would be needed, and in the “animation maker” you could define zones for what each section of the chain is, but the STM32 has enough timers for me to support more. I’ll definitely consider it.

As for power, the 5v regulator i’m using is rated for something ridiculous like 5 or 6 amps, which should be plenty for any string. My main concern for power delivery is actually more the length of the strip than the current.

1 Like

A single meter of 144LED/m neopixels uses 7A. I think you could exceed 5A if you do something excessive like this: https://i.imgur.com/m1Dnx2sh.jpg . Our regulator was maybe in the 15A range on that robot. That being said, no reasonable person should use that many LEDs, or if they do they should have the ability to wire it themselves. IIRC we had double-sided power for the long strings on the robot in the picture and wired power and data separately. Maybe 22AWG for data then 18 for power.

It would be nice to have the extra ports so that the LED layout could be more of an “octopus” and less of a “snake”.

I’d retitle the thread - I was worried you were skirting copying IP from Rev or someone else in the community (a parallel to the Limelight-related “did u kno its a raspberry pi in a case?? flash your own!” stuff that pops up every few months), rather than just making fun of your name. At the very least, if you need to keep the lawsuit reference, something like “that Disney might sue me to rename”.

For those of us who work in professional Engineering, lawsuits especially about IP are not something to joke about.

Cool project though.

@jnicho15 your image link 404’s for me :frowning:

2 Likes

I agree. But I have known some students, and one electrical mentor, who would not.

3 Likes

Changed!

You mean no reasonable person should use less than that many LEDs, right?

@21brownz: As long as you’re not selling it, I doubt Disney will even notice this exists. I’d probably change the name if this goes beyond an internal team project, though.

1 Like

I’m building a lightsaber in our team colors for the upcoming competition season. Check out the opensource Proffieboard. One of these might end up on our robot to make lightsaber swinging and clashing sounds (and lighting effects) during matches.

3 Likes

If I’m understanding it correctly, the Blinkin requires no programming. Push buttons and turn screws

For one you’re using the wrong team logo :rage:

2 Likes

When building switching regulators for the load current levels you are wanting to work at, the PCB layout has to be done correctly. If not, electrical noise from the switching devices can cause mis-operation in the microcontroller. It can also interfere with the switching regulator chips, possibly leading to instantaneous catastrophic failure as one coworker found out.

The buck inductors need to be chosen correctly to avoid saturation under full load conditions.

That’s not excessive, compared to this: Team 3574's Big Bling Board running on the Beaglebone Black

Yes, I’ve done a simulation in SPICE to optimize the regulator parameters.

SPICE does not address the issues I described in my previous post.

In your simulation, did you do the steady state, load transient response, input transient response or start up behaviour?

Yes, I’ve run everything on the power side through TI’s Webench simulator as well.

Okay. Good luck with the implementation!

1 Like

Thanks!

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.