Costume Light Project Timed to Music

This is my first Arduino/electrical project, but I have a lot of experience making all kinds of stuff and some pretty in-depth coding work... just not using Arduino. I have done a lot of research to prepare for this, but would love some insight and direction!

THE PROJECT: I am adding lights to a dance costume. Obviously it needs to be battery operated. I plan to run individually addressable lights off of 4 pins, with each pin controlling 8-12 lights. When the lights are turned on via the switch on the battery pack, they will run through a simple loop, allowing the lights to be a part of the costume when not dancing. When a button is pressed, the lights will run through a 2 minute sequence, the length of the song, that will create the effect we want through the dance routine. I plan to place a button on the costume that can be pressed on a specific beat in the beginning of the dance so that the code will run in time with the song. When the sequence is over, it will return to the initial loop.

CAVEAT: I want to prevent any accidental button pressing while dancing from messing with the lights, but I also want a way to override the sequence in order to return to it's initial loop, ready to begin again. My plan is to use 2 buttons. Button A starts the dance routine sequence. Once started, if either button A or B are pressed, nothing happens. To exit the sequence, both buttons must be pressed at the same time. I would have no trouble coding an app with this sort of logic, so I assume this is possible and, with some diligence, I can figure it out on Arduino. Let me know what you think.

SET-UP: My primary questions right now are in regard to set up. I have attached my wiring layout, and I'd really appreciate confirmation or advice on my approach. I also have a few questions:

  1. Is one diode ok, or should I instead use 5 diodes (after the fork)? I saw a similar project using the one diode, but the 1A concerns me since 48 lights drawing up to 60 mA each needs almost 3A, right?

  2. Once the battery pack is connected to the Metro, are there any concerns about plugging in the USB to transfer data (getting power from 2 sources)? I was planning on not making the + connection to the Metro until after it’s coded and I’m ready to get it into the costume, but I will likely tweak things and download new data after it’s up and running.

  3. Regarding the buttons: I’m uncertain where the ground wires connect. Can they connect to any other ground wire at any point?

CODE: Finally, if you have some direction to point me in regarding code, I'd be very grateful! I've found some stuff, but would love links to code running in a sequence like I explained, making it easier for me to make my own adjustments. Same with the buttons.

Thanks!!

Sorry, but 4 AA batteries will not supply enough current to power all your LEDs

I should probably explain further. Here's my understanding:

It's a single strand of 50 5V LED lights, broken apart into 4 smaller sections. The controller also runs on 5V. With each battery providing 1.5V, 4 batteries would give me 6V. From my research, 3 or 4 batteries are commonly used for wearable LED projects of this size, but 4 batteries requires the diode to drop the voltage into an acceptable range.

I can draw 2100mAh from the AA battery pack.

My sample shows I'll only use 48 LEDs, but I may use less than that (10 on two strips and 6-8 on the other might be more likely). If I used all 48, though, we're looking at about 2mA per light and 25 for the controller as a constant, so roughly 125mA.

Each of the 3 RGB LED colors for each light uses 20mA if they are all set to 100% brightness. I will sometimes use white, using all 3 colors equally, but I don't expect to use full brightness. Either way, we'll call it at 60mA to run white. The other 2 colors I will use are red and blue, so only one LED on each light = 20mA max.

With this, if I used 48 lights at 100% brightness the entire time (no blinking), I could run:

12 white lights x 60mA = 720mA
36 red or blue x 20mA = 720mA

Add the 125mA constant, that's 1,565mA... converting to hours: 2100mAh/1565mA = 1.34 hours of battery life.

The final consideration is that this a show - each of the lights will represent a button on a robot costume. I will never have all of the lights on at once, and most of them will only blink or turn on periodically. I also expect to use red and blue more often than white. The initial simple loop will only use a handful of lights, so the load when not running the dance sequence should be very light. We want the costume to look cool when it's worn, but also want a wow! factor once the dancer is on stage and the music starts.

Does this change things? I realize I can't just run all the lights at full bright white, and I do need to keep an eye on how many mA I'm using at a single instance when coding the sequence.

If you still think the 4 batteries won't run it, I can wire in another 4 pack, using each to power half the lights (and one the controller), right?

2mA per light? Where on Earth did you figure such a thing?

Sorry... didn't specify: the 2mA per light is a constant simply to run the PROCESSOR for that light, regardless of whether the light is on or not, not the light itself.

You can see that after that I looked at the mA usage of each light itself, when on. The amount I added to my lighting scenerio (the 125mA) addresses the constant use from the controller and the processor in each light.

Here's my source: Overview | Battery Power for LED Pixels and Strips | Adafruit Learning System

Really hoping, with this being my first post and expressing that this is my first project, that the tone can change. I would love to have things explained if you're going to say my setup won't work, have some of my other questions addressed, and be asked more respectfully to clarify if something seems off. Thanks!

Like I said, I've done a lot of research. If there's something else I'm missing, whether is battery or other, please point me in the right direction. I here to learn, and am eager to solve problems, not scrap the project:).

Just had a look at that diagram you posted, but even though\ that it's the Fritzing style of unreadability I did spot an issue, and it's your diode. You mention about 3A of current, a diode dropping about 0.7V means a 2.1W power dissipation. That's a lot of power, you have to deal with that.

Li-ion batteries are probably a better bet than alkaline for this kind of currents. To get to 5V consider a boost or buck converter instead, much more efficient.

I'm not sure what you mean about the diagram (even though/that it's the Fritzing style of unreadability?).

I wrote that the lights would need close to 3 amps if fully lit all at once, but then showed that at most I would only use max 1.5 amps at a given moment... and even that is unlikely, and would only be for a blinking moment:). The visual I'm going for will be various lights turning on and off with the beat or in a certain location to enhance a dance move rather than upstaging the dancer by using too many lights at once.

This is often what is expressed in the articles I've read concerning LEDs... while it's good to know what the max energy needs could be for the lights your using, it's most important that energy is supplied based on the actual needs of the project, especially when it's a wearable project.

Digging in deeper, my estimate is that the initial loop would consistently use about 3-400mA, and this includes the controller and light chip. This loop will be super mellow, and could use even less than that if the brightness is reduced (sounds likely - I've read that 100% is super bright!). When running the 2 minute dance sequence, that will increase a little bit - maybe 5-600mA. The sequence would now utilize all the lights and blinking would be more frequent, but there would actually only be a few more lights lit up at a time. There are some intense parts where more lights may be on at once, but I also imagine I will have to further reduce the brightness some so that it's not too overwhelming, which may balance out the amp usage.

I read further on the diode. My understanding is that it drops the current .7V and can provide up to 1A constant current. If full voltage and ampage are used constantly, it can dissipate 3W of power, but the load I just described won't cause this.

In addition, the diode can handle more than 1A for a moment, maxing out at 30A (but would cause it to fail within milliseconds at that point). My understanding of this, which was also expressed in my research, is that I am safe going over the 1A briefly for a more intense blink. I guess this eliminates my first question above.

Yes, li-ion batteries with a boost are more effective, and the recommended option for long-term projects. But this has been one of the recommended set-ups when trying to be cost effective for short-term projects such as this one. Those batteries simply not worth the investment for a single dance at a handful of competitions before the costume is retired. Knowing I can pop new batteries in before going on stage is also critical.

I'd rather do the math and reduce my amp usage to manage the energy supply... or wire in a second battery pack so that the watt draw on each of them is less (another option discussed in wearable LEDs when needing a bit more amps or wanting a longer battery life).

Let me know if you think this sounds right or not! Thanks!

If you want to use AA batteries I would go for NiMH rechargeable batteries. They have much better ability to supply bursts of current when needed than alkalines and with the slightly lower voltage you can probably do without the diode. Modern LSD (low self discharge) rechargeables no longer lose charge just sitting around so you can charge them back at base and still carry a few charged sets with you to swap in at the last minute.

There is an up-front cost for rechargeables and a charger, but it's difficult to imagine that you will not have anything else that uses AA batteries so it's not like you will use them a few times then throw them away.

Steve

I've been frustrated with rechargeable batteries in the past, but I'll take a serious look at them again. I would love to replace our household use with rechargeables, so maybe I'll invest in a good charger and a startup package with a reliable brand. It's a great way to rationalize that I'm not adding to the cost of the costume!

So aside from the diode, it sounds like the rest of my setup is ok? I just got the last of my supplies and will build today... with a multimeter on hand!

Eneloops made in Japan are one of the best options to go with if you're looking to move over to rechargeable batteries. They're all I use for AAA and AA needs.

Eneloop are excellent batteries. I swear by them, when lithium is not an option.

Eneloops were the original LSDs and they are good but quite pricey. Lately I've been using Energisers, they're excellent too and often a little less expensive. They're both made in Japan. Of the Chinese versions GP Rechargeable and Maha Imedions are the only ones I've used enough to trust and they work as well as anything.

Steve

... of course in this global manufacturing environment we now have, the Country an item was made and the race and nationality of the person who made it have absolutely no bearing on the quality of the product, it is the quality of materials used and the manufacturing processes and procedures that denote the quality of an item.

Not entirely true.
China produces a lot of cheap crap because labour and land are cheap, so they can actually get the final price down that much.
Other countries like Japan and Taiwan have much higher overhead, and also much more a culture/mentality of quality and willingness to pay for quality.
That said, most of the cheap crap that comes out of China is of the brandless kind, where the only thing they have to compete on is price. Branded stuff costs more but the brand gives some form of quality indication/guarantee (not necessarily very high or so - just a certain level) so at least you have an idea of what you're getting and why you pay more for one item over another.

Yep, plus we all know that one way of reducing costs is not to bother with quality control. Just make the things and throw them out there. The Chinese know this too. That's why most of us only use known brands. Otherwise you risk a fair percentage of what you buy being DOA. I suppose it could be worth it if they're cheap enough but I can't be bothered...particularly with things like batteries which are likely to sort of work but have much less than advertised capacity (mAh) and lower lifetime in terms of number of charges.

Not that any of this is helping to OP to get his project going....

Steve

Ha ha... yes, I was appreciating the change of topic there! I actually saw that the Eneloop batteries had much better reviews on Amazon before checking back in, including reviews where people going through a lot of batteries at work compared a number of rechargeable brands they bought.

Sounds like a relatively quicker turnaround on the battery use/recharge cycle increases their life. We don't go through batteries here very quickly, though. We're putting batteries in new items far more than replacing dead batteries, and I think I'm buying a 20-pack of batteries about every year or more. I'm not convinced we're ideal users to get the maximum about of rechargeables, making it worth the investment...