In which all the pieces finally come together...

gplus
(Robert Atkins) #1

In which all the pieces finally come together…

In the US now (Seattle) and have taken delivery of the 4x Limefuel LX130 batteries that are powering the Technocolour Dreamcoat. Got it all back together and everything is working, but it seems I’ve got a couple of loose joints somewhere which could end up causing me grief. I set it up hanging off the back of a chair and everything worked fine, but when I put it on and moved about a bit, all of a sudden it went dark. I took the batteries off and they were unresponsive (pressing their buttons wouldn’t make their charge indicator LEDs light up). But plugging them in to the wall for a second made them come back again!

I guess I may have shorted out and tripped a polyswitch in the battery and plugging them into the wall reset them? What a pain…

(Ryan Clough) #2

Oh no! You’re making me nervous about what may happen to mine in transit!

(Robert Atkins) #3

Whoah! This just in, left it on the back of the chair running the brightest pattern and it’s cut out again after about 25 minutes. Batteries quite warm to the touch. I’ll let them cool and plug them in again and see how we go.

(If it were cutting out due to over-current this would actually be good news, as I can fix it by dialling the brightness down a bit—I have a literal dial for that.)

(Steve Eisner) #4

@Robert_Atkins as you guessed, that’s usually the sign of overdraw on the batteries. It used to happen to me pretty frequently with the smaller batteries from pre-iphone5 days, they just weren’t ready to push out 2A. The limefuels are what, 5A total between the two ports?

Unfortunately, it really sucks to be caught out and away from a power plug when this happens. I used one of my jackets with the kids during Halloween once and I had to ask to go into someone’s house and plug my jacket into their wall ;(

I know you put a lot of effort into power management but my recommendation is to turn down the brightness a bit – you’ve probably already noticed that at night the jacket’s gonna be really visible even at partial brightness!

(Robert Atkins) #5

(On the other hand this would be bad, as either I’m drawing a lot more current than I think (which I doubt) or the batteries are’t capable of delivering what they’re spec’d at. Or there’s some other problem, possibly caused by connecting the output of four batteries in parallel.)

(Steve Eisner) #6

ok, your comment hit while I was writing mine. Yes I do think that it’s due to brightness but more than just that, it could be some kind of inrush current when your patterns go suddenly from mostly dark to mostly bright.

1st, never trust battery specs. :frowning: I’m not much with power calculations but I also wonder if there isn’t something you need to do when trying to balance +5 from across multiple different “smart” batteries. Dial down your patterns for a test, drop the max_current to 1-2A and run off a single pack for a bit, just to see if it happens at all. Then start to parallelize the batteries and ramp up?

(Robert Atkins) #7

Aww shit.

One of the four batteries didn’t light back up when I put it back on the charger. The sniff-test reveals The magic smoke has come out :frowning:

(Robert Atkins) #8

@Steve_Eisner I have a solution for that—I got another “lipstick” battery. If I take that with me I can plug the switched-off battery into the dead Limefuel to reset it.

(Steve Eisner) #9

in one of my older jackets I used to store ongoing data about which pattern was running in nvram. Invariably after powering back up after this kind of battery crash, I’d find it had just transitioned into my pattern with the most chance of flashing lots of white LEDs. There were other patterns that consistently had more lights on, with slow fades. So I think it’s more than just total LED power consumption, also something about sudden need for current during extreme brightening.

(Ashley M. Kirchner [Norym]) #10

@Robert_Atkins Magic smoke come out … you’re drawing more than what they can provide, heating them up, and poof … Likely an internal cell popped …

(Steve Eisner) #11

sorry to hear about the dead battery. BTW that’s another thing you need to watch out for in a parallel battery circuit, if you’re pushing the max you can get from all 4 (or 3 :frowning: ) when one battery goes off you will definitely be overdrawing from the rest (and you won’t even know it’s happened… vs a single battery where you know because your whole jacket goes dark)

(Steve Eisner) #12

Also, I’m in Seattle and have a LimeFuel if you need one immediately. (I could give this to you & order a new one via Prime)

(Robert Atkins) #13

Oh dude I would heart you forever… (sent you a message with my number)

(Steve Eisner) #14

Happy to help. It’s fully charged etc. Send me a mail (my full name all one word) at http://gmail.com and we can arrange

(Robert Atkins) #15

Gone and run some errands and am thinking about my situation, and possible solutions (with beer in hand).

Firstly, massive thank you to @Steve_Eisner for helping me out with on-Playa delivery of a replacement battery. May sexual favours from attractive specimens of his preferred gender(s) rain down upon him.

My thinking:

I have 640 LEDs for a peak theoretical draw of 640x0.6 = 38.4A

The Limefuel batteries claim 2x2.4A ports each, and I have (had!) 4 of them, for a total of 19.2A

I set FastLED to clip the power draw at 80W (80W/5V = 16A)

Therefore by the numbers, I ought to have had plenty of headroom. So, I conclude that either my numbers are wrong, my theoretical numbers are different to the actual numbers, or one of my batteries was just flat-out defective.

Informed speculation on the above invited.

However: does my course of action change in any case? Change my power cap to 60-70W or so, which is comfortably under what 3 batteries can supply by themselves, giving me some redundancy in case this happens again?

(Ashley M. Kirchner [Norym]) #16

I would question the ‘2x2.4A’ ports to be honest. Are you sure that’s what they are, or are they 2x1.2A ports for a total of 2.4A?

(Robert Atkins) #17

http://www.limefuel.com/pages/product-blast

(Click links marked “tech specs” enough times and you get a popover saying):

TOTAL OUTPUT 5V/4.2A
OUTPUT 1 5V/2.4A (MAX)
OUTPUT 2 5V/2.4A (MAX)

Well that’s interesting. 4.2A “total”, but 2.4A max per port! I had not noticed that cheat.

So: Max 16.8A over 4 Limefuels. But I still should have been under the limit of 80W.

(Ashley M. Kirchner [Norym]) #18

See, that’s why I never trust those numbers. It makes me wonder if they’re not listing “burst” outputs instead of of a constant load. See, I can tell you that the batteries I use are 20A batteries. However, that’s not true. They are 1.5Ah batteries that you can burst to 20A for a short period. Only way to find out what the batteries actually are is to rip the case open and figure out a) how many, b) individual output, c) how they’re wired together.

(Ashley M. Kirchner [Norym]) #19

I’m also noticing that ALL of them have the same total output, 5V/4.2A. Even the ones that have four ports on them. Each port is labeled as 5V/2.4A, yet the entire pack is listed as 5V/4.2A … Uh, 4x2.4A does not 4.2A make. Waaaay too fishy.

(Robert Atkins) #20

“This is a 1.5V, 1800mAh battery*”

*Every word of this sentence is a lie.