That minor triumph out of the way, I’ve now managed to smoke another battery, despite splitting the two across separate power buses.
That is, the current configuration is two Limefuel L130x-es, with each of their own outputs wired together in parallel, but the positive rail of the power bus split in two so the positives from different batteries aren’t connected. (Does this make sense?) I had thought doing that would prevent the boost converters on the batteries’ outputs from “seeing” each other and going nuts, but apparently not? Or maybe I’m still trying to pull too much current, despite my max power output being set to 28000mW, which should be well within spec across two two-port batteries—2.1A * 4 * 5V = 42W.
I can’t think of what else I can do, besides joining the power bus back together setting the max power output to below 20W.
If the battery modules are Li-* -> USB boost converters, you oughtn’t to have too much trouble hooking the outputs in parallel. My admittedly basic knowledge set gives me the impression that with a decent capacitor on the output side, they won’t notice.
@Daniel_Tullemans , I guess I’ll try the capacitor trick next. My information about not hooking the outputs of boost converters together comes from a couple of Electronics Stackexchange threads (as does the idea that two outputs from one pack are ok to tie together, @Herb_Smith .) Without the schematic to the battery pack’s circuitry there’s no way to know for sure. Relevant threads:
Finally @Garrett_Mace , I had thought of that but decided to wing it. On average they will be even—the kinds of animations I’m running where FastLED clips my power outputs are plasmas and noise kinds of things, so relatively evenly distributed.
“Capacitor trick” not recommended by Dude At Local Electronics Shop, who thinks it just won’t help.
So I’m trying two batteries, one port each, power limited to 20W; running Noise with the black and white palette it’s clipping almost continuously but not quite. We’ll see how we go…
Some good news, for those playing at home: I got three hours free and clear no problems on the B&W palette noise, and at about that point the Teensy started resetting every 10-15 minutes; switching on to my Snake pattern (only 16 LEDs lit at a time) got me another good hour with no issues and the batteries still show a quarter charge each. There’s definitely something wrong with my first pixel; I’ll re-solder that tomorrow night and get it solid. But I’m looking good for London Decom on the weekend!
