SparkFun Forums

[RootlessAgrarian]

This story starts, like a lot of 'em, with someone giving me something that doesn't quite work any more usually it's something simpler, like a toaster or a vacuum cleaner, but this is a bit more high-tech.

In this case it's an iAudio portable recharger -- a battery pack in a slick little case with 4 usb sockets at one end, so you can recharge your various iGizmos on the road. Nice consumer packaging. However, this one has an issue.

At the opposite end from the USB ports is a cute little 5-led bar. The r-most LED is green and seems to indicate only on/off of power to the USB ports. The other 4 are blue and they're supposed to indicate the state of charge of the pack. Unfortunately, this blue LED bar is now acting rather strange. I've got it to light up just once, when the unit was cold; but as soon as I attach the wallwart and start charging the pack, this bargraph goes dark. Moreover, when you press the "check battery state" button when disconnected from the wallwart, you get a high pitched whine instead of light. Interesting.

I'm wondering whether someone's got a "typical" led-bar voltage display circuit I could use as a reference after I oh-so-carefully winkle out the small, fragile PCBs from the tight packaging and take a look at the LED area. I'm more of an electrician than an electronics tech so this teeny fragile stuff is new territory for me. But hey, it's a learning opportunity.

I see sparkfun offers breakout boards for USB so I should be able to test the other end of it (USB power) pretty easily. I suspect the unit still works -- the user just has no way to monitor the pack voltage. Seems like it should be a simple fix, like one chip if I'm lucky... and I hate the idea of just throwing away a serviceable product because of this apparently trivial failure.

So my naive dinosaur questions are: are these LED bargraph voltage indicators a commodity, same-old circuit that everyone (except me) was familiar with a decade ago? do they all use the same controller chip, just with different resistors for different voltage ranges? Am I right in believing this could be a pretty simple fix (so long as I don't destroy the board getting it out of the box)? Sorry to display my vast ignorance among all you experts, but we all have to start somewhere eh.

[RootlessAgrarian]

CORRECTION -- the brand name of the gizmo is iSound not iAudio -- faulty short term memory

[Mee_n_Mac]

Is this it ?
http://www.isound.net/shop-by-type/port ... 0-mah.html

[RootlessAgrarian]

@Mee -- Similar package, but the one in my possession is the 16000 mAh model. DGIPAD 4544. Output 5v 2.5A max, charging input 12v 1A. Claim is 6oWh on the battery, they think it can run a 60w incandescent light bulb for an hour!

I've just been digging into it a bit and it's frustrating. The lithium polymer "pouch cell" battery is made up of (looks like) 4 units; its resting voltage is about 8.4vdc. I have no idea however whether this is what it's supposed to be, as there is absolutely zero labelling of any kind on the battery pack! I've cut away its blue shrink-wrap cover to expose the hokey-looking taped-together cluster of li-poly pouches (like mini boil-a-bags), and there is not one shred of labelling on any of 'em. No mfr, no part number, no specs. Argh. I'm guessing (from a lot of mostly pointless googling) that each pouch *might* be about 4vdc, so I'm guessing that *maybe* they are paralleled and seriesed 2x2 (in whichever order) for 8v at big mAh. So maybe 8.4vdc is a healthy just-charged voltage. But that's a WAG. How the heck is a bench tech supposed to know whether the pack is healthy, when there's no clue as to its intended voltage?

I dare not try to split the pack to look for labelling elsewhere on the pouches -- online disassembly videos for li-poly packs emphasise that bending or wrinkling the pouch will kill the cell, period. Extreme finesse (dental floss was mentioned) is needed to separate them w/o any violence. Not going there

Output is 5.25vdc or so at the USB connectors, which sounds right. The LED bargraph looks ominously custom. It's now behaving in a *new* strange way -- LEDs 4 and 2 are lighting w/o 1 and 3, which (per manual) is not supposed to happen. There's a piggyback 2-pcb assembly loaded with mini components; those sandwiched between the two pcbs are impossible to get at. I had hoped to find a standard TI LED bar controller chip in there (like LM3916 or 3914, I hope I have those part numbers right), or at least a Chinese knock off of same, but no such luck I fear.

Anyway next step is to pull the piggyback board and have a look with loupe and mirrors to see if I can read any part numbers off the componentry. I don't know why I'm still fooling around with this thing -- it says "disposable overpriced consumer cr*p" all over it really -- just stubborn I guess.

If anyone is interested I can post photos ... I checked youtube and found no documentation of disassembly for this unit -- guess it's too expensive for people to hack into it for grins and giggles. I'm still kind of shocked at the anonymity of the battery pouches; guess the mfr isn't very proud of their product

[RootlessAgrarian]

OK if anyone really is interested

http://www.flickr.com/photos/rootlessag ... 203502525/

here are 4 views of the piggyback mini PCB. It's pretty devilish if you only have my crude, large-scale bench tools to work with. I doubt I can get it apart with destroying it. There's another staggeringly anonymous component on the board -- an IC without a lick of labelling on it! Others have stencilling that can barely be read if the light is at just the correct angle

[Mee_n_Mac]

I think you're screwed wrt any repairs. I doubt it's the LEDs. Your only "out" might be if one of the battery packs has died or if the AC charging circuit has obviously blown. I'm suspicious re: the whining sound, that might indicate the charging circuitry.