Gecho2 lipo charging

[alefoot]

Trying to charge a new [i.e. not recycled  / repurposed] 603450 fitted to my diy Gecho2.
Have tried a number of different charge sources and cables, including the workhorse set which we charge our phones with, and most other lipo powered 'toys'.
Starting from discharged - i.e. Gecho ceases to function - connecting the charge cable illuminates the red 'charging' led.
After a significant period of time, the blue "charged" led lights, but the red charging led also stays lit. unit remains in this state indefinitely, certainly overnight, so many hours.
Cell output voltage at the Gecho pcb measures only 3.7v at this point [not 4.2v]. Multimeter being used is a generally credible unit [Fluke 117]. That 3.7v is of course downstream of the cell protection circuitry.
I see here


that you have the pad pairs for the leads well bridged with solder; me, maybe not so much. Is that likely the cause of my perceived problem?  Any suggestions as to what else I may have messed up? My soldering on the jumper closing blobs looks clean and sound.

[mario]

A lot less solder should be enough on those pads, as long as there is no cold junction under 

Both LEDs glowing normally means that the charging circuit does not detect the battery. I haven't observed exactly the same behaviour, as there was either no charging at all, or normal charging all the way. The combination of red LED for a while, then going to both, is strange. My unit did not charge from certain adapter + cable combination, but for example the supplied cable + laptop is working well. It will not charge when it does not have enough supply current, however it only needs slightly above 500mA and I believe most adapters / phone chargers can supply at least that much, if not couple times more. Oxidized contacts or not entirely good connection will have the same effect, as it will limit the flow of current to some degree. The cable that we include has gold plated contacts so it should last long, but it also depends on what is on the other side.

Is it the same regardless whether Gecho is running (not necessarily playing, just in idle mode), or shut down?

It might be a defective charging circuit too but then it would probably not show red light in first place, I am not entirely sure though.

The connection between board and Li-Po should be OK as it does charge at least for a while. The built in protective circuits were never a problem either, at least with those two dozen batteries that I've tested.

I hope you can figure it out, if not, and it looks like a HW problem, it might be worth trying to replace the MCP chip, I can send you a spare one, or fix the board for you, if it's worth the hassle sending it back and forth.

[alefoot]

Let's see.
Behaviour appears consistent regardless of USB power source - I've tried with three different "wall-wart" and two li-po battery backup units, all of which have been recently used to charge and or run other devices including Glo and Gecho 1.
Cable supplied with Gecho 2. exhibits the same behaviour as several other reasonable quality cables used as above.

Of the three different types of source voltage [PC USB port, Wall-wart, and battery back-up unit, I am least confident of the battery back-up, since the various ones I own have different behaviours when they detect a device draining power from their supply-out port. My guess is that it's to do with how much load they read and their circuitry is effectively black box for me.

I can confirm that both red and blue leds light when the unit is USB powered without the battery being connected.
Occasionally [maybe one time in ten] the orange 'usb' led comes on and remains illuminated when testing with a battery back-up unit. Weird.
Connecting to a live [PC] usb port results in 'data flicker' from the usb led, as would be expected.

Regardless of whether the Gecho2 is idling or turned off, the same "both red and blue illuminate" behaviour occurs.
 

[mario]

Have you tried any other Li-Po in the same board?

From my experience it is rather unusual for it to not work, when you find that "magic" combination of charger & cable. So the problem seems to be in board or battery. I've charged it from power banks without any problem. I don't even think it needs precisely 5 Volts or that much current, what matters is rather that there is no micro-resistance anywhere in the circuit.

Occasionally [maybe one time in ten] the orange 'usb' led comes on and remains illuminated when testing with a battery back-up unit. Weird.


-- looks like the USB driver thinks CPU is transmitting some data, if it happens while powered off, it's just a glitch. Those are 2 independent circuits, whatever the USB driver does should not influence charging.

[alefoot]

Had the opportunity to try a different cell - an 18650 this time. No change to reported behaviour.
examined the soldering of the leads, and reflowed the battery lead to pcb joints. They didn't get any prettier, but I'm pretty sure the joints are sound.
Me replacing the chip sounds like an opportunity for disaster Available options look like ignore behaviour, or send the board home for a visit...

[mario]

Thanks for the update. Now I'm wondering if the built-in protective circuit is the culprit. I can send you one that is tested to work well and you can add it between the board and 18650 (there is place for it at the inner side of the bottom panel). It isn't required for Li-Po as they have own protection, that's why you don't have it now, but for Li-Ion it might influence the behaviour. However I never tried charging Li-Ion without it. Might be worth a try and it's easier to solder than changing the MCP73831 chip. Of course you're welcome to send it back for replacement.