I bought days ago this cheap external battery charger (I have a 5800xm, BL-5J battery):



I think they are all the same, the price was right, 1.99 Gbp (inc shipping), and it is the only thing which is right with this unit.

The quality is next to nothing:





The screw was already broken, both two wire fell off as I opened the unit, the plastic assembly feels (is) very cheap.

So, before letting it torture my batteries in a unspeakable way, I was wondering about it's regulation circuit.. so I opened it and took a couple of photos:




I know that "reverse engineering" is not good, in some places this leads to bad things.. but with this cheap object, I think I can still sleep thigh in the night, so I tryed to trace back it's circuit:




And produced this schema!



http://img641.imageshack.us/img641/8428/circuits.png

(I inserted the full link to the image of the circuit, handy).

Now.. Much work is still to be done, currently I have a terrible split-by-two headache, so I'm going to link the last file, is the archive with everything inside:

_ images
_ models
_ datasheets
_ circuit file (LTspice)

Link --> http://www.megaupload.com/?d=7EYM531E <--

The circuit actually is simuable, it tends to a voltage of 4.4 circa. I do not know if it is right (the simulation!), I measured with a multimeter the output (before wires broke up) and it was around 4.2 - must check later.

First thing to do is:

1. Verify traces and plotted circuit, the two images (PNGs - in the above archive) are overlappable, one of them is mirrored so it is easyer to check traces; can someone help me veryfing them? reports here all the suggestions, rights and wrongs.

Then:

2. Find reliable substitutes for the original transistors or reliable models.

3. Find correct component value I'm 90% they are correct, there are some normal 4 band resistors, a couple of 5 bands resistors and one ceramic cap which reads "104" -->180pF?

The very-very-very last step will be to... err.. improve the design

Now, I must really go to sleep, aber I'll pass out

I was inspired in making this post by a thread on replacements batteries BL-5J where battery capacity was verified - playing around this charger could be usefull to many people around here

See you later!

I have the same style of unit but it has the micro usb connector - never tried it but I'm gona hook it up and see if the thing actually works. The ceramic cap value= 0.1uF

UPDATE I'm finding some errors in the circuit tracing, I'm working on the new schema, meanwhile I'm posting a new image of the back of the PCB without the pen marking.



Full size --> http://img845.imageshack.us/img845/3162/31052011123.jpg

I found I included a wrong datasheet in the above archive on megaupload, the right one for transistor S8050 is:

http://pdf1.alldatasheet.com/datashe...RON/S8050.html

As soon as I have some satisfactory result I'll post a new archive

All right, here are the latest news:

The latest circuit is this:



Full size: http://img38.imageshack.us/img38/1665/circuit11.png

I'm 90% confident it's correct and without big mistakes, even if I wish.. why?

All right, simulating this we get some interesting results:



Open this in a new window: http://img696.imageshack.us/img696/94/diag.png

As you can see from the graph, the upper pane, the current going into the RED led cut off when battery voltage reach 4.2, which is acceptable; so, when we put the battery in the charger, RED led turns on. When battery goes up to 4.2V the led turn OFF. Nice, uh? battery is full

What is not so *nice* is the remaining current flowing into the battery past 4.2V: (bottom pane) there's still 64mA going! - at 4.3V battery voltage, mAmps are only 10 left.. but the charger is still pushig the voltage up to 4.4 circa.



So, I'm in a good mood, and I want to see some fireworks happening! here's the setup



I'll let you know, if I don't get too distracted at the computer, I'll keep a watchfull eye on that multimeter reading

Ka-boom?

P.S. The battery, don't get confused by that hologram.. for 2.88$ I got it (shipping included) from eBay.. - so it's perfectly expendable.

Yep, mine got both, a variation of a mini usb + the cable in the picture.

Are you sure about the cap value? where you referring to yours or mine?

Yours mate - I've not opened mine (yet).

The capacitance value is often marked using a 3 digit code. This works in the same way as resistor coding but using numbers instead of colours. The first 2 numbers give the value and the last number is the multiplier. These give the value in Picofarads (pF), e.g. code 103 = 1 0 000pF (=0.01uF). Alternatively the value may be marked directly, for example 2n2 is 2.2 Nanofarads (nF).

Ops, you are perfectly right! (http://en.wikipedia.org/wiki/Electronic_color_code) - when I googled I wrongly took the first page that showed that strange value "180pF"..

I'm finding an incredible number of errors in my diagram when I traced the circuit!! I must have been drunk in those last 3 days 'couse otherwise I should change hobby

Moreover, I found in my charger the red LED indicating the charge was going on and off every 5-10 minutes.. very odd. I should have found the cause: apart the bad solder job made on this board, one of the wires going to the battery terminals is burnt inside, the insulator is transparent, and is possible to see a black area....

I'll fix that, and other issue; meanwhile I'm correcting the diagram of the circuit inside LTspice and publish new finding - for now just assume the material linked is.. "terribly" wrong

Howdy!

New update: revised and corrected schema, 98% sure it's right.. for now



Full image

What I still don't like, maybe I'm wrong, is the direct +5V connection to the emitter pin of the PNP transistor.. more than 80mA will flow out of it's base..

Then, removed all those cheap-burnt wires and put some more serious ones.

Then applyed the charger to a test: a nice 1F@5.5V capacitor! the perfect battery replacement



And the charge was done in 5-10 seconds, topping up to 4.25V... uhmmm



But I'll check again circuit tracing (other info are on the schematic).

I've got one of these too. Very pleased with it so far...but I'm intrigued if you found problems. I usually pull my batteries as soon as the led goes green though.

Non-terminating chargers are rather common, and they do wear out batteries faster. Some would even consider them dangerous, though I think with this low rate trickle, it'd take quite some time for overcharge damage to accumulate (and it IS an accumulating damage).

How much extra would it have cost them to make the circuit "stop" once the "Done" led is triggered? Because that led is triggered at the correct stage in charging process