I recently bought AK-47 driver boards from KD http://www.kaidomain.com/ProductDetails.aspx?ProductId=10697
I allready had some of these, which I have got from DX http://www.dealextreme.com/details.dx/sku.6190 , so I'm quite used to work with it. I installed one of the KD boards in a flashlight and it didn't have modes at all. I didn't put any jumpers at all, so it should have 5 modes. What happened? The KD drivers look exactly identical like the DX ones. Now I'm confused. Why they are not performing the same? Did somebody else have the same problem with these?
Maybe they're empty , not flashed, very good for Tido or as spare 7135 source, I ordered a few days ago some from KD, if that is the case not problem for me since is very difficult pull such chips from the 8 ones. Those chips are a lot fragile and pull off it is a hard task.Half dead is my 'score' with that.
This is also what I thought, but I didn't really know what is actually the information that is stored on the chip. I didn't know if it's only the mode arrangement, or if there also are other parts of the functions managed by the chip. I also got a five-pack. I didn't test the other four yet, but when I have one faulty, it's absolutely possible the others are as well. I will check another tomorow.
I was allready thinking of getting inside to program the chips later one day, but I don't like to be forced into doing it. I look forward to a long story, when I try to explain to the staff members at KD, what's wrong with these.
Btw, I have also got five of the new 13mm driver boards http://www.kaidomain.com/ProductDetails.aspx?ProductId=11063. These work well. It seems they are only for Nimh or alkalines. They are a bit like the drivers in the BLF MrLite. They allways start on low mode first. I tested shortly a 14500 with the ampere meter. I had allready 1100 mA at mid mode. Then I touched just for a tick on high, just long enough to see it going ower 3000mA. However, the bord did take it without damage and still works fine. But now I'm sure, it's definitely not made for Li-Ion.
Edit: My score on harvesting AMC7135 chips alive, is also only around 50%. I also tried to get off an ATiny to fix this board. Well, it was a wash out.
The components on the driver are really small and close together. It is easy put too much solder on there when attaching the leads to the LED. Get a magnifying glass and make sure that nothing is being shorted out or touching something it isn't supposed to. I had a 1-mode driver one time and it just turned out I had oversoldered. I was able to fix it by getting rid of the extra solder.
It could be worst than empty,corrupted. I remember reading some time ago somewhere ( edited, never mind ).As saying , some guy reported that all of his drivers which were stored in a drawer one year become as yours.
The bad thing with the 8 one's is you can't put it in a controlled hot plate to pull out that stuff safely and easily.
I've managed to fix two drivers, one with the DC DC converted faulty and another with the core of the coil broke. Keep all the faulty one's for spares, you never know.
I don't like less than li-ion drivers because with one AA give little power and 2 AA are out of my size factor preference.
I think rechargeable 3,7 cells are optimal to feed leds and cheap in the long run also.
You should check if the 7135s are connected to the MCU first. If you look at the picture (shamelessly stolen from brted, btw), the right hand pins of the regulators need to be connected to pin 6 on the MCU. If you can find no problem there, check if there is a connection to ground. This would permanently disable the regulators.
Next thing on the checklist is whether the MCU gets power. Wire it up, turn it on and measure pin 8 against ground. If it is powered, you're out of luck. The MCU then is probably defective or at least the program flash is empty or corrupted.
As for de-soldering components, I have found that the easiest way to do it is with a hot air gun. Grab the part you want to remove with a pair of tweezers and heat up the board with the hot air for a few seconds. Three to five seconds is usually enough and won't bake the chip.