FREEME ✌ ASTROLUX EA01 SST40 (2300lm)/ XHP50.2 (3500lm) TIR 26650/ 26350 Flashlight Group Buy - ALIVE!

So you touch one lead to the battery tube and the other to battery - ? While light is off with lighted switch on low? Or lighted switch off? I’ll test mine if it is possible with my cheapy Amazon MM…

Yes, that’s right. The red probe needs to be plugged into the A mA socket to measure milliamps and the dial should be set to A mA also. You could try the micro amp setting (uA) but it probably will show OL unless you are lucky and got a low drain EA01.

code please

What reading did you get?

Sorry I’m not able to try it tonight. I will do by tomorrow though.

Found the leaky part. Removed a 3 legged IC, maybe a LDO, as part of the charging circuit and on the bench, the standby drain went down from 250 uA to 32 uA. This was on the EA01 that read 640 uA on standby with switch LED's off. It's probably the only cause but won't know for sure until it's fully assembled again, repeating the setup for the 640 uA reading. It was the only resistor outside of the standard LED driver circuitry I could find connected to ground (i.e. Batt -).

Thanks for troubleshooting Tom.

It bugs me enormously, I’m a great fan of USB-C charging within a light so I do not want to disable that. But with the casual way I handle flashlights (and so it should be, one should not have to think when using a flashlight) this light will continue to kill a battery every now and then.

And with 5mA parasitic drain (with or without the switch light on) I got the worst from the lot.

:frowning:

Showing the part removed:

Haven't been able to get a good pic of it with the #'s:

Checked two caps in the driver circuit and one is 40 uF (C1) and other is 5 uF (C2). Our usual FET+1 design originated by DEL is to use a 10 uF primary (C1) of Batt+ to GND after the R5 4.7 ohm resistor, and an optional C2 of 0.1 uF between the V+ and V+ of the MCU. I'll have to re-check this - might be inaccurate to measure caps inline. Also know DEL wasn't too tight on specs for the cap values so these might be ok to use.

This was the light that didn’t charge if I remember correctly. Do you plan to replace the part and how would removal affect a working charge circuit?

Yes - this one didn't work for me. I don't have the know-how to redesign the charging circuitry, so only option for me is to remove/disable it. Would be great if the 3 pin part could simply be replaced with a non-leaky one, but usually it's not that easy.

The three-legged component is an N-FET.
3416
I think it connects the charging circuit to the battery when you’re feeding the usb-c from the outside.
I’ll try to put resistance between Gate and Source in case it stays open and that’s what’s causing the problem.

Thanks for researching this! So the "Gate" is the pin marked "G"? Think the "D" pin was connected to GRND and another pin connected to Batt+ -- would that be the place to add a resistor? If so, what value do you think I should try?

I’m suspicious of this resistance on Gate, only 47Ohm. (R2) Around 1K is commonly used.


By default, I was thinking 10-20K resistance to R1.

Ok, was looking at that resistor, yes 47 ohm. Problem with this piece is I don't think the charging was working. On my other light, the MT01 (same driver), I know the charging works and the leak is even bigger. I've been inspecting for shorts, bad solder, etc. - haven't found anything definite. The other pin seems to go through that big diode - is that supposed to be? The thru-holes there got me confused though. The USB processor chip (I think??) is on the other side of the driver. It's a square package, 16 pins, 4pins each side.

This pic is from the MT01 - exactly the same as the EA01, but the MT01 is a lot cleaner. Been cleaning the EA01 driver up with isop alcohol:

The 16-legged IC is the controller of the charging circuit.
TP5100
Unfortunately, I don’t really understand why it’s built this way.
The Gate resistance is connected to the positive of the battery, not the usb. It’s close, and I mistook it.

I thought it was strange, but something I'm not seeing/understanding.

I think the 3414 FET is the reverse polarity protection of the SS52 diode which would short the battery otherwise. When inserting a reverse battery.

The charging circuit itself could be the leak that you turned off by removing the FET or just the diode. I don’t know.

So component failure or design flaw ?

I do not know, we should compare one with one of which has no problem. Mine consumption decreased from 2mA to 0.6mA after soldering and cleaning (isopropyl alcohol). I’m trying to figure out the reason for the high power consumption, but the parts are very small, and the circuits are hard to trace.

Well, thank you and Tom for trying.