ML-103 charger ?

[hank]

I just pasted above the URL that you get when you type
Miller “4.35v”
into the site search tool window — upper left corner
the percent sign stuff is what HTML makes of quotation marks and spaces

[wight]

@MRsDNF

I bought a ML style charger from a member here that charges to 4.35 volts but do you think I can find the thread. It was around August.

@hank

Search results for '' - BudgetLightForum.com and sort by date

I think that you are both referring to led4power’s MCP73831-based driver. led4power did not abuse MAX1879 for that driver, it uses a pair of entirely different charging IC’s.

@HKJ

@wight

@eas

Another detail: there are some chinese boards with this chip that are somehow using it out of spec as a 4.35v charger.

Any tips on how that’s being done, links to info, etc?

The obvious way to do it to raise the gnd pin 0.15 volt above 0V.

Right, that’s what I assumed. I couldn’t think of a straightforward way to do it. I think I’d end up using an LDO + a voltage divider to do it. That doesn’t seem very elegant.

If EAS is aware of it already being done, I would like to know how it’s being done.

[eas]

@wight

led4power did not abuse MAX1879 for that driver, it uses a pair of entirely different charging IC’s.

@HKJ

@wight

@eas

Another detail: there are some chinese boards with this chip that are somehow using it out of spec as a 4.35v charger.

Any tips on how that’s being done, links to info, etc?

The obvious way to do it to raise the gnd pin 0.15 volt above 0V.

Right, that’s what I assumed. I couldn’t think of a straightforward way to do it. I think I’d end up using an LDO + a voltage divider to do it. That doesn’t seem very elegant.

If EAS is aware of it already being done, I would like to know how it’s being done.

I came across it on TaoBao. The boards were field switchable between 4.2, 4.35 and, I think 4.1v charge termination. The first place I found them, there was no mention of how that was done, but at another, they said something to the effect that “the stock MAX1879 has very precise termination voltage detection for 4.2v. We apply an offset voltage for 4.35v termination, and the precision is lower as a result.” This is all from memory, and all from Bing or Google translate, but it suggests that it may be using the method HJK suggested. I’ll see if I can find the links. If I can, I’ll update the post.

[wight]

@eas

I came across it on TaoBao. The boards were field switchable between 4.2, 4.35 and, I think 4.1v charge termination. The first place I found them, there was no mention of how that was done, but at another, they said something to the effect that “the stock MAX1879 has very precise termination voltage detection for 4.2v. We apply an offset voltage for 4.35v termination, and the precision is lower as a result.” This is all from memory, and all from Bing or Google translate, but it suggests that it may be using the method HJK suggested. I’ll see if I can find the links. If I can, I’ll update the post.

Thanks eas. I’d love to eyeball the boards and see if I can spot exactly how they are doing it.

[eas]

This is the one with the explanation.

This is the first one I saw.

[wight]

Hmm.

Well (looking at the blue PCB) I’d say that the 3.6v charging is certainly making use of the MAX1879’s “Adjusting the Battery Regulation Voltage” feature by pulling ADJ down towards GND with a 60.4k resistor. As for the 4.35v charging… not sure, but I’d assume that they’re pulling ADJ up (which is not spec’ed) with 260.4k (or more?). What are they pulling it up against? Vin, which would normally be allowed to be anywhere from unregulated? Ah, hence the warning: “Input voltage: 4.5-5.5V (not recommended too high, otherwise hurt the battery)”

[Halo]

Then instead of “Input voltage: 4.5-5.5V (not recommended too high, otherwise hurt the battery)”
IMHO it should be “Input voltage: 4.5-5.5V (not recommended too high, otherwise burn house, murder loved ones, bomb in flashlight, hurt battery)”. For lack of regulator. Pair this with a spewfire 18650.

[wight]

@Halo…

Then instead of “Input voltage: 4.5-5.5V (not recommended too high, otherwise hurt the battery)”
IMHO it should be “Input voltage: 4.5-5.5V (not recommended too high, otherwise burn house, murder loved ones, bomb in flashlight, hurt battery)”. For lack of regulator. Pair this with a spewfire 18650.

I agree. Unfortunately I’m not surprised, but I was hoping that this wasn’t what they’d do. I’m happy to have someone prove me wrong, but I find it almost certain that these things have components only on one side and I see no regulator so I see no way they’re getting the higher voltage from anywhere else…

[Halo]

Never surprised with this stuff. :weary:

[eas]

This ebay listing seems to be the yzxstudio board. It has shots of both sides. Nope, no components on the backside.

[wight]

Good catch. I saw the listing (expensive, 2.5x as much as Taobao…) but I didn’t note the shot of the back side.

So the 4.35v jumper leads back (by way of a trace on the backside) to that little area between IN+ and IN-. That area features a 2k-ohm resistor pulling it up to IN+ as well as an SOT-23 sized component doing… what? It could be an SOT-23 linear regulator… they appear to be available in whatever pin configuration your little heart desires… but to me it appears that pins 1 and 2 are soldered onto the same pour…

Am I looking at it wrong?

[eas]

The two pins do indeed seem to be soldered to the same pour. My basic electronics knowledge isn’t great, but if one of those pins is the gate, then wouldn’t that island tend to be at the threshold voltage of the transistor?