# How does this battery discharger work?

I'm trying to figure out how this rather simple discharging circuit works and what the SOT23 component is. What kind of component am I looking for? Some sort of transistor, right? But what kind? It doesn't match the common pinouts of anything I've looked at and I can't wrap my head around the functional principle of this circuit to derive it from that.

Specifically the 5.1 MOhm resistor confuses me because while it was obviously chosen to ensure a negligible standby current, it also drops the divider's output voltage into the mV range.

Functional description: It is designed exclusively for a specific battery type/voltage (4S LiIon/LiPo), which it discharges to approx. 15.5 V before it disconnects the load.

Component description: A voltage divider consisting of a 500 Ohm (R1) and 5.1 MOhm (R2) resistor, connected to pin 2 of an unidentified SOT23 component, a 430 Ohm load resistor connected to pin 1 of said SOT23 component, a fuse¹ and—in parallel with the load resistor—an activity indicator LED (+resistor).

¹based on the silkscreen label on a virtually identical product

Schematic (sans LED and mystery SOT23):

simulate this circuit – Schematic created using CircuitLab

• As others have suggested - a TL431 fits the need very very well. The two resistors gove a 6.1 divide ratio so the TL431 turns on above 2.5V x 6.1 = 15.25V (approx). || Rload to Cathode, Battery negative to Anode, gate to resistor divider. Commented Aug 7 at 7:22

This might be many things – from exploiting the exponential current vs voltage curve of a BJT (so that there's a sharp decline in current through the collector-emitter junction when the base-emitter voltage drops below a certain value) to simply a battery protection IC whose job is to cut off a battery when its voltage drops below a specific value.

That's the thing with unmarked ICs, nobody can know what's going on inside.

If you want to design a circuit with a purpose yourself, you might want to ask about that design; of course you can refer to existing circuits, but generally, there's usually good reason to not copy commercial designs 1:1 in your own designs; the parts you choose when you produce many of a thing can be pretty different.

it also drops the dividers output voltage into the mV range.

no, that's not true. Your R1 is labeled "105", that's 10·10 Ω , so 1 MΩ. So this is a 1:5 voltage divider, so 1/6 of the 16V fall of over R1, and 5/6 over R2, in other words, at 16V in, the voltage at the node between these two resistors is 2.67 V.

The load resistor says "4300", so that's 430·10 = 430 Ω, not 100Ω. That makes a lot of a difference in practice, because assuming there's 16V across that resistor, there's 0.6W of heat being produced when it's 430 Ω, and 2.56 W if it was 100 Ω; that's a lot of heat.

• Thanks for the correction regarding the resistors, I have edited the schematic accordingly. I simply forgot to adjust the default value for Rload and wrongly assumed that R1's marking was "501", based on the similar product's (see the link in the original question) marking of the apparently corresponding reistor being "310", which is still way off from 500 Ω, but I chalked that up to different specs of the SOT23 component. It's most likely "01E", though. The reason why I chose this product as a starting point is that I'm looking to eventually create 50+ similar units, not just one or two. Commented Aug 6 at 22:16
• The classic TL431 or one of its clones would perfectly cut off the load at 15.5V.
– Jens
Commented Aug 6 at 22:16
• @marvp 50 units is still "very few". You'd design for this from scratch, with a constant voltage source, probably. Which, as Jens correctly points out, might be exactly what they've done, but it's probably not the "classic" TL431 (that one would melt to a beautiful clump at the current that would flow through it in this circuit, which is 4 times as high as the rating), but a modified clone. That's what I meant with "mass-production demands different design choices". When you produce 50 of something, you'll not cut a bargain with a part supplier for a meagre 50 pieces. Commented Aug 6 at 22:21
• @MarcusMüller There’s about 40mA maximum flowing through that TL431. It’s rated for 100mA. Furthermore, the load resistor drops most of the voltage. That TL431 drops about 2V only and dissipates less than 0.1W. It is a perfect fit for the schematic. I’d say the circuit is exactly what it should be around a 431 used as a discharger. Commented Aug 6 at 22:47