Lipoly discharger - troubleshooting

Thread Starter

IrejectYourReality

Joined Jul 28, 2016
7
Hello! I am working on a simple project, a device to discharge my lipol batteries. Fully charged, they have 4.2V per cell, and I need to discharge them to 3.85V per cell for storage. I am using 2 cell batteries, so they have 8.4 V fully charged. This is my circuit, which I built on a breadboard:

The idea is, that the TL431 will provide 3.85 volts, which is compared by the comparator(LM2903) to the battery voltage divided by two (2 cell battery). If the battery voltage is higher than 3.85 V, the mosfet will turn on and burn the energy on a power resistor (the resistor is actualy only 4.7 ohms, forgot to update the schematics). The problem is, that the comparator only puts about 2-4 volts on the mosfet gate, which is not enough to turn it fully on. If I disconnect the mosfet and measure the comparator output voltage, it is at saturation, about 8V, but when I connect it to the mosfet, it drops. I measured the voltage drop of the battery when directly connected to the 4.7 ohm rezistor, it dropped from 8V to 7.99V, so the voltage drop of the battery should not play a significant role. And another observation, I just tried to connect my only 3S battery (11.1V) to the circuit, which basically means that on the noninverting input, there will be about 6V instead of 4V and it worked flawlessly, mosfet turned on completely. So the problem seems to only arise when the voltages on the comparator inputs are close to each other. What am I doing wrong?
 

Thread Starter

IrejectYourReality

Joined Jul 28, 2016
7
Yes, it is a dual comparator, I just put one in the schematics, not to make it confusing.

EDIT: And I connected all the pins of the other comparator to GND, as advised in the datasheet.
 

#12

Joined Nov 30, 2010
18,224
I think you have a broken part, but I wish to make some words clear. The comparator does not apply voltage, R8 applies voltage. The comparator is an open collector output which means it can only short current to ground. It seems your mosfet is dragging the gate voltage down when it should be not conducting current through the gate.
 

Thread Starter

IrejectYourReality

Joined Jul 28, 2016
7
Is there a way to test if the mosfet is bad?
I tried this: I replaced it with a BS170, which is not rated for such a high current, but I disconnected the resistor. The LED was turned on. When I connected the resistor back, the LED turned off and the gate voltage dropped to something like 2V.
 

#12

Joined Nov 30, 2010
18,224
You have to be clear about, "the resistor". Is that the 4.7 ohm load resistor or the 5k gate resistor?
 

#12

Joined Nov 30, 2010
18,224
If it's the gate resistor, that's just backwards. If it's the load resistor, it isn't even connected to the gate and should have no effect. Now I'm thinking about a wiring error.
 

Thread Starter

IrejectYourReality

Joined Jul 28, 2016
7
So I pulled all the parts from the breadboard and placed them again, trying to eliminate any possible wiring errors. The result is the same.

Tried to measure the gate voltage depending on load, if I disconnect the LED and the power resistor, I get 7.79V at the gate. If I only connect the LED, I get 7.77V. If I connect the LED and a 110 ohm load, I get 7.74V. And finally, if I connect the 4.7 ohm resistor, I get only 2.63V.
 

Kjeldgaard

Joined Apr 7, 2016
476
I think we need to look more closely at the construction of the circuit. It's in fact a large current, which can be a problem if the cables are too long, or to measure in the "wrong" end of a wire.
How long are the wires from the battery to breadboardet?

I also see a possible problem in that there is no hysteresis in the comparator circuit. In fact, we must have a negative feedback by the battery voltage falls below preset, so decreases the load and the battery voltage will then stabilize around this voltage.
Here I would suggest to try with a resistance in the range of 470K Ohm to 1M Ohm, connected between the comparator output and the +input.
 
Top