0
\$\begingroup\$

I previously asked asked an advice on power control of battery operated system.

The problem was to cutoff the battery when low to avoid overdisharge. I have built the device according to Olin Lathrop's advice to add hysteresis to avoid oscillations. Unfortunately this does not work. The oscillations still appear in very unpredictable way. Is there a way to add a control with stable hysteresis and preferably with reset button. My current schematic is: stable voltage is compared with battery voltage at the comparator, when certain threshold is achieved the comparator switches and switches off the load via power mosfet, 200mV positive feedback from is added to input of comaparor to add hysteresys. It work perfectly fine in simulation software but still oscilates in real life. The problem might be in capacitance of the load or whatever. I need a different approach, something that would latch and hold the OFF state.

\$\endgroup\$
  • \$\begingroup\$ Did you experiment with R2? It may be that adding more hysteresis by using a smaller resistor value will work. Also, can you provide a more complete schematic showing exactly how the output of the comparator is connected? Is it connected to the coil of a relay? And is the relay between 12V and load, or between load and GND, as you show in the original question? One schematic will answer all these questions. I wonder if the relay coil inductance could be part of the issue. \$\endgroup\$ – mkeith Aug 30 '15 at 16:55
  • \$\begingroup\$ Yes I have experimented with R2, no help. I don't use a relay, the output of the comparator is connected directly to the gate of beefy power mosfet. \$\endgroup\$ – dmitry8080 Aug 31 '15 at 10:57
  • \$\begingroup\$ I assume it is high-side PMOS with the comparator set up to pull the PMOS gate low when you want to turn on the power. If so, you can add even more hysteresis by connecting a feedback resistor from the PMOS drain (VOUT) to the inverting input of the comparator. You can add as much as you want. \$\endgroup\$ – mkeith Aug 31 '15 at 17:57
0
\$\begingroup\$

200mV positive feedback from is added to input of comaparor to add hysteresys.

This may not be enough. A discharged battery has higher internal resistance, the voltage drop can be much bigger than 200mV depending on load current and battery chemistry.

Some sort of latch is probably a good idea, but keep in ming that this circuit should consume as little current as possible - otherwise the battery might get damaged in under-voltage condition.

\$\endgroup\$
0
\$\begingroup\$

Solved this with a circuit similar to this one.

enter image description here

Works fine in sumulation. All circuitry including comparator and latch consuming 50mW, that's nothing for big lead acid battery that I am using. I decided to use both techniques, hysteresis and latch, since it requires only one resistor to add for hysteresis.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.