Battery Supply: 2.4V (2x NiMH 1.2V - 1500mAh batteries )

Charging Supply: 5V (USB Power Adapter / Wall Wart)

Switching Regulator/Controller: NCP1450A 3.3V Step-Up Boost Supply Voltage

Circuit Description

Power Consumption: The circuit will be fully on for 12 (15 second) intervals every day consuming 150mA of current (3 minutes total in 24 hours). The circuit will consume approximately 50uA of current while sleeping.

Circuit Composition: Put simply, the circuit is composed of an MCU that reads sensor values and controls output to peripherals based on sensors and timing intervals.

Battery Charging: The primary MCU, or optionally a dedicated secondary MCU, will control the battery charging and maintenance.

Usage and Application: The device is primarily intended for battery operation, but may just as easily be placed within range of a power outlet where it can remain plugged into the provided dedicated 5V power supply for extended periods of time or even indefinitely.

Questions & Concerns

Concerns: In a circuit engineered to constantly pull current from the batteries, even when the optional dedicated power supply is connected for charging and maintenance, the battery life will be reduced and the batteries damaged from being subjected to a constant charge and drain. Also, accurate battery capacity may be difficult to ascertain once the initial charge sequence is completed which may lead to over or under charging.

Question 1: Are my concerns founded? Or, alternatively, will any potential life-reduction/damage be sufficiently inconsequential to omit the necessity for an alternative design solution?

Question 2: Assuming my concerns are founded, what will be an appropriate design to isolate the batteries from the circuit while the device is plugged into the dedicated power supply yet will still facilitate the charging and maintenance of the batteries?

Please reply with your ideas and design considerations. Your input will be greatly appreciated as I do not have much experience in this area.

  • \$\begingroup\$ I'm trying to understand Q2 - Why do you want to isolate the batteries from the 3.3V switcher(?) when they are charging? Doesn't the 3v3 power the MCU that you need to manage the battery charging? \$\endgroup\$
    – Andy aka
    Commented Jul 5, 2013 at 15:40
  • \$\begingroup\$ I was proposing that the batteries be isolated "only" while the secondary power supply is attached which will provide power to the MCU through the 3.3V switcher when connected. \$\endgroup\$
    – GabeNix
    Commented Jul 5, 2013 at 15:45
  • \$\begingroup\$ How are you planning to do the battery charge voltage/current regulation? With a dedicated IC or under microcontroller control? Also, your boost regulator may operate inefficently at very low current draw. \$\endgroup\$
    – pjc50
    Commented Jul 5, 2013 at 16:11
  • \$\begingroup\$ To elaborate further, the proposed scenario would swap the power supplying the 3.3V switcher from the battery to the wall wart. The batteries will then be isolated from any current draw. The switcher will be receiving power from the wall wart, and will still be powering the MCU which will maintain the battery charging. To put it simply, once the device is "plugged in" via a wall wart, the unit becomes a battery charger and does not pull current from the batteries. The prior proposed method in my "concerns" is the batteries will "ALWAYS" be supplying the current draw to the switcher & circuit. \$\endgroup\$
    – GabeNix
    Commented Jul 5, 2013 at 16:18
  • \$\begingroup\$ @pjc50 - I am planning to use only one MCU for all aspects of the circuit as well as battery charging. The amount of available I/O pins may necessitate the need for a dedicated charging MCU though. I will employ voltage readings and temperature sensing of the batteries to control the charge current. I havn't looked at the efficiency of the boost controller at low current. The boost controller does have a sleep mode though and the circuit may be designed to put the boost to sleep and draw directly from the batteries during MCU sleep time perhaps.. \$\endgroup\$
    – GabeNix
    Commented Jul 5, 2013 at 16:26

1 Answer 1


I think the answer to that question is "yes": e.g. Panasonic - Nickel Metal Hydride Batteries - Technical Handbook 2000 explicitly says that you can't trickle charge NiMH, except for a defined period of a few hours after rapid charging. With NiMH you need to carefully measure the voltage while charging to look for the "end of charge" condition. Having a circuit run directly off the battery will also make that harder.

You could however do:

  • on plugging in, start charge
  • stop charging as normal (voltage dip detection) and set a timer
  • after approx 24 hours, retrigger the charging cycle

Or select a different battery chemistry which can be float charged such as NiCad.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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