I am trying to design a battery powered remote controller that has the following components, TPS63802 (converter), AP9211 (battery charger/protection), Atmega328 (MCU), NRF24 (transceiver). The following circuit uses a push button to send a signal to the MCU to tell it to go to sleep or turn on. My questions is regarding the small current drawn my the MCU under sleep mode about (5uA-20uA depending on the sleep mode used) and how it effects the boost converter. More specifically will the Boost Converter be able to supply the 3.3v at such small currents and will the output be stable. Also this is my first time designing a battery powered device that doesn't have a physical switch that decouples the battery from the boost converter. Also is such a method recommend or are there better ways.


The front page of the datasheet shows operation down to light load; 80% at 100 uA load. Although the 2 A output seems excessive for a remote, this device is a reasonable choice. If TI has one that has a lower maximum output current, you will gain a little more efficiency at low loads because the gate charge in the power MOSFETs will be lower. Gate charge is all loss.

From page 16 (typos are TIs):

8.4.2 Power Save Mode Operation

Besides Continuos Conduction (PWM) Mode, TPS63802 features Power Safe (PFM) Mode operation to achieve high efficiency at light load currents. This is implented by pausing the switching operation depending on the load current.

This means at light load there will be ripple on the output. The converter will run for a few cycles and bring the output voltage up. The converter tuns off and the load (your uC in sleep state) draws the output back down. Page 17 of the datasheet has a drawing of this.

The TPS63802 will still draw current to run its circuitry. The datasheet quotes 11 uA.

  • \$\begingroup\$ do you think the TPS63031 will work better since the MOSFETs are rated for a lower current? \$\endgroup\$ – Fabian Cojman Mar 12 at 2:21
  • \$\begingroup\$ The curves on the front of the datasheet show the TPS63802 having better light load efficiency. If the advance copy of the TPS63802 datasheet is to believed it has lower quiescent current. Also you can hope that the TPS63802 is a follow on to the TPS63030/TPS63031 and consequently incorporates improvements. \$\endgroup\$ – jherbold Mar 13 at 20:20
  • \$\begingroup\$ Another difference is the TPS63802 incorporates more features like soft start. \$\endgroup\$ – jherbold Mar 13 at 20:21
  • \$\begingroup\$ Finally consider that the TPS63030/TPS63031 are in production and available for sale now but the TPS63802 is in a preview datasheet of a preproduction part.You will only be able to get samples from TI. As a former analog IC designer I can tell you that some of the quality assurance testing, production test development and so on can cause a big delay in IC production release. That is some re-design and a 2 month turn through the fab may be necessary. \$\endgroup\$ – jherbold Mar 13 at 20:25
  • 1
    \$\begingroup\$ I noticed that yesterday when trying to order the part from Digi-key. I decided to use the TPS63031 in the current design but when TPS63802 becomes available I will compare them and see if the quiescent current is actually lower. \$\endgroup\$ – Fabian Cojman Mar 13 at 21:04

The TPS63802 can certainly regulate the output at very low output currents.

It has a mode select pin that allows you to select between forced PWM mode (where the frequency will be constant but the efficiency low at light load) and PFM mode where the frequency will depend on load current at light load, but efficiency will be higher.

PFM mode would be the right mode for a battery powered application. The output ripple will be slightly higher, but to power an MCU in standby mode it won't make a difference.


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.