0
\$\begingroup\$

I am working on a zigbee wireless remote. MCU datasheet says it operates in wide voltage range of 1.8 to 3.8 volts.

Besides zigbee MCU, there will be a few LEDs and tact switches (fairly simple and straight-forward circuit).

I am leaning towards using a li-po cell in my design. (previous thought was a CR3032 coin cell but moved to li-po for higher capacity and to make it easily rechargeable by any mobile phone usb-c charger. Rechargeable battery appeared to be more user-friendly.).

Are there any major draw-backs in terms of safety and product life just because a li-po cell is present? I am targeting a product life of 5 years or more. Remote will have to be charged every 6 months or so.

Below is the discharge profile of a li-po cell:

discharge profile

The usable capacity of the cell lies in a voltage range of 4.2 volts to 3.6 volts. As such, how do I make the most of it? What kind of voltage regulator will give me best efficiency? Should I just do away with a regular diode? 0.7 V drop keeps me in fairly safe zone.

At full charge, I get 3.5 V and at low charge (3.6 V cut-off), I get 2.9 V. These are very comfortable voltages for my zigbee MCU.

\$\endgroup\$
1
  • \$\begingroup\$ You shouldn't power the MCU directly from battery regardless. If cost-sensitive, there's various MCUs with built-in regulators for the purpose of battery applications. Otherwise just add an external regulator. \$\endgroup\$ – Lundin Nov 26 '19 at 9:15
0
\$\begingroup\$

The simple answer to your first question is yes, there are lots of draw backs to LiPo cells. They are expensive, large, sensitive to temperature, shock, over charge, under charge and subject to various regulations due to their fire/explosion risk. And that's all in storage, charging gets more complicated (temperature, current and voltages all need to be monitored to charge them safely). It is up to you to see how they will effect the end product. If you are just doing a quick one off for yourself, it'll be fine. If you're selling 10 million a year, you'll need to analyse that yourself, or pay someone to do it for you.

As for your second question, you say your processor can take any voltage from 1.8 to 3.8V, and claim your cell is fully charged at 3.5V, empty at 2.9V, so from those numbers why would you use a regulator at all?

\$\endgroup\$
1
  • \$\begingroup\$ 3.5 V = 4.2 V - 0.7 V. So essentially diode working as a voltage regulator. Calling it a voltage regulator is technically incorrect. Diode is just shaving off the extra voltage so that my MCU doesn't blow up. \$\endgroup\$ – Whiskeyjack Nov 26 '19 at 9:21
0
\$\begingroup\$

Battery powered RF embedded systems are sensitive. You should be care about transmission current and maximum current your battery and cap could provide. You can use a battery for your application but you need a "PMIC" for your design, it handles your power delivery to your circuit and it's very a useful component with low BOM for battery powered design.

The second thing is the type of your battery, your system condition (temperature, humidity, etc) is important. Li-Po batteries are sensitive too:) If you want to use them check the maximum tempreture in datasheet and buy something much bigger than your requirements, because Li-Po batteries lose about 20% of their capacity after 500-1000 charge cycles.

\$\endgroup\$
3
  • \$\begingroup\$ While the issues you raise are things that should be considered in general, this is not a very useful answer as it shows no real familiarity with the technology proposed at all, guesses at things which are contrary to the facts of the actual situation, and completely misses the concerns that actually are important. When the original design worked on a CR2032, pulse power demands are not going to be challenging for a LiPo at all. \$\endgroup\$ – Chris Stratton Nov 26 '19 at 16:55
  • \$\begingroup\$ @ChrisStratton,Hi.He want to use Li-Po batteries in his design,A Li-Po or Li-Ion with PMIC is good option instead of CR2032 ,because they are rechargeable. \$\endgroup\$ – Danesh_sa Nov 27 '19 at 15:16
  • \$\begingroup\$ Maybe, maybe not. The problem is that due to an apparent lack of familiarity with the differences in capabilities, your posting gets sidetracked into things that are not problems with the idea, and ignores the actual problems with the idea. \$\endgroup\$ – Chris Stratton Nov 27 '19 at 15:50

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.