4
\$\begingroup\$

I am trying to design a very simple circuit consisting of an accelerometer IC and a Bluetooth IC, all rated at 1.8 up to 4.something volts and thinking to use a single CR2032 cell. Seems like these usually come at 2.8 - 3V and about 250mAh current, depending on manufacturer and other factors.

I have already planned to add required capacitors around each IC, but I am wondering if I should still use a voltage regulator with this circuit?

So far the only good reason I could think of would be to prevent frying the ICs in situations where the user would manage to somehow cram up 2 batteries instead of one in the enclosure, aka making the circuit idiot-proof.

\$\endgroup\$

3 Answers 3

3
\$\begingroup\$

Adding a voltage regulator would dissipate more power (although not much, if it's a good switching regulator), which isn't desirable in a battery-operated device. If those chips are meant to run over that full range, then I would take advantage of that. They were probably designed that way to address battery-powered applications.

If you want over-voltage protection, put a zener diode across the battery input. That would also give you reverse-battery protection and won't draw additional power (except possibly for a little leakage). Select one whose zener voltage is greater than one battery and is less than two batteries. The current carrying capability should be greater than the current capability of the battery.

\$\endgroup\$
0
2
\$\begingroup\$

From your description, you do not need a voltage regulator on a single 3.0V lithium primary cell to prevent damage to your two devices.

Note that there are still some reasons you may consider regulating the cell down to a lower voltage:

  • The quiescent current of the circuit may be lower at a lower VDD (though you have to add the regulator current)
  • The circuit may change performance as VBATT changes (due to state of charge, or large pulse currents)
  • The circuit outputs an analog voltage proportional to VDD
  • You want a lower transmit power from your Bluetooth

That being said, you probably don't need to include a regulator for your application.

\$\endgroup\$
2
  • \$\begingroup\$ Thank you for your answer. The BT IC has a register where I can configure the RX power so there's no need to play with the voltage for this reason. You could be right about the current at a lower voltage though, although I'm not sure if will compensate the regulator current even in case of a very low current regulator which would add to the price tag...guess it's easier replacing the battery a couple months earlier (so the user pays for it) as opposed to me paying for it 1-2 dollars extra in the front-end. \$\endgroup\$
    – Nick M
    Apr 26, 2016 at 5:25
  • \$\begingroup\$ performance/drift/gain/offset versus Vbatt is pretty much the only reason to add a regulator, what's the accuracy spec? \$\endgroup\$
    – Neil_UK
    Apr 26, 2016 at 5:43
2
\$\begingroup\$

Most available linear regulators have a "drop out" voltage by which the output will be less than the input. This can range from under 100 mV for good examples lightly loaded, to well over a volt for others. So putting one in means you reduce the available voltage, which may not be something you want if it is already in range.

If all of your parts can tolerate the open-circuit voltage of a new CR2032, then you probably do not want a linear regulator.

There are, however, some tiny switching regulators designed to save power when running devices that can operate in the sub-2v range, from a coin cell by efficiently reducing the voltage to make them consume less energy. Good low power design of course is an art that extends beyond the supply - you need to make sure never to leave a voltage across a pulling resistor, and to sleep your processor and its main clock the overwhelming majority of the time.

More importantly, if you ever think of using a different chemistry, such as lithium ion or lithium polymer, then the voltage of a fresh cell may be around 4.2v and so likely too much for some of your components. In that case you might try using a very low dropout 2.8v or 3.0v linear regulator, or else looking at the switching solutions.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks Chris. I'll double check the rating on the ICs and buy a bunch of different cells to see what sorts of voltage they output out of the box. Surprised to see they can go as high as 4.2, didn't expect that nor noticed previously. \$\endgroup\$
    – Nick M
    Apr 26, 2016 at 5:31
  • \$\begingroup\$ Lithium Ion and Lithium Ion Polymer are different chemistry than the typical Lithium / Manganese Dioxide coin cells, and so have a different voltage. They tend to come in different form factors, but I mentioned them as a counterexample where you typically can't go without a regulator, and in case your plans move in that direction at some point in the future. \$\endgroup\$ Apr 26, 2016 at 5:34

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.