I've read some related questions, but I was not able to find/understand a proper solution

I'm using 2 AA batteries to power a SAMD21 MCU (arduino mkr series), a DS18B20 probe and a MCP3424 ADC. My problem is that temperature probe and ADC need more than 3V or 2.7V (respectively) to work properly, and I've seen that with voltages lower than 3V, measurements works, but have an important offset.

I'm trying to solve the issue, and my constraining factor is having the longer autonomy I can get, and the simplest circuit with less components. So I'm thinking on using a easy to solder, low footprint and very low power consumption step-up converter that could take the 2xAA batteries an step up to 3.3V for the instruments just during the measurements periods (about 30s every 12min)

Do you have any hint here? Thanks


closed as off-topic by Andy aka, Elliot Alderson, Leon Heller, RoyC, Chupacabras Jul 1 at 19:45

This question appears to be off-topic. The users who voted to close gave this specific reason:

If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ Shopping questions \$\endgroup\$ – Andy aka Jul 1 at 12:14
  • 1
    \$\begingroup\$ Does your MCU have a few spare I/O? Consider using a charge pump boost using I/O output pins for drive (perhaps PWM). \$\endgroup\$ – glen_geek Jul 1 at 12:37
  • \$\begingroup\$ What's easier ? , buying and wiring up a boost converter or adding a battery in series? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jul 1 at 12:40
  • \$\begingroup\$ @glen_geek Yes, its a SAMD21 and I've spare I/O pins, but I've not a consumption problem but a power voltage one. VCC is at battery voltage, so I've to power it up \$\endgroup\$ – avelo Jul 1 at 12:46
  • \$\begingroup\$ @SunnyskyguyEE75 obviously the latter (+batteries), but I cannot power the board with more than 3.7V, so using 3xAA is not an option here \$\endgroup\$ – avelo Jul 1 at 12:47

I would suggest to use the TPS61221 from TI.

It has no enable pin but you don't need to disable it because it has a very low quiescent current of \$5.5\,\mu\$A

  • \$\begingroup\$ Seems good thanks! Anyway, I'll try to find a way of powering it off, as my board is consuming about 25uA in idle, so 5.5uA is low, but higher than expected. I'll also try ti find a non SMD version of that chip \$\endgroup\$ – avelo Jul 1 at 12:36
  • 1
    \$\begingroup\$ seems to have an enable pin when I clicked your link! \$\endgroup\$ – Neil_UK Jul 1 at 12:57
  • \$\begingroup\$ None of the good ones are going to be other than SMD, it is very nearly a functional requirement, and for the supporting parts too. Get familiar with SMD before you do this project, options are far too limited in trying to avoid it. \$\endgroup\$ – Chris Stratton Jul 1 at 12:58
  • 1
    \$\begingroup\$ Ok, I've also seen that there is enable pin. This makes this option way more interesting. I've also seen that there is a board option here but will try to learn if I could use SMD version. Thanks all \$\endgroup\$ – avelo Jul 1 at 13:04
  • \$\begingroup\$ Your suggestion seemed very good. Digging into that, It seems that TPS61291 could even fit better there. I'll try to get one to test. Thanks! \$\endgroup\$ – avelo Jul 2 at 7:55

Best bet

3 batteries and a CMOS LDO to Vdd


No noise for ADC, more power from 3 cells, adequate efficiency especially when low bat with Vin~=Vout with 50mV dropout type CMOS LDO’s or 100mV dropout.. many to choose from.

Some have UVLO or DC OK status and extremely low quiescent current x uA.

Action item: Go search for CMOS LDO’s or ultralow dropout LDO’s. With your unstated specs

  • \$\begingroup\$ Sorry for do not post any specs, the reason is that actually the specs or capabilities I can reach are defining the scope of application (sensors in a low footprint mooring). About LDOs, that was my first bet, thinking even to use solar panels but, AFAIK, that implied high power loss and cutout when voltage drops significantly, so I discarded them (as I though I wouldn't have problems with voltage for sensors or ADC, and got very good autonomy with batteries). Probably I'll have to take a look again (I'm very newbie in electronics) \$\endgroup\$ – avelo Jul 1 at 19:22
  • \$\begingroup\$ Yes consider primary and alkaline x3 or secondary NiMH 1.2V x 3 =3.6 with no regulator but disconnect load on charger \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jul 1 at 19:37
  • \$\begingroup\$ NiMH would be a good idea, but I think that self discharging would be a serious problem for having the board working for 6M-1Y. Alkaline would give around 4.5V in the beginning, so way higher. The MKR series of arduino comes with a LDO (probably AP7215) for powering with up to 5.5V, but in my tests, a lot of energy were lost, and consumption was way higher, so the reason to discard that. \$\endgroup\$ – avelo Jul 1 at 19:56
  • \$\begingroup\$ Then I would suggest 2AA (2S) and run off 3V for even lower power consumption than 3.6V or 2D cells if that voltage works. Then take Vbat/2 to also monitor battery voltage and send info. But you should sate ALL of your power Specs in your question instead of comments with ideas. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jul 1 at 19:59
  • \$\begingroup\$ Edited and specs posted. I've just tried your bet by powering with 4.5V the mkr internal LDO (AP7215) which was supposed to be a very low dropout, low power consumption CMOS LDO. I've also removed power consumers as led. Idle current was 1.28mA... previously it was about 25uA without using LDO, so its a factor of about 50x or 1/50 in autonomy... \$\endgroup\$ – avelo Jul 2 at 7:49

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