Pi PICO W on batteries

Why not 18650s? 2 would give you a little over 7V and battery holders similar to the AA types are available. The TP4056 charging modules are inexpensive if you want to incorporate USB charging.

"The maximum recommended input voltage for a Raspberry Pi Pico or Pico W is 5.5V. Powering the board above 6.0V will permanently damage the onboard voltage regulator"

I've been using Li-Ion rechargeables from disposed vapes to power many battery devices. Although the supply has dried up now, I managed to collect over 100 3.7V 500mAh, 550mAh, 1400mAh and 1800mAhr cells when out walking the dog.

You still find such vapes but rare now. You do find rechargeable vapes now and then and can recover the USB rechargers from them to use with previously recovered batteries. Rechargeables are normally disposed because the battery is dead. All the chargers have been fine.

The Energizer lithium primary cells are a lithium iron sulphide technology. I have measured the cell voltages of about 20 new cells and found them all to be between 1.800V and 1.805V open circuit. This would make them completely safe for this usage. They maintain a voltage above 1.5V for much of the discharge curve and most importantly they are unlikely to leak. John

One cell should be sufficient for powering the Pico. The regulator lower limit is 2.3 V so there is a need for deep discharge protection on the cell.

Exactly. I have lithium ion polymer cells coming out of my ears, but one cell is too low and two cells are too high.

This technology is higher voltage, greater capacity, and better performance over the discharge curve and in heat and cold than alkaline cells. I estimate over a years life.

Which is good because its a ladder climb to access the unit.

I bench tested the device., Its not just a pico W. There is an ultrasonic device in there as well.

It stared to fail around 4V.

Then two 18650 in serial and a 5V switching regulator should do the job. Perhaps a small USB powerbank may also work if you find one that does not shut off if the consumed current it to low.

Sigh,

The nano power switch will not handle that voltage Putting a regulator in front negates the point of having a nano power switch,

Or drain the power in a month...

Sheesh. Why do some people always think they know better?

"I have examined, X,Y,Z,A ,B and C I found Q to be optimal" "Why didn't you use A B or C?" "

nano timer ??

What's that then?

TPL5100

-- Future generations will wonder in bemused amazement that the early twenty-first century’s developed world went into hysterical panic over a globally average temperature increase of a few tenths of a degree, and, on the basis of gross exaggerations of highly uncertain computer projections combined into implausible chains of inference, proceeded to contemplate a rollback of the industrial age.

Richard Lindzen

It's disappointing that it's not available in a higher voltage version since, as you say, it needs to be directly connected to the battery.

I suppose one workaround would be to use a single lithium ion cell to give you down to 3V and then a boost converter which you only run when the CPU is operating. Timer goes off, boost converter spins up to power your CPU at a stable 5V (or whatever), then CPU tells it shut off again once done.

That's likely going to take more power than just running directly off battery, but the flipside is you can install several 18650 cells in parallel if you want to get more capacity.

The primary lithium AAs aren't a terrible option though, especially if you are more likely to be killed by self-discharge of the battery than by your own load.

Theo

One workaround is to use 3 lithium primary batteries,.

The primary lithium cells have more capacity. Better temperature response. Better discharge curve.

They are, having exhausted all other alternatives, by far and away the best option. This design took nearly 3 years from idea to completion, and this is the optimal solution

The ultrasonics have no voltage regulation. Just very high voltages to drive them Over 20V. They fail if they don't get at least 3.9V input

I kinda thought modern microcontrollers have this sort of thing built in? Oh, looks like the Pico (no W) draws 2 mA in "deep sleep" and this timer draws 35 nA.

Yes.

Plus there is the advantage that you get a complete reboot every time things fire up.

And its very very simple. Drive pico pin high. get power cut.

That does sound surprisingly bad for the Pico. Seems that the RP2040 is

If you've got other stuff hung off the MCU that isn't super low power, you're going to have to power-gate it anyway, in which case the nano timer will do that at the same time.

Theo

The broadcomm wifi chip.

Indeed. I have an ultrasonic unit.

I looked at every combination I could think of, and ended up with the nano timer after trying to construct my own in analogue.

I can say that the three lithium primary cells plus the nano timer (MUST have a BIG electrolytic on its back end or it goes unstable) is THE very best way to have a periodic remote sensor 'phone home' if a cable can't be laid

I think (ultra) low power was not a primary design goal for the Pico.

I think that is very true. What attracted me was the amount of flash, and ram, the availability of wireless and ADCs and the price. Less than a pint of beer in a nightclub. Arduinos were 3 times the price

The ZeroW running debian is a damn sight easier to code for, however.