6

I have several pi-zero boards and I utilize them to monitor environmental conditions (such as humidity, temperature and intrusion) of each room in home.

For continuous operation, I want to eliminate power blackouts. So, I decided to make a basic UPS for pi-zero boards. (need to be as small as possible)

I'm not a low level hardware guy, but a software guy. Here is my design: Power comes from AC/DC adapter to the battery charger. The battery charger charges 3.7V battery and powers 3.7V DC/5V DC converter concurrently. Then the (assumed) uninterrupted power comes to pi-zero board.

In theory, the design is expected to fulfill its mission. But the real world is different. I have 2 questions to the electric/electronic experts.

  1. In the case of no power blackout, can the charger operate successfully to power the DC/DC converter and to charge the battery? In other words, does continuous/non-stop output current cause the charger burn out or decrease in the battery lifetime? I think the charger is designed to cut the output current when the battery is full.
  2. In the case of power blackout, can the battery and the DC/DC converter operate successfully to power pi-zero? Does having the charger connected (with no purpose) to the circuit have a negative impact, in this case?

Pi-Zero UPS Design

#Update March 29, 2017

The linked question and comments at DIY uninterruptable power supply (UPS) for Raspberry Pi Zero with power supply selector

Improve this question
5
  • 2
    Kindly refrain from making breaking changes to your questions. You can't keep updating it to reflect information provided in answers, or the answers provided will very quickly become nonsensical and your question will become an unmanageable mess of updates. I'm rolling back your latest update, but please feel free to reformulate a new question and ask that.
    – goobering
    Commented Mar 22, 2017 at 13:23
  • Breaking changes??? There was no change, but amendment. And it was strictly related to the post title. Tired of this kind of misleading managements.
    – vaha
    Commented Mar 22, 2017 at 13:29
  • While your update was related to the original question, it introduced a completely new circuit design which you wanted an evaluation of. Consider the state of the question after 6 or 7 more rounds of receiving an answer, updating the circuit, and asking for yet another evaluation. The question now refers to a design which is completely different to the original, and the first answer makes no sense whatsoever.
    – goobering
    Commented Mar 22, 2017 at 13:34
  • 1
    I'm afraid that's not a sustainable approach here, where we require single-issue questions which have single-issue answers. You can post the revised design in a new question, and receive a new answer directly related to it.
    – goobering
    Commented Mar 22, 2017 at 13:34
  • Also, be wary of a "Power-bank" solution - very few can tolerate both charging and supplying current to a load at the same time. Possibly because then they could be called a UPS and could be priced higher - OTOH I am sure that I have heard of such devices so perhaps they are not entirely mythical after all.
    – SlySven
    Commented Mar 23, 2017 at 12:38

2 Answers 2

Reset to default
13

Absolutely DO NOT connect the battery in parallel with the load. The charger will not know if the battery is full or not and will continue to pump power both to the battery and the step-up converter.

Since the battery is still receiving power and is still charging, it will overcharge and explode.

If you want to do this properly, you should use a proper LiPo charge controller with an integrated power output (link) which will eliminate the charger+booster combo in your setup.

The module linked above is featured on Adafruit's blog (link). You can also search the internet for something similar and cheaper than that.

This is an illustration of what overcharging will do. Thankfully, I caught it expanding before it exploded.

enter image description here enter image description here enter image description here Right one is normal. The inflated one was extremely firm.

Improve this answer
8
  • thanks for your explanation. I updated my post for an alternative solution. I will be glad if you comment about it too.
    – vaha
    Commented Mar 22, 2017 at 12:27
  • 2
    If anyone is wondering why I'm very sure the battery will explode, I had it happen to me (thrice).
    – Aloha
    Commented Mar 22, 2017 at 16:12
  • 2
    @vaha What will happen depends on the battery. LiPo usually inflates then explodes, lead-acid car batteries will boil until there is no more fluid, modern AA Li-ion batteries will vent instead of explode, etc.
    – Aloha
    Commented Mar 22, 2017 at 17:16
  • 2
    For the record: all of which are bad things...!
    – SlySven
    Commented Mar 23, 2017 at 12:35
  • 3
    Come on, this isn't Philosophy SE. 8-)
    – SlySven
    Commented Mar 24, 2017 at 7:59
0

I am devising a systems that might just help you. As pointed by other answers, use a dedicated systems that both charges the battery and powers your system, just like this UPS from AliExpress.

I am using 6 li-po (2 parallel 3 series). In my case I use the UPS to generate +15V/5A for a milliohmmeter: then I generate isolated 9V/5A (for the heavy resistance measurement) and +5V/2,5A for the Pi 4 and 7" HDMI screen (via a LM2576).

It gives me 8+ hours in "standby" (just the Pi and the screen ON) or 2+ hours of measurements (9V/5A for 3 seconds every 2 minutes).

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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