I use the 3.3V output of Nano to power a 3V device that, once it is initialized after power-up, can remember its initialized status for a power interruption up to 10 seconds. I want to back that up with a lithium battery that is always on standby charge so it will let the device survive power interruptions of 8 to 10 hours or longer. I'm confused by all the talk about safety when recharging various types of lithium batteries, warnings against trickle charging, over charging, etc. Anyone know of a reliable and safe way to handle this?
Lithium based batteries do not tolerate this scenario. NiMH or NiCd batteries do, using "trickle charging", but their lifetime is limited to a couple of years.
Also consider supercaps.
That is why you should use a chip specifically designed for the purpose - don't try to "home-brew" it.
See above.
jremington:
Your supercaps suggestion is intriguing, since the 3V device only draws about 0.5uA on standby. Are there supercaps special-made for low voltages like that?
How much current does the 3V device draw?
Sure:
https://www.mouser.co.uk/c/passive-components/capacitors/supercapacitors-ultracapacitors/
EDIT
trying that mouser link again: https://www.mouser.co.uk/c/passive-components/capacitors/supercapacitors-ultracapacitors/
What is the device you want to backup?
The formula for the discharge rate of a capacitor is (deltaV/deltaT) = I/C, so a 1 Farad, 3V supercap could power that device for 100000 seconds (27 hours), with a voltage drop of about 0.5uA*10^5 s = 0.05V.
STRICTLY obey the voltage rating of supercaps! You might be interested in this solar powered, supercap sensor project: Gammon Forum : Electronics : Microprocessors : Solar powered Arduino
jremington:
I'm at risk for over-simplifying this. Digikey has a 100F 3.8V supercap (part # 4688-LIC1030Q3R8107-ND) for a little under 5 bucks. If I just hang this across the 3.3V output of the Nano it should keep my device's memory alive during an overnight power outage. But is more needed to get past the initial powerup? For example, will the Nano 3.3V regulator shut down the first time power is turned on because it thinks it sees a short circuit? And if it does get past that, when power to the Nano fails, will the supercap reverse discharge through the Nano 3.3V regulator?
That is a 100x overkill for the sensor, but now you seem to be wanting to power the Nano. To do that you first need to know the maximum and average current draw of the system you wish to run on backup power.
Please post the details of your system, including the exact product descriptions (e.g. which Nano?).
jremington:
I must have said something wrong. I'm using anybody's Nano, especially the clones with CH340 chips. The Nano supplies 3.3V to my device. The Nano itself gets its power from the USB cable plugged into a PC. When the PC is switched off overnight the Nano loses power, and that in turn causes loss of its 3.3V output to my device. There's no problem with power on and off to the Nano itself. It will resume operating normally when power is restored. But losing power to the device will cause it to lose its memory, requiring that it be reset manually. It's the need for that reset I want to avoid by sustaining the 3.3V output from the Nano.
I'm not sure what you mean by "which Nano". That's an open design product that is replicated by many manufacturers. As for the "device" that uses the 3.3V power, it's a GE model 33701 Universal Remote Control, if that makes any difference.
What is the usual way of powering the remote, batteries?
Arduino sells about half a dozen different Nanos (Classic Nano, Nano 33 BLE Sense, Nano ESP32, Nano Every etc.). You evidently have an imitation of the Classic Nano.
To back up the gizmo, a 1 F supercap will be more than enough. A Schottky diode between the Nano 3.3V output and the gizmo+supercap will act as a switch.
However, with this setup, you run the risk of unreliable communication. The 5V Classic Nano is not guaranteed to recognize a 3.0V signal as a HIGH or logic 1. A 3.3V processor would be a much better choice to interface with the gizmo.
Why wouldn't you power the remote with batteries all the time? The Nano could come and go, and it would never notice. And if the remote only uses 0.5uA unless you're pressing a button, two AAs would probably last for years.
However, I didn't know that a universal remote looses all its settings when you change batteries. Are you sure about that?
Edit: The 3.3V output of a Nano clone is actually provided by the CH340, and I think you only get 40mA or so. That's not enough to drive the LED on the remote unless you are very close to the receiver. Could you possibly provide more information about what you're trying to do?
Gentlemen:
The sytem I'm designing must be able to silence a TV under certain conditions by sending it an infrared MUTE command. I'm trying to make it very easy for anyone to use it. That means it has to operate with any brand and model TV, and there are hundreds of different IR protocols. I have used three different methods of accomplishing that, and I'm trying to select one of them. Method 1 cranks out the IR signal using Arduino code. That takes around sixty lines of code. There's no problem with losing power to the Nano, but there's also no way to expect non-technical users to find changing a bunch of code to a different bunch and uploading it with IDE is easy. Method 2 uses a learner type remote control. All the user has to do is point his regular TV remote at the learner and press a few buttons to teach the learner how to mute/unmute his TV. That's I would call "fairly easy", and there's no problem with power interruption because the learner remote saves everything in an EEPROM. Then there's Method 3, my favorite, which uses a universal remote. This type remote comes with a code book that lists literally hundreds of consumer products, each with a 4-digit code stored in the processor's memory. All the user has to do is find his TV in the codebook, and key in the code. Now that's what I call easy. If only the designers of the universal remote had found four bytes of non-volatile memory to save the last code entered. They must have thought it was so easy to program, it wasn't worth a few more pennies to save it. After all, as ShermanP says, the batteries would probably last for years.
Thanks, ShermanP, for convincing me I am trying too hard to worry about losing the TV configuration during power outage. I'm going back to battery operation.
Did you know about the TV-B-Gone? It turns just about any TV off, so all you would need to do is change the codes (in flash memory) to mute instead.
Arduino based!
1 Like
Sorry I forgot to answer your question. On page 3 of the excellent instruction booklet that comes with my GE Universal Remote are two short paragraphs.
Battery Saver
Your remote automatically turns off if the buttons are depressed more than 8 seconds. This will save your batteries should your remote get stuck in a place where the buttons remain depressed (e.g.,between the sofa cushions).
Code Saver
You have up to 10 minutes to change the batteries in your remote without losing the codes you have programmed.
That's what I call "user friendly".
A classic Nano comes with 1024 bytes of eeprom.
And then there is non-volatile FRAM.
Leo..
My laptops and Tivoli radios have been plugged in for decades and my house is still standing. Lithium is not the boogieman.
So is the Nano going to trigger the remote to send the Mute command? How will it do that?
I'm not sure it would be reliable enough, or work for every model TV, but the Nano could be programmed to be your learning remote. With an IR receiver, it could receive the IR output of the remote when the user presses the Mute button, then store that raw code in eeprom. Then you wouldn't need the universal remote at all. Of course you'd have to rig up an IR transmit circuit too, but that's pretty simple.
If your device will mute the TV automatically during commercials, I will buy one.