Powering up arduino

Hello everyone!
I am currently working on a small camera lightmeter based on an arduino nano. The board is connected to a 4 pin 0.91" oled display and to a light sensor via I2C, and is powered up by a 3.7v LiPo battery, charged via a charging module.

My issue is that when I try to turn on the arduino with it does not seem to work, but if I plug in the USB cable and remove it, the battery is able to keep it up for a long time.

To mee it seem like the small 3.7v battery is powerful enough to keep the arduino running, but not powerful enough to turn it on.

Can anyone help me? I am sure that the battery is fully charged as I left it for 4 hours plugged to the current. Also everything else seem to be working fine, so I don't know what could be wrong.

Is the Nano a 5V 16MHz version? I think that you are right in that the battery does not have enough voltage to start the processor but enough to run it once started with the, higher, USB voltage. You may need a 3.3V board to run with the battery.

That voltage is outside of the spec for a 328 processor running at 16MHz .if you drop the clock rate to 8MHz and check the BOD voltages , then it will be fine.

hammy:
That voltage is outside of the spec for a 328 processor running at 16MHz .if you drop the clock rate to 8MHz and check the BOD voltages , then it will be fine.

Correct me if I am wrong (as it is the first time I am doing this). To speed down the clock rate I should open the board.txt file and change

"nano.build.f_cpu=16000000L" into
"nano.build.f_cpu=8000000L"

As for the BOD voltages what does it mean? How can I check it? Could you please explain me how to do this?

shevaar:
As for the BOD voltages what does it mean?

Google brown out detection.

“nano.build.f_cpu=8000000L”

I am not sure that that would make a difference in terms of the power required. I think that the processor is still running at 16MHz.

You need to reflash the bootloader to get it to run at 8MHz so it requires an ISP programmer like a USBASP or similar. Here is a page that documents the issues.

I did a search for 3.3v nano an it turns out that you can’t just burn an 8MHz bootloader to the Nano, there are other changes necessary.

Here is one page on converting a Nano clone to 3.3V.

I think that I would just find a 3.3V board.

You did not tell us HOW you have connected the LiPo battery to the Nano.

The 5volt supply will probably be about a volt lower than the battery when connected to V-in,
because of the drop-out voltage of the LM1117 5volt regulator.

You could connect the battery directly to the 5volt pin,
but then the LiPo will get ~4.7volt when you connect the Nano to USB.
And that will eventually result in smoke and fire.
Leo..

Wawa:
The 5volt supply will probably be about a volt lower than the battery when connected to V-in,
because of the drop-out voltage of the LM1117 5volt regulator.

If it is connected to the 5V regulator you will get nothing out. The input has to be the drop out voltage above 5V.

windoze_killa:
If it is connected to the 5V regulator you will get nothing out. The input has to be the drop out voltage above 5V.

Uh, no. The dropout voltage is not where the output goes to zero, it’s the point where they no longer regulate the output voltage. The exact verbage of the LM1117 datasheet:

The dropout voltage is the input/output differential at which the circuit ceases to regulate against further reduction in input voltage. It is measured when the output voltage has dropped 100 mV from the nominal value obtained at VIN = VOUT + 1.5 V.

Wawa:
You did not tell us HOW you have connected the LiPo battery to the Nano.

The battery is indeed connected to the 5v pin through a battery charging module.

Quick update. I tried to connect to the arduino (at the 5v pin) a 1.5v battery. Now I have that the pow led on the arduino is on at all time, even though the arduino only turns on when i use both my 3.7v and my 1.5v batteries.

I find this a rather unelegant solution, but for now it works, so I wanted to ask you guys:

If I leave the 1.5v battery always attached to the 5v and GND pins, how fast will it discharge? Is it a solution worth considering or you guys think that I will have to change battery every other day?

I think that 1.5v are barely enough to power up the POW led, which does not fully turn on, and thus it seem to me like it could stay on "forever", but I have no experience in this. Do you think it would be better to add a second switch that is used only to turn on the arduino, and then let it be powered with the 3.7v battery?

I'm not clear about your setup. Where are you connecting USB power? To the charger? Directly to the Nano? Both? Could you post a drawing of your circuit?

If your display and sensor work ok on the battery, then the solution is to either replace the Nano with something that runs on the battery at 8 MHz, or change the bootloader and fuses on the Nano so it runs on the internal 8MHz oscillator. A 3.3V 8MHz Pro Mini might be an option for a replacement. It's the same as a Nano except it has an 8 MHz resonator and does not have a USB connector or USB-to-serial adapter. So you would need an FTDI adapter to program it and communicate via serial. You would connect the battery directly to the Pro Mini Vcc pin.

As for redefining the Nano, I believe there is a board definition for a breadboard stand-alone 328P-PU running on the internal oscillator. It might be possible to flash the bootloader and fuse settings for that board to your Nano. You might have to calibrate the internal oscillator if accurate time is important.

But I'm still unclear about your charger setup. Are you charging the battery and running your device at the same time?

Hi,
How many mAh is your Lipo?
Can you post a picture of your project and a schematic please?

Thanks.. Tom... :slight_smile: