Battery powering

[flopp]

@Jason-Brunk
You can try with a cap on main pwr/gnd.

Maybe you can measure the voltage on main pwr/gnd and see what happens when it starts?

[CrankyCoder]

@flopp i changed batteries and made sure i don't go below 2.6v and still the furthest i get is

TSP▒Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP▒Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK

Looks like it is locking up as soon as the radio is initialized.

[flopp]

@Jason-Brunk
Do you have an other NRF, if you have try with that

[CrankyCoder]

yeah, this one works if i give it a 3.3v power source. But i can try my other one.

[Yveaux]

@Jason-Brunk Do you know/did you change the brown-out level of your atmega? See BODLEVEL at e.g. http://cryoarchive.net/tutorials/avr-tutorials/avr-fuses-explained/
What Arduino hardware are we talking about?

[flopp]

This post is deleted!

[CrankyCoder]

Im using a 3.3v pro mini.

From what I gathered BOD level only affected the power down. Would it affect the power up and radio initialization?

[flopp]

@Jason-Brunk said:

yeah, this one works if i give it a 3.3v power source. But i can try my other one.

Do you mean NRF?
NRF should work down to 1,9 volt

[CrankyCoder]

@flopp That is correct, everything registers ok and debug shows the connections when i give it 3.3v from my usb power supply

[flopp]

@Jason-Brunk said:

@flopp That is correct, everything registers ok and debug shows the connections when i give it 3.3v from my usb power supply

Then it can be bad battery.

Measure battery, should be about 3.1.

Connect something to the battery, like an LED, battery shall now show maybe 2.9, not lower. If it shows lower I guess the battery is almost empty.

[Yveaux]

@Jason-Brunk said:

Im using a 3.3v pro mini.

How are you pwering it? Through RAW or VCC? Did you remove/disable the regulator?

[CrankyCoder]

@Yveaux VCC. Yeah i pulled the regulator off

[Yveaux]

@Jason-Brunk Does it run stable without the nRF? E.g. disconnect nRF and simply print something to serial.

[CrankyCoder]

tried to do the BOD fuse update. bricked a pro mini. grabbed another one. it took the updated boot loader.

if i disconnect the radio it just loops

TSM:PDT
TSM:INIT
!TSM:RADIO:FAIL
!TSM:FAILURE
TSM:PDT
TSM:INIT
!TSM:RADIO:FAIL
!TSM:FAILURE
TSM:PDT
TSM:INIT
!TSM:RADIO:FAIL
!TSM:FAILURE
TSM:PDT

as you would expect.

If i hook the radio up i get this.

Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP:Á

I do still need to remove the regulator and led on this one. Not sure if that's the issue. I have also tried a brand new cr2032 battery

[CrankyCoder]

So here is my latest. I modified the arduino pro mini. Changed the BOD fuse value and reburned the bootloader. I wired it all back up. Got it to register. I got 6 button presses where it sent the data. Now I am right back where it was.

Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP:ASSIGNID:OK (ID=1)
TSM:FPAR
Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP:ASSIGNID:OK (ID=1)
TSM:FPAR
TSP:MSG:SEND 1-1-255-255 s=255,c=3,t=7,pt=0,l=0,sg=0,ft=0,st=bc:
Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP:ASSIGNID:Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK
TSP:ASS▒Starting sensor (RNNNA-, 2.0.0)
TSM:INIT
TSM:RADIO:OK

I am just not having much luck getting this to work via battery. I have a 100u capacitor on the radio.

If anyone has any thoughts I am all ears. I kind of at a loss now. I am using the 2.0.0 release.

[CrankyCoder]

More testing update:

I tried 2 AA batteries. Which would be the same voltage, just more mah. 0 Issues. Using the battery monitor with the AA batteries it shows the voltage at about 2.84v

When I hook a brand new cr2032 up though, I press the button and it shows voltage is already down to 2.48v. I have the BOD disabled (I believe) per the instructions to modify the boards.txt and reburn the bootloader.

The batteries I have are energizer cr2032. I would think they are ok, but they could all be junk. I have tried 4 so far.

[m26872]

@Jason-Brunk I would try to put in a delay of the radio startup in the sketch. Maybe there's to much or fast simultaneous power consuption initially with respect to the coincell.

[CrankyCoder]

Another quick update, seems to be making progress.

I decided to try to flash a 1mhz boot loader onto my pro mini. Then reuploaded my switch sketch. So far it's looking good. Button presses are registering as expected. 2 things I have seen.

1. i think i need to read up some more on changing clock speed. My sleep command should wake up and report battery level once an hour. This did not happen at all through the night.
2. now that I am running at 1mhz when i tried to connect via serial for debugging using the default baudrate it was junk. I forced the baud rate to nice and slow, and it looked like that was going to do it. but no go.

[Nca78]

Hello I gave you some tips for button cell on your other thread.
In addition: if you are using interrupts only then your power consumption should be extremely low, I don't think it's necessary to wake up regularly to send battery level.
If you don't do it then then you have even lower consumption because the watchdog timer is turned off, my CR2032 door/windows nodes using the sleep method from MySensors draws less than 2uA, there can be no visible change in battery level except if you leave your node in a closet for one year ;)