Burrowing Owl Relocation Sound Box

Hi @papajoe0418,

  I hope it isn't a problem, but check current consumption when it isn't playing ... it will be lower than when it is playing, but it may drain too much current to be left connected 24 x 7.

Of course, you might be lucky!

Best wishes, Dave

@papajoe0418 

Looks neat. Well done.

Are using little solar cells an option?

I read they also use fake poop to make the relocated owls feel at home 🙂

Thought this was interesting as I had not heard it before.

"Some rabbits growl like large carnivores; some birds hiss like snakes; and burrowing owls produce a rattling sound to mimic a rattlesnake."

@robotbuilder Yes! We're planning on splattering non-toxic white latex paint around the area. I hope this all works.

I don't know about the solar cells because it's a nighttime device, and I don't want the added cost of storage battery.

@davee Yes, I was planning on hooking up an ammeter but my VOM is acting up. So I ordered a new one tonight. I was measuring voltage instead and can see it trickling away hour by hour. Testing overnight to see what the total damage is. After I get the new VOM, I can experiment in earnest with different battery (3.7 lith-ion) and methods of deep sleep mode. I'm not a fan of external RTC, just because I prefer to keep components down to a minimum. We'll see.

Posted by: @papajoe0418

↑

I don't know about the solar cells because it's a nighttime device, and I don't want the added cost of storage battery.

If you're going to be using a 3.7V li-ion battery anyway, there's no reason that you can't use solar to charge it up during the daylight to extend or even permanently power your hooter.

@will Yeah, that would be nice...especially if we get to put them on larger projects that may run for weeks. I do have a solar cell I can experiment with. I was thinking I would be hiding this in the bushes, so the amount of sun is iffy. Thanks, I will put it on my to-do list.

While I continue to wait for my new multimeter (so I can measure amperes), I'm establishing a power consumption benchmark, of sorts. Using three Duracell AA batteries, I set the device on continuous loop for approximately 72 hours (and counting). Voltage reading started at 4.85v and is now (3 days later) at 3.70v. The sound module continues to work without issue. 

The test is greatly accelerated ("burn-in") in that the audio is firing off many more times than it will in the field. In total, the test is firing off 480 15-second "hoots" in a 24-hour period, for a total of 2 full hours of audio. In the field, I expect a total of ten 60-second "hoots" in a 24-hour period, for a total of 10 minutes of audio. Audio volume is set approximately the same as it would be in the field, but power consumption at the audio device ought to be 92% less (10 minutes vs 2 hours).

The total sleep time difference is negligible (22 hours vs 23 hours and 50 minutes). However, if a more efficient sleep mode is possible, then that saving would be realized over at least 22 hours out of every 24!

When the ammeter arrives, I'll redo the experiment with different types of batteries. (Note that if I gift the device to the owl relocation organization, they are more likely to have access to AA batteries than to li-ions or a recharger.) I will also spend some time looking at software solutions for deep sleep mode.

@davee I got my new multimeter and started measuring current. I can verify that sleep mode current is indeed 1.3mA. When the sound player is playing, the current only goes up to 1.8mA. My readings were taken inline with Pico VSYS & GND (pins 39 and 38).

I managed to perform two tests where the AA batteries lasted a little over 3 full days before the player got wonky. Voltage at that point was around 3v (out of 4.85 when batteries were fresh). Based on all of this, I'm off to sit down with pen and pad to calculate the potential savings if I add an RTC, which seems to me a better solution than C code.

I also printed up some nice custom boxes to insert into my Burrowing Owl field container (a whey protein cannister). 😀 

Hi @papajoe0418,

  Well done with your progress.

  It looks like sleep mode uses nearly as much current as when you are hooting! Clearly dormant mode is worth looking into.

  You mention maximum of 10 hoots per 24 hours, so I am guessing about an hour between each 'possible' hoot . On this basis ...

  I don't know if you can find an RTC to give a pulse every so often ... say once an hour ... but if you can then it could wake up the Pico ... which in a few milliseconds could check the time .. then either go back to sleep immediately or  do a hoot and then go back to sleep.

If not,  maybe use a 'crude' timing circuit to do the 'hourly' wake up, then follow the same path.

You could look to the lowest current types of 555 for this 'crude timing' purpose, though the quick Google I did suggested this might use 0.2-0.5 mA.

Alternately, maybe look to a more modern and specialised chip ... e.g. https://www.ti.com/lit/ds/symlink/tpl5111.pdf which claims to be nearer 35 nA !

Obviously, these are only suggestions to consider, and you probably have much better ideas yourself.

Good luck and best wishes, Dave

@papajoe0418

the rpi pico has on on board RTC, it may be suitable for your requirements.  See the docs

https://docs.micropython.org/en/latest/rp2/quickref.html

I'm not sure just how deep a sleep you can set on the pico I've not played with that aspect as yet but this doc may be of interest

https://ghubcoder.github.io/posts/deep-sleeping-the-pico-micropython/

@davee @papajoe0418 FYI @byron A separate RTC has the advantage of its own multi-year battery. With it, you can use deep_sleep (not sure of the name but that's what esp calls it). When I am well I will be looking into the RTC + PICO as my game camera solution unless the 2024 Pi5 ZERO is better (just my guess at a future product)

I have a photography device that is used for a camera timer/trigger, it NEVER turns off and the coin battery is replaced every 2 years. It even has a tiny display. Those CR2032's are great!

@davee @papajoe0418 @zander,

A separate RTC could also have the benefit of more accurate timekeeping where an accurate timestamp is required, as the rtc on a pico has to be seeded with the start time and will loose some accuracy as the days pass by.  However if a timestamp is not important, then the deciding  factor is how deep a sleep can you put the pico and and how long will the pico battery last.  I'm no expert in deep sleeping but I doubt if the pico rtc compared to an external rtc will have any bearing on the life of the pico battery.

One of the best boards for deep sleep is possibly the esp32 boards made by the Unexpected Maker who designs his boards for the best low power consumption during a deep sleep.  

Hi @byron,   @papajoe0418, and Ron @zander,

  The miniscule Googling I did suggested that the Pico was unable to wake itself from dormant mode ... of course that 'information' could be flawed.

If you use an external RTC, instead of using 'dormant' Pico, you could use a FET to control the Pico's power line ... so that it boots every time the RTC creates a 'wake up' .. and the Pico uses no current at all during the 'off' periods.

The common DS1307 doesn't seem to include an alarm, but it's newer sister does.. see reference if you are interested.

https://www.analog.com/media/en/technical-documentation/data-sheets/DS3231.pdf

Thus, each time  the Pico powers up, it finishes the session by setting this next alarm time and powers itself off.

Sorry if I am encouraging 'scope creep', but I am thinking it might be better to spend some effort on the design, as it could pay for itself in reduced battery costs.

Hope this is useful, and not too confusing.

Best wishes, Dave

ps Bearing in mind the number of changes, I was holding off from mentioning the Pico might not be the only processor to consider, but as @byron has mentioned another one, I agree it is a possibility. However, all processors take very little power when their power supply is disconnected, so it might be a moot point!

@byron I did some digging and it appears that the PICO sleep code is unstable for now. In the perfect world using an external RTC allows the attached board to go into the lowest power level it can thus saving the board battery. The RTC boards can go for years.

@byron @davee @papajoe0418 As I said earlier, the PICO sleep code is still unstable. Maybe a switch to the esp32 with it's deep sleep mode and an external RTC to wake it will give many days perhaps even weeks of battery life.