What Hardware Are You Using?
You can find a full hardware list here, but generally speaking, I am using "off-the-shelf" components. I am not an electrical engineer and barely a programmer so I kept things simple. The major components include a Raspberry Pi Pico W, a couple DS18B20 temperature sensors, Cat6 ethernet cable, various power packs, some antennas, an IP68 enclosure and a 3D printed buoy.Why a Rasbperry Pi Pico Instead of Arduino or [Other Option]?
I've chosen the Rasbperry Pi Pico because it is a versatile microcontroller with plenty of aftermarket adapters, a strong community, and I wanted to learn how to use one (a selfish reason, I admit). Plus, it is programmable with MicroPython, which is very similar to programming languages I already know, like Perl and PHP. Yes, it draws more power than an Arduino or other customer-built electronics, but as I mentioned, I'm not an electrical engineer. So far, early testing shows the Pico works perfectly for this application. We are not aiming for years of battery life - just a "hands-off" period of at least 30 days up to 6 months. Larger batteries and small solar panels can easily be incorporated into the design.How Are You Measuring Temperature Underwater?
The DS18B20 temperature sensor uses the 1-Wire protocol that is natively supported by the Pico and MicroPython. An added benefit is that you can 'daisy-chain' multiple sensors on a single wire (within reason), making it easier to construct the temperature probe array. This also allows me to use pre-built cables, such as "waterproof" shielded Cat-6, by tapping into 2 of the 8 wires (+ ground). I can seal the connections from the elements with heat-shrink "waterproof" tubing. Another benefit is that we can use standard "waterproof" RJ45 connections at the buoy/box end of the cable array, making it easy to 'plug' and 'unplug' the transmission box from the deployed buoy+temperature cable for maintenance.What about power consumption and battery life?
So far, this is the most important item to figure out because when you are in the middle of an open water area you have no access to grid power so you will be relying upon batteries and maybe solar panels. The Pico W has the following power consumption measurements with one transmission per day:| Situation | Draw | 24hr Need | 18650 life | Solar Panel Needed |
|---|---|---|---|---|
| WiFi On | ~60mAh | ~1500mAh | ~1 day | ~3W |
| WiFi Off | ~18mAh | ~500mAh | ~5 days | ~1W |
| Sleep (PowerControl) | ~8mAh | ~200mAh | ~12 days | ~0.5W |
| Sleep (+USB off) | ~5.6mAh | ~140mAh | ~14 days | ~0.5W |
| Hardware On/Off | ~0 mAh | ~1mAh | ~100 days | 0.1W |
What about [hot/cold] temperatures?
I have not done extensive testing, but my main concern is heat. The device won't be deployed during winter months or submerged at depth, cold temperatures. It will, however, be inside a waterproof container and/or buoy. My hope is that because the electronics will only "power up" for transmission once or twice a day, the interior temperature of the buoy/box will not be excessive. Time will tell.How are you transmitting the data back to the Internet?
I have extensive experience in transmitting, collecting, aggregating and displaying various data types. The challenging part is not the data payload but creating a reliable transmission medium be that WiFi, LoRa, cellular or satellite. The initial prototype is being built with WiFi, and eventually, I plan to experiment with products from Blues for the cellular and satellite needs.For most deployments outside of Lake Michigan, I anticipate that the transmission buoy will be within WiFi range of land. If necessary, I could deploy an AC-powered WiFi repeater on shore or a battery/solar WiFi repeater using another Raspberry Pi hardware package or something "low cost" off the shelf. In cases where WiFi is not available, then cellular service is almost always guaranteed at these locations. However, deployments in Lake Michigan will almost certainly require satellite service, as cellular service is marginal unless deployed close to shore.
How come you are not using LoRa or [network blah]
LoRa is great for long-range, low-power communication when WiFi or cellular options are not available. In an IoT setup across open ground, the distances that can be reasonably expected without custom antenna packages are generally:| Transmission | Distance | Cost |
|---|---|---|
| WiFi | 1 mile | Free |
| Cellular | 15 Miles | $0.01 per transmission |
| LoRa | 10 Miles | Free |
| Satellite | Unlimited | $0.05 per transmission |
How many temperature probes and at what depths?
I would eventually like to support deployments of the bottom temperature probe to around 100 feet. Anything deeper places the device at a depth that significantly limits the time a recreational scuba diver has for deploying or removing it (~10min of bottom time). In my opinion, measuring beyond that depth is better left to an in-situ gauge deployment that is already covered by the 'standard' Midwest Underwater Explorers - Project Baseline deployment.I would additionally like to be able to place a temperature sensor every 10 feet in an attempt to identify thermoclines. In theory, this is completely reasonable but I have not put it into practice yet.