Beginner building a solar-assisted ESP32 buoy (temp + pH) -- design review & major concerns

Projects General Guidance
1

Hi everyone,
I’m a Grade 12 student and a complete beginner in electronics. For my school project I want to build a solar-assisted buoy with an ESP32, DS18B20 temperature sensor, and SEN0161 pH sensor. It should measure every ~10 minutes and run safely for days/weeks on a LiFePO4 battery with small solar panels.

I adapted my design from a RandomNerdTutorials solar ESP32 project (which uses Li-ion + TP4056). Since I want to use LiFePO4 for safety in water, here’s my plan:

  • Battery: 1S LiFePO4 (3300–6600 mAh)
  • Solar: 2 * 5–6 V, ~1.2 W in parallel
  • Charger: CN3058E or TP5000 (LiFePO₄ cutoff ~3.6 V)
  • Regulator: RT9193 LDO -> 3.3 V (or maybe buck-boost?)
  • Fuse + capacitors for safety/stability
  • Sensors: DS18B20 (OK at 3.3 V), SEN0161 (unsure if 3.3 V compatible)

My main beginner questions:

  1. Are CN3058E/TP5000 modules reliably safe for LiFePO₄ charging, or are there better beginner-friendly options?
  2. Should I use an LDO or a buck-boost regulator so the ESP32 doesn’t brown out as the battery discharges?
  3. Can the SEN0161 pH board run properly at 3.3 V, or do I need 5 V + level shifting?
  4. Am I overestimating battery life with deep sleep (~3–5 mA average)? How much margin should I plan for?
  5. Does my exact ESP32 devkit board matter for power draw in deep sleep (USB chip, LEDs, etc.)?
  6. For the SEN0161, should I post the exact board link/product? Do different versions behave differently?
  7. For the solar panels, do I need to worry about panel Voc vs charger input, or is parallel wiring safe for small 5–6 V panels?
  8. I’m in the Philippines -- about 5 sun hours/day. Is 2.4 W of panels overkill or realistic for cloudy days?
  9. For Wi-Fi: is it better to transmit every 10 min, or log data and send less often to save energy?
  10. Any advice for waterproofing and safe battery enclosure (venting, sealing) for long deployment in water?

I’ve kept this short -- I also have a much longer write-up document with calculations, wiring plan, and failure concerns if anyone’s interested. Thanks so much for any guidance!

2

At point 4:
Some MCU's can use almost no current in deep sleep.
But: the switching power supplies may have significant current, even without load. They may need a miniumum load for stable opearation...

In general: a pH probe needs regular cleaning and calibration. In the lab they ar calibrated every day... how will you compensate for sensor drift over a 2 week period?

At point 8: on cloudy days your panels may yield 10 times less power than the max rating...

At 9: larger chunks of data will save energy. Your data chunks will be very small. Over 10 minutes it may be less than 100 bytes. The sending of the protocol may take more... (message header, adresses, handshake, confirmation, etc.)

3

Thanks a lot for the insights -- that really helps!

Just to clarify, I haven’t bought any parts yet and I don’t have prior experience -- this is all still research.

  1. Since my LiFePO₄ is ~3.2–3.6 V, is a low-Iq LDO (like RT9193) simpler/safer than a buck/boost for a beginner?

  2. I didn’t realize pH probes need daily calibration -- thanks for pointing that out. For this 3-week prototype I’ll note that as a limitation, but would proper rinsing/storage at least slow down drift?

  3. If cloudy days cut my 2.4 W panels down a lot, would a 6600 mAh LiFePO₄ still be enough buffer for a short project?

  4. Would sending Wi-Fi data once per hour (vs. every 10 min) noticeably extend runtime?

Really appreciate your calibration point especially -- I hadn’t thought of that.

4

what wireless technology are you using?
what is distance between buoy and shore station?

5

Thank you! I haven’t researched connectivity in depth yet, but my current plan was to try the ESP32’s built-in Wi-Fi.

For testing, the buoy would be about 10–30 meters from shore. Do you think it makes more sense to connect through a regular Wi-Fi router on shore, or would using a phone’s hotspot nearby be more practical?

6

try it - it may work over that distance
otherwise look at LoRa

7

Few notes,
if you connect solar panels to regular battery charger module, the efficiency is really bad. You need solar charger circuit with (pseudo) MPPT. And it's hard to find one for lifepo4.

Lifepo4 battery voltage is 3-3.6V (3.2V nominal). Which is within specs of Esp32 operating voltage and for best efficiency you connect it directly to esp 3.3V

There are also esp32 boards built for solar, but they use li-ion/lipo battery.
Like ESP32-S3 PowerFeather (I have never used though).

If you need boost converter for your sensor, pick one that has enable pin to shut it down when not used for sensor.

8

If your design is not safe for a Li-ion battery then it won't be safe for an LiFePO4 either.

9

As far as I can see, you haven't mentioned what sort of water you are hoping to operate in.

You mention a buoy, but is it in a lake, a river, open ocean? From shore, I'm assuming this is coastal saline water.

I've done a lot of deep sea investigations in the past, including in situ buoys and submersible sondes operating in oceans in the middle of nowhere.

It's a hostile environment. The deep water sondes would only expose sensitive instruments as and when necessary. This would mean periodic measurements and withdrawing sensors into protection when not using them. This had the added advantage of keeping them clean and calibrating automatically.

Fouling has been mentioned, a problem in most environmental waters. It won't affect temperature, but pH could be a problem.

Sea water is corrosive, so everything is going to have to be suitable. For long term use, metal fittings need to be marine grade stainless steel and any enclosures, IP 68 minimum, and that means everywhere moisture can get in.

In seawater, you are working in a conductive electrolyte.

For pH readings to be accurate, you need to calibrate in saline standards, at least two. Don't expect to calibrate in a lab with normal buffer solutions and hope it will work in seawater.

It's an ambitious project, doable, but you might need some more research first. It's hard enough dangling a probe into a freshwater lake or river, but if it is open coastal water, keeping it working is going to be a major challenge.

10

Thanks for the tips! My plan is solar-assisted, not purely solar-powered. I also want the option to charge from a wall supply.

I chose LiFePO₄ for safety in water, but as a beginner should I stick to the tutorial and revert to using Li-ion + TP4056 for simplicity? My main worry is battery safety -- I want to avoid leaks or explosions. Any beginner-friendly advice for keeping it safe would be nice!

11

That makes sense, thanks. I’m new to batteries -- do you have any tips for making a safe setup in water, or would using Li-ion + TP4056 actually be simpler without increasing risk?

12

Probably best to do so but don't assume that the tutorial is correct.

13

safe for who, the fish?

14

For us, lol. I guess for the fish too. We just don't want batteries exploding since we'll be working with water.

15

Thank you so much for the detailed advice. I really appreciate your perspective. Our project will be freshwater only, mostly in waterways and small bodies of water. We’ll be working in a team with just three weeks, so we’re aiming for a short-term prototype rather than a long-term deployment.

Given that, do you have any tips for keeping the electronics safe and functional in freshwater, even for just a few weeks? Are there simple ways to reduce fouling or protect the sensors without spending too much time and resources on complicated enclosures?

16

Well you need to be careful. You can't let any of the electronics, wires or battery get wet. You won't need water until you are ready to test the pH probe, so there is a lot you can do without involving water. You can even use a pot to simulate the pH probe during testing

18

We will definitely keep the electronics and battery dry. One thing I’m a bit worried about is condensation inside the enclosure once it’s deployed -- even if no water leaks in, humidity might build up over time. Our team hasn’t done much hands-on engineering yet, so we’re trying to figure out simple ways to minimize that without overcomplicating the design. Do you have any beginner-friendly tips for keeping electronics dry inside a sealed enclosure?

19

That doesn't resolve the fact that the efficiency is really bad, especially with small panel.
So if you switch to li-ion, find solar charger. There are lot of them around.
If you stay with lifepo4 and regular charger you have advantage of voltage conversion free power for esp.

20

You can spend big money on IP rated enclosures that are 100% water proof. I have also seen someone here on the forum that put everything in a plastic tupperware container and used silicon caulk to seal the places where wires came out

21

Depending on the polution in the water, you might get reasonable readings for a week or so (without calibration). In our lab we do pH values with 2 decimals... that accuracy cannot be reached without daily calibration. For you, one decimal may be sufficient...
Keep the electrode in a dark spot. That will prevent growth of algae...