Hi all,
Side project, not commercial, so cost is the main constraint. I’m trying to put a small vertical string of temperature sensors out in a freshwater lake, run a cable back to shore, and feed the readings into a hobby weather site I run. The whole thing is intentionally semi-disposable. A fishing snag or anchor drag could wreck it any year, and I’m not going to spend money armoring against things I can’t actually prevent.
Looking for feedback from people who’ve done moored sensor work before. Mostly curious what cheap and effective actually looks like at this scale, and where you’d draw the line between fine for a hobby install and cheap in a way that I’ll regret.
Setup:
- Freshwater lake, about 40 ft deep at the install point, soft silt bottom
- Roughly 850 ft cable run from a private shore structure
- Five PT100 RTDs spaced down a vertical chain to capture the temperature profile
- ESP32 + MAX31865 boards live on shore, dry and reachable
- Subsurface mooring buoy at maybe 25-30 ft of depth to stay below where people fish
What I’m thinking right now:
- Direct-burial Cat5e (24 AWG bare copper, UV-resistant jacket) for the trunk cable
- 8 conductors carrying 4 probes in 2-wire mode with a software offset to handle the cable resistance
- Probes get spliced to the trunk at their position on the chain using 3M DBR-Y gel-filled connectors
- Maybe a 50 lb mushroom anchor or a DIY concrete pour with rebar
- Short bit of galvanized chain at the abrasion zone, polyester rope above that
- Either a DIY 8" sealed PVC float or a used trawl float for buoyancy
- Two-point anchoring with a small secondary anchor to keep the chain from rotating
Things I’d love input on, in rough order of importance:
The big one: what would a minimum test build look like? Before I commit to 850 ft of cable and five probes, I’d really like to validate that the splices and the calibration approach actually hold up underwater. Could I prove the concept with a much shorter run and one or two probes first? I’d rather find out a method doesn’t work after spending $100 than after spending $500.
Flip side: what does the “upgrade” version look like? If v1 makes it a couple of years and I want to do v2 properly, what would you reach for that I’m currently skipping? I want to see what I’m trading off, partly because some of those upgrades might be cheap enough to pull forward into v1.
Cable: is direct-burial Cat5e actually fine for ~850 ft of fully submerged run on silt for a few years? I’m worried about long-term water ingress at the splice points more than I am about the jacket itself.
Splices: are gel-filled DBR-Y connectors the right call for permanent submerged splices at this scale, or is there a better answer at this price tier? I see them recommended a lot but I haven’t done one before. I could also just run Cat5 to the base, do one splice into the end and have the leads from the probes all going to one waterproof box? This is probably way easier than trying to splice in and pull out a single pair from the cable.
Anti-rotation: I’ve landed on two-point anchoring after some reading. Anyone solved chain rotation differently for a single sensor string?
Install: any sequence advice for getting this whole assembly into the water off a small boat without tangling everything? Plan is to lower the anchor on a temporary line first, pay out the chain and cables as the boat moves away, and release the float last.
I know someone’s going to say “just buy a real industrial sensor package and a proper moored station.” Totally fair answer. The whole point of this one is that it stays cheap and gets built on a weekend by one person, so I’m looking for the cheap version of “right.”
I’m also not sure on the sensors, if there’s a better solution out there I’m all ears. This is just kind of cobbled together from a weekend of Google searches.
Happy to write up results and post photos once it’s in the water. Thanks for any input.