A Digital Beehive: A scientific study proposal.

GoForSmoke:
This is where my understanding of Dallas 1-wire is not so good. AFAIK the 1-wire library only reads on 1 pin but I don't see why an AVR can't keep up with a few such wires. If it's 1:1 then go with Tiny AVR's, I bought 10 years ago from Mouser for $1.30-some each.

I think for at most we'd be using 160 sensors (two boxes, 80 per box) that 1 arduino can probably be setup to query all those sensors. We only need a temp reading every 10 minutes or so per, at most 5 minutes, I think that even with that number of sensors it can read them all out.

The DS18B20 ID of the 1st sensor in the frame is not good enough?

Yea, I wasn't thinking that the chips had an ID that you can get out of them, now that I know they can that's more than sufficent to tell the frames apart.

Arduino boards are development boards. UNO costs about $25. UNO chip costs about $2.50. The board has the USB chip and USB jack and power regulator, clock crystals and connectors that you don't need as long as you can supply regulated power, solid ground, be happy at 8 MHz or less and not want USB.
The chip is a complete self-contained small computer. Load the code and wire it up, that's all it takes. Add what you need, in-deed.

If i have a good wiring diagram I can probably translate that into a custom board for this, instead of buying prebuilt arduinos. But the cost of a prebuilt one is probably not bad, that and they're likely to be more reliable than one with my noob skills could make, and it's easy to replace if it fails since buying another is trivial. We're looking to make sure it's easy to replace a failed piece of hardware, so off-the-shelf stuff would be preferred over custom stuff, custom stuff will need to have extras made initially so they're ready to be swapped if something fails.

You can add an SD card to each hive and swap out chips as you do the physical inspections you will be doing anyway. I get SD modules from LC Tech in China, below costs I see on eBay. If you're okay about soldering, a micro-SD to full SD adapter (the thing they usually sell with a micro-SD chip so it can fit & work in a full size slot) can be turned into a module with a few resistors for voltage dividers.. that you only need for 5V operation. If you run your AVR's at 3.3V (I think you can at 8 MHz, know you can't at 16.) then you can wire the SD adapter direct.
I think that SD would be cheaper than long cable runs and the chips you'd need to handle the extra voltage required for long cable runs. Also with cables, you need something running all the time to receive data.

Yea... SD cards could probably work. There is a laptop that is running all the time anyway collecting the weight measurements from the scale. So having a computer on to collect data isn't an issue. Alternatively we could get a Raspberry Pi that sits at the scale site and have it collect data from the hive and timestamp everything with it's central clock.

It would be nice if someone could figure out a way to get the scale data and have a Raspbery Pi read it or arduino.. lol. But that isn't necessary. I'm trying to think of something crazy, like a WWVB receiver that way everything is timecoded to the atomic clock, then probably one for USB on the PC that will also synch it' s clock. Then everything will have the same time stamps.

Just making this more complex I'm affraid... lol. :-\