Why use so many interior temperature sensors in just one hive?

Now I am sure you’ve done much more research than I have.
That’s why I am asking.

How necessary is it to know the temperature in so many parts of a hive?
What information does it bring?

Also, bees like to seal everything with propolis. Is it a big problem? Is it a problem with internal Humidity sensors? How would you deal with such a problem?

Thanks for the replies and sorry if this is the wrong topic.

The purpose for so many temperature sensors was to monitor the size of the broad nest. We know that a temperature of 35 °C indicates broad activity. In hot summers a higher temperature is possible especially in the honey supers and of cause less temperature is nearly all over the year outside the broad. So if we measure 35 °C ± a half degree we know very well that there is broad.

My idea behind measuring the broad nest size was also to detect phases without broad in the winter time. E. g. for oxalic acid treatment in the winter against the varroa mite you have to know the broad status. The treatment is more successful without broad in the hive.

A “perfect” setup would be a sensor between every comb. Best would be to have two sensors in any comb alley. A small and not centered broad nest (as you can see it sometimes end of winter) could be detected so also. And you like this for every broad box. With 10 combs you would have 9 or 11 (in case you count the space between comb and box also) alleys, two broad boxes (x2) and two sensors per comb alley (x2) that means 36 sensors.

But this is perhaps overkill. ;-) So let us use only one sensor per comb and It is also unlikely to have broad in the border area, so we will concentrate on the central nest area. In the most cases we know wheater the broad is in the fist or second broad box, so let’s skip the second layer also and we will come to a number of 5 to 7 sensors to do the job with as much effort as needed but the minimal necessary hardware.

You can see on the left side of the picture a prototype for a hive with two broad boxes. I would not recommend this. It is a PCB on top, expensive, not so easy to remove and with the solder joins on top not usable with some queen excluder that lay directly on the combs. It is also difficult to pass the sensors for the upper box through the combs in the top box.

Wax and propolis is no problem for the water proof temperature sensors. It will take a bit longer till temperature changes reach the sensors but this is no problem, we have not fast changing deltas. This is different for the humidity sensor. Here you need an entrance for floating air. We are using a simple queen cage to protect the sensor. In case it is built up with wax you have to change the cage or clean it.


Wow, thanks for the detailed explanation!

I didn’t think about measuring nest size that way.
I was thinking the weight sensors would provide that, but I guess you’re correct, a lot of things attribute to weight.

Do you have any data available that shows how it looks?
Also, from what I understand you use basic Thermistors, right?

I guess I could avoid those questions if you could direct me to an article explaining the hardware, I haven’t found any yet.

You can find a detailed technical description under Open Hive Temperature Array

Sorry a long and German thread but you find a translation icon in the footer of every posting for auto translation.

We are using digital DS18B20 sensors with a good accuracy and you can use this sensors as “one wire” so all sensors on the cable share Vcc, GND and only one data pin for all sensors! According the spec more than 100 are possible.

A nice data visualization example made by @wtf can be found here:

And a very interesting research outcome with this sensors is

1 Like

So even better that you’ve asked!

I guess we’re getting some blind spots for what new people here want to read when arriving. As long as one can’t find these answers easily, please ask for them! :)

1 Like

So many questions at a time!

This serves for ‘measuring’ the size of the brood nest. Illustrating @clemens’ links with a combined image from there:

No thermistors here (yet). No need for high update rates, and in very most cases no energy budget for that. This would be a science use case involving a discrete OPAmp (example with KTY81 and AVR or another even simpler one) or other dedicated RTD front-ends.

Here only counterfeit Chinese night-shift DS18B20 are used! :)


Only the finest quality Chinesium!

What if I wanted to try this software out? Can you direct me to an article explaining it well?

The wiki is… well in a bad shape. Doesn’t explain much.