External wall temperature sensors

[DavidL]

Hi all,

We've recently had an extension finished, and we've moved away from the wireless Loxone switch we had in there to a cabled switching solution.

As with most other rooms, I've added a 1-wire temperature sensor behind the switch plate for input in to the heating schedule within Loxone.

I've applied some adjustment to the value of the temperature sensor to account for the difference between actual temperature in the room and what the sensor thinks it is.

What i've noticed is that due to the wall the switch is on being external, I'm getting vastly different temperatures depending on how the weather has affected the temperature of the wall.

Does anyone have any working temperature adjustment for when a sensor is on an external wall?

David

[Duncan]

you may have to play with the values depending on your insulation and how exposed/windy your location is

[DavidL]

Thanks Duncan. An interesting formula, I'll give it a whirl.

I'm curious about your external temperature sensors. Are they reasonably close, but on the outside, to where your internal sensors are?

I have a couple of external sensors. The one I'd use in this instance is mounted at roof level, mostly shielded, but probably around 10metres away.

We have reasonable cavity wall insulation, 100mm insulation under the floors and similar above in the ceiling, so I'm assuming i'd be adjust the number 14 to reflect any changes I might need to get the right balance for us?

Thanks again.

David

[Duncan]

my external temperature sensor is in the shade on the north side of the building, i dont think the location is important as long as its not in direct sunlight

14 is the offset and /7 is the gain - if you find your output is too low at low temperatures then reduct the 7 so try 6.5 or even 6

its all trial and error and if your house isnt particularly airtight then you might use perceived outside temperature (which includes wind chill factor) rather than real temperature

[DavidL]

Thanks Duncan.

[Joakim Arfvidsson]

If you really want to nail the formula, collect at least 20 data points during different conditions, with these variables:

- Real indoor temperature as measured by independent sensor in middle of room

- Your (bad) indoor sensor

- Outside air temperature

Then pull this into a Excel/Google Sheet, and experiment with various formulas to see how you can approximate real indoor temperature using the other two. You could probably also use various builtin kinds of regressions in the Excel/Google Sheets tool.

[DavidL]

Good idea Joakim.

I've been trying this formula out on the side without using it for controlling the heating yet and I'm struggling to find a good balance of offset and gain.

David