Remote temperature sensor sensitivity

[lugnut]

I just installed one of the ELK remote temp sensors with the 7 foot probe and the "silver bullet" on the end. The box is mounted indoors, and I drilled a 1/2" hole through the wall to the outside, then put a 1/2" OD steel tub through the wall, then slid the remote sensor through the tube.

My question is, how much of the silver bullet needs to be exposed to the outside before the readings are reliable? I was hoping to only have about 1/4" protrusion from flush with the outside wall (i.e. just the tip of the silver bullet is exposed). Will that read low?

 

[rfdesq]

I've only used one of the 7 footers. I stuck as much of the silver bullet out as possible. I guess you could do a test to determine if there is any difference based on how much sticks out. You may get more accuracy by running a test.

 

[lugnut]

I've only used one of the 7 footers. I stuck as much of the silver bullet out as possible. I guess you could do a test to determine if there is any difference based on how much sticks out. You may get more accuracy by running a test.

Thanks, that would work. Maybe someone will chime in that has already done the test.

By the way, what is the significance of the flashing orange light at the connector? It seems almost arbitrary the way it flashes. Sometimes 3, sometimes 4, and variable interval in between.

 

[rfdesq]

By the way, what is the significance of the flashing orange light at the connector? It seems almost arbitrary the way it flashes. Sometimes 3, sometimes 4, and variable interval in between.

I have found on all of them that I've installed, if they flash they work. I haven't spent too much time on the flash pattern.

 

[Spanky]

The flashing shows an 8 bit manchester code of the temperature with a value that is + 60 degrees above the real temperature. The slow data rate is so that the M1 filtered zone inputs can read the temperature value. It also takes two temperature reads in a row that match before the temperature changes. Each temperature read time is about 17 seconds.

The temperature sensor is at the tip of the stainless steel tube.

 

[lugnut]

The flashing shows an 8 bit manchester code of the temperature with a value that is + 60 degrees above the real temperature. The slow data rate is so that the M1 filtered zone inputs can read the temperature value. It also takes two temperature reads in a row that match before the temperature changes. Each temperature read time is about 17 seconds.

The temperature sensor is at the tip of the stainless steel tube.

Perfect; thanks

 

[BrettS]

My question is, how much of the silver bullet needs to be exposed to the outside before the readings are reliable? I was hoping to only have about 1/4" protrusion from flush with the outside wall (i.e. just the tip of the silver bullet is exposed). Will that read low?

You might want to re-think this mounting option. I was looking for a good place to mount a temperature sensor outside and did a bit of experimenting with a little battery operated thermometer. Compared to a reading out in the yard in the shade (Which should be pretty accurate) I was getting an increase of 4-5 degrees in the shade with the thermometer a few inches from my house, and putting the thermometer up against the house gave me a reading almost 10 degrees higher than the reading in the yard in the shade.

Note that these were done in the late afternoon with the Florida sun shining on the wall of the house I was using, so depending on where you live and how much sun hits your house you might not have as much of a deviation as I did, but I bet you'll still be off by at least a few degrees in the afternoons in the summer from the heat radiating from your house.

In the end, I've become convinced that it's not possible to get an accuarate temperature reading right next to a house, so I'm looking for other options to mount a sensor out a bit in the yard.

Brett

 

[beelzerob]

In the end, I've become convinced that it's not possible to get an accuarate temperature reading right next to a house, so I'm looking for other options to mount a sensor out a bit in the yard.
Brett

Or just apply an offset to the temp reading, assuming it's consistently high (I don't imagine bearing near the house would cause cooler temps).

 

[BrettS]

In the end, I've become convinced that it's not possible to get an accuarate temperature reading right next to a house, so I'm looking for other options to mount a sensor out a bit in the yard.
Brett

Or just apply an offset to the temp reading, assuming it's consistently high (I don't imagine bearing near the house would cause cooler temps).

Yeah, but it's not quite that easy as it depends on time of day (it's not too far off during the middle of the night or morning) and how sunny or cloudy it is, or even if it's raining as the water will cool off the building and it will radiate less heat. I considered trying to build some sort of basic offset calculation into it based on the time of day or maybe even some sort of light meter, but it got pretty complicated pretty quick and in the end I'm not sure that it would have really increased my accuracy much at all.

Brett