pre-testing how we check radiators


I thought I’d try some of what we’ll be doing at St Columba’s, but from the relative comfort of my own home – and potentially answering some questions at the same time.  Here’s the set-up.

Question one:  How much does the taping affect the readings of waterproof DS18’s on the (metal) surface of radiators? Of course we want maximal contact, metal on metal – but the sensor is round, and the radiator is flat. I’ve insulated one sensor with some curtain blackout lining, which is known as a reasonable thermal insulator. The question is whether it reads higher than the sensor next to it at peak, and how much it slows down the sensor response.



Question two: Is the lounge radiator behaving as expected? The work room is fine; the radiator is probably a bit bigger for the space, and it doesn’t have a big bay window or as big a fireplace, but still, it doesn’t seem right. I say “work room”, but it’s really just the room where we do just about everything, because it’s warm. Our demonstration of the principle of adaptive thermal comfort would be complete if we just moved some comfortable chairs into the work room and were done with it. Alas, they won’t fit through the door, and the one we bought from Lewis’ mysteriously vanished from the van on delivery day.


No answers yet, by the way. We’re waiting for the heating to go on and off again. There’s also the question of how to tape so that the sensors reliably  stay on warm metal, at least for me – but I know Iain already has that one sorted.

3 thoughts on “pre-testing how we check radiators

  1. Why does the insulated sensor record a higher temperature than the metal to metal taped one? Should we use insulated sensors in St C’s?

    1. Further to what I put last night -I could at least test the inside and outside of a saucepan full of water while I bring it to the boil. My nice cold kitchen should approximate the conditions fairly well 😉

  2. I think those waterproof DS18B20s are meant to be surrounded by what they’re trying to measure – and so they’re probably influenced by the cold air moving past on the non-radiator side. At least, I think the higher temperature must be a more accurate reflection of the water temperature inside the radiator (since how else would the reading be higher). The fact that there’s a difference between the two readings, and the insulation doesn’t seem to slow down the response rate, suggests to me that the insulation helps. Does that make sense to you?

    I’m going to see if I can also take some readings with infrared devices, although judging the emissivity of the radiator surfaces might be difficult. I don’t think I can set up a test rig where I know the water temperature to be sure of what the reading “should” be, unless we find a spare radiator somewhere that isn’t plumbed in!

Leave a Reply

Your email address will not be published. Required fields are marked *