Measuring water temperature

Post by **bjorn** » Sun Dec 17, 2006 11:43 pm

Hi all. At least Bwired is measuring the water temperature in, for example, the heating pipes. Does placing the 1-wire sensor against the pipe with isolation around the sensor and pipe give an accurate enough reading or do you have the water flowing around the sensor? How do you have it connected on or in the pipe to get the temperature?

Post by **John Downe** » Mon Dec 18, 2006 1:14 pm

Bjorn, my experience of this when trying to measure a swimming pool temperature was that strapping the sensor on the pipe in the filter/pump box with surrounding insulation wasn't satisfactory - not helped I assume by plastic piping. I had to put the sensor actually in the pool water itself - encased in a metal tube filled with epoxy resin to seal it.

Post by **Bwired** » Mon Dec 18, 2006 6:11 pm

Hi Bjorn,
I have good experience with gluing the 1-Wire sensors to the tubes of my floorheating. It's very simple to see if it is working. The temperature in of the floor heating is at the most 50 degree Celsius and the temperature out is 34 at the most.
I glued the sensors very tight to the pipes and carefully picked the best place to put them. Of course it's better to do it the way like John told for his pool, but sometimes it's just not possible or very hard to do:-)
Regards Pieter

Post by **Digit** » Tue Dec 19, 2006 12:28 am

Hi John,

Could you give me an estimate on how much the temperature readings deviated from the "real" water temperature when you had them strapped on the pipe?

I'm having plastic pipes in some parts of my house, so i'd like to know what to expect or maybe just never start trying to read the temperature on plastic pipes.
I think the plastic must have been *the* major factor in not being able to get reliable readings from your sensors that way. What's your opinion about that?

Regards,
Digit.

Post by **John Downe** » Tue Dec 19, 2006 9:58 am

Hi Digit, the deviation varied a lot - partly due to the surrounding air temperature in the pump house which is in full sun and partly due to the fact that the water isn't actively flowing through the pipework all the time - only approx 50% of the time when the pump is on - so static water gets to ambient temp before too long. So the deviation ceratinly wasn't predictable enough to be calibrated out. To be honest when I discovered the problem I didn't spend too long analysing it though. If the plastic is a problem in your case is it possible to imbed small metal sections of pipe to put the sensors on?
John Downe

Post by **beerygaz** » Tue Jul 21, 2009 2:20 pm

I have some RFXCom sensors in the little plastic boxes. On opening them up there seems to be one IC on the board, I'm assuming I need to find a way to get that IC as close to my hot water pipes as possible to get an accurate reading? Using some sort of thermal paste and taping/bandaging to hold it in place.

Has anyone come up with an alternative to accurately measure water temp? Is there merit in adding a T-piece to the pipe and inserting the sensor and trying to seal it in? Or is that just a flood waiting to happen?

Post by **Bwired** » Tue Jul 21, 2009 2:47 pm

Use the Oregon THC-138 Much easier and can be received with the rfxcom receiver.
See my answer here:
http://www.domoticaforum.eu/topic.asp?TOPIC_ID=1089