Temperature differential between rad sensors and wall sensors

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • rs1987
    Automated Home Jr Member
    • Nov 2019
    • 44

    Temperature differential between rad sensors and wall sensors

    I just got my Evohome setup prior to it being installed this coming Wednesday. I bound the wall sensors and rads to the various zones.

    Then I put all 8 HR91s, the two T87Rs and the evohome touch next to each other on the table and left them there for an hour with the door closed to let the temperature stabilize in the room.

    It appears that the evohome touch controller (which reads way off if handled) and the T87Rs wall thermostats read about 2.5 degrees higher than the HR91s. Is this by design?

    If only one of the wall sensors read differently I would think it perhaps faulty, but the rad sensors read one temperature and the wall sensors read a cooler temperature. See below.

    Untitled.jpg
  • mtmcgavock
    Automated Home Legend
    • Mar 2017
    • 507

    #2
    Originally posted by rs1987 View Post
    I just got my Evohome setup prior to it being installed this coming Wednesday. I bound the wall sensors and rads to the various zones.

    Then I put all 8 HR91s, the two T87Rs and the evohome touch next to each other on the table and left them there for an hour with the door closed to let the temperature stabilize in the room.

    It appears that the evohome touch controller (which reads way off if handled) and the T87Rs wall thermostats read about 2.5 degrees higher than the HR91s. Is this by design?

    If only one of the wall sensors read differently I would think it perhaps faulty, but the rad sensors read one temperature and the wall sensors read a cooler temperature. See below.

    [ATTACH=CONFIG]1499[/ATTACH]
    Are your HR91s actually fitted to any radiators yet?

    Comment

    • rs1987
      Automated Home Jr Member
      • Nov 2019
      • 44

      #3
      Originally posted by mtmcgavock View Post
      Are your HR91s actually fitted to any radiators yet?
      No, they're not. They're all sitting on the table right next to each other. So theoretically the HR91s and the T87Rs should read the same temperature, unless the HR91s are deliberately setup to read cooler than the wall stat due to their proximity to a radiator.

      Comment

      • G4RHL
        Automated Home Legend
        • Jan 2015
        • 1580

        #4
        Originally posted by rs1987 View Post
        No, they're not. They're all sitting on the table right next to each other. So theoretically the HR91s and the T87Rs should read the same temperature, unless the HR91s are deliberately setup to read cooler than the wall stat due to their proximity to a radiator.
        This does not surprise me. I can have three temperature sensors side by side and each one will give a slightly different reading. I also use such to check n humidity (because I have a couple of classical guitars in one room) and they give slightly different readings. I use a free standing Honeywell Thermostat (the round one) to control the temperature in my living room, it is about two foot six off the floor on a side table well away from the door and the radiator, it invariably reads a different temperature than the HR92s on the radiators. Often about 1 to 1.5 degrees out.

        I understand the HR92s, and presumably the HR91s, are designed to allow for the proximity of the radiator and draughts across the floor so one can expect them to read slightly differently. I am sure that having your HR91 off the radiator and alongside the other thermostat is bound to give you a different reading because of where it is designed to be used.

        What I do find is that if say I have set the room temperature to 21c, if the HR92 is controlling the temperature, and not an external thermostat, then it will show 21c and shut off when the actual room temperature is less. All because of where you are and positioning etc. Don’t tell my wife though for when she says it is cold and the HR92 is consulted it reassures but in fact the room temperature, other than in my lounge, is actually lower. The free standing Honeywell thermostat in my lounge controlling the temperature always reads a degree or so less than the HR92.

        I think you will find a variance with many different thermostats unless they have been built for specific use where 100% accuracy is needed.

        Comment

        • mtmcgavock
          Automated Home Legend
          • Mar 2017
          • 507

          #5
          Originally posted by rs1987 View Post
          No, they're not. They're all sitting on the table right next to each other. So theoretically the HR91s and the T87Rs should read the same temperature, unless the HR91s are deliberately setup to read cooler than the wall stat due to their proximity to a radiator.
          The HR91s will compensate for the radiator so will show less.

          The Main Panel is over due to the heat from the panel. I've had to offset mine -1.5oc because of this to get a true reading.

          My T87RFs are bang on when checked.

          To be fair my HR92s are pretty spot on too with the reported/actual room temperatures. Mine do have plenty of free flowing air around them though and not obstructed by curtains etc.

          Comment

          • rs1987
            Automated Home Jr Member
            • Nov 2019
            • 44

            #6
            Originally posted by mtmcgavock View Post
            The HR91s will compensate for the radiator so will show less.
            It certainly looks that way, which I suppose would be a logical way of dealing with their proximity to the rad.

            Originally posted by mtmcgavock View Post

            The Main Panel is over due to the heat from the panel. I've had to offset mine -1.5oc because of this to get a true reading.
            Good tip, thanks. Looks like I'll need to do the same. The reading on the touch controller will approach the TR87 over a period of time if it is not touched and on battery power. I did read some threads talking about this issue on this forum. I think I may just get another TR87 and not use the touch panel as a sensor as I find charging it or handling it will cause the reading to jump up by a few degrees.

            Originally posted by mtmcgavock View Post

            To be fair my HR92s are pretty spot on too with the reported/actual room temperatures. Mine do have plenty of free flowing air around them though and not obstructed by curtains etc.
            I have a couple of HR91s that are fitted to rads that have been installed in some idiotic wooden cabinets as seems to be the fashion in so many London houses, so in these cases I'm using TR87s to read the room temperature. Glad to know that in the case where they're not obstructed that the temps are accurate.

            Thanks everyone for the input.

            Comment

            • DBMandrake
              Automated Home Legend
              • Sep 2014
              • 2361

              #7
              Originally posted by rs1987 View Post
              I just got my Evohome setup prior to it being installed this coming Wednesday. I bound the wall sensors and rads to the various zones.

              Then I put all 8 HR91s, the two T87Rs and the evohome touch next to each other on the table and left them there for an hour with the door closed to let the temperature stabilize in the room.

              It appears that the evohome touch controller (which reads way off if handled) and the T87Rs wall thermostats read about 2.5 degrees higher than the HR91s. Is this by design?
              Unfortunately, yes.

              The HR91 has a fixed "calibration" setting of about -2C (or it might be -3C, can't remember as it's a while since I tested one) so it sends a temperature reading that is 2-3 degrees below what it senses. This is to compensate for directing heating from the radiator, however unlike the HR92 you cannot adjust this built in "calibration" offset for different circumstances, it is fixed.

              On an HR92 you can adjust it from -3C to +3C in one degree steps including being able to set it to 0C or no offset. The degree of offset required depends greatly on the room layout, the size of the radiator, whether the radiator is a convector or not, whether the radiator has "end plates" (which act as a heat shield and improve the accuracy of the reading) and many other factors.

              In short if you're fussy about the accuracy of the reported temperature (as I am) you'd need to determine the required calibration setting on an HR92 for each room with a standalone thermometer in the room and adjust the calibration to get the best accuracy. Typically -1C works well for some rooms but sometimes more or less is needed.

              Unfortunately the HR91 doesn't let you adjust it so for this and other reasons I really can't recommend the HR91. I trialed one in my house and did not like it, and also had issues with it having problems with my (then at the time) Peggler Bulldog valve bodies, although I have since replaced those. If you use a remote wall sensor for a zone which uses HR91's then the calibration offset of the HR91 no longer applies since the reading is coming from the wall sensor and the internal sensor is not used. (Except for open window detection)

              The YT87R and DTS92E are both very accurate and don't require any calibration - the DTS92E's I have are within 0.1C of the stand alone weather station unit I have. The built in temperature sensor of the Evotouch always reads a bit high due to internally generated heat - I have mine mounted on a wall mount in the hallway and from trial and error I found a calibration of -1.0C is pretty accurate vs the weather station unit hung beside it.

              If you have the Evotouch on a table stand, move it around or pick it up, I would not use it as a temperature sensor as even holding it in your hand for a minute will cause the reading to shoot up as you've noticed and then take a long time to cool down to the correct reading again. (Meanwhile your radiators in the zone will be shut off due to it thinking it's too hot) However if its on a wall stand 99% of the time and is not handled then it can act as a "free" thermostat for one zone.
              Last edited by DBMandrake; 12 January 2020, 06:49 PM.

              Comment

              • rs1987
                Automated Home Jr Member
                • Nov 2019
                • 44

                #8
                Originally posted by DBMandrake View Post
                Unfortunately, yes.
                Thanks for confirming

                Originally posted by DBMandrake View Post

                Unfortunately the HR91 doesn't let you adjust it so for this and other reasons I really can't recommend the HR91. I trialed one in my house and did not like it, and also had issues with it having problems with my (then at the time) Peggler Bulldog valve bodies, although I have since replaced those. If you use a remote wall sensor for a zone which uses HR91's then the calibration offset of the HR91 no longer applies since the reading is coming from the wall sensor and the internal sensor is not used. (Except for open window detection)
                I think I can live with this. I'll check the differential between a wall sensor and a rad sensor in each room and just adjust the temperature accordingly in rooms where I'm not using a wall sensor. It should get pretty close to what I need.

                I deliberately chose the HR91 for two reasons: the first, and most important, is that it is way quieter than the HR92. I'm super noise sensitive and the actuation of the HR92 is guaranteed to wake me up. The second is that the price of an HR91 based system when purchased in a value pack was about £100 lower than a similar HR92 system, although this was a secondary consideration.

                I did actually read your post about your Peggler valve bodies before I went ahead and purchased. Very useful information that report. As a result, I'm having brand new Honeywell Valencia valve bodies fitted to each radiator so as to eliminate any potential problems from old, sticky valves. The existing valves are also Honeywells, but they look pretty old and I don't really trust them to work flawlessly or actuate smoothly.

                Originally posted by DBMandrake View Post

                The YT87R and DTS92E are both very accurate and don't require any calibration - the DTS92E's I have are within 0.1C of the stand alone weather station unit I have. The built in temperature sensor of the Evotouch always reads a bit high due to internally generated heat - I have mine mounted on a wall mount in the hallway and from trial and error I found a calibration of -1.0C is pretty accurate vs the weather station unit hung beside it.

                If you have the Evotouch on a table stand, move it around or pick it up, I would not use it as a temperature sensor as even holding it in your hand for a minute will cause the reading to shoot up as you've noticed and then take a long time to cool down to the correct reading again. (Meanwhile your radiators in the zone will be shut off due to it thinking it's too hot) However if its on a wall stand 99% of the time and is not handled then it can act as a "free" thermostat for one zone.
                I'll be buying another T87R and using that instead of the evohome touch panel I think.

                Comment

                • rs1987
                  Automated Home Jr Member
                  • Nov 2019
                  • 44

                  #9
                  Originally posted by rs1987 View Post
                  I did actually read your post about your Peggler valve bodies before I went ahead and purchased. Very useful information that report. As a result, I'm having brand new Honeywell Valencia valve bodies fitted to each radiator so as to eliminate any potential problems from old, sticky valves. The existing valves are also Honeywells, but they look pretty old and I don't really trust them to work flawlessly or actuate smoothly.
                  I've just read your other post about the Valencia bodies (I've bought Valencia VT117-15As) and it seems like I should probably not use these given the experience you and others have had?

                  Comment

                  • DBMandrake
                    Automated Home Legend
                    • Sep 2014
                    • 2361

                    #10
                    Originally posted by rs1987 View Post
                    I've just read your other post about the Valencia bodies (I've bought Valencia VT117-15As) and it seems like I should probably not use these given the experience you and others have had?
                    Jump to the last post in my Valencia thread - with Stroke set to 1 I've been happy with their performance on an HR92.

                    I can't comment on how they will perform on an HR91 as I didn't have the Valencia's back then, and as you point out the motor actuator in the HR91 is quite different to the HR92, so let us know how it goes for you.

                    Comment

                    • Jack007
                      Automated Home Jr Member
                      • Nov 2018
                      • 32

                      #11
                      The built in temperature sensor of the Evotouch always reads a bit high due to internally generated heat
                      Wasn't there also an issue with the heat produced by charging the internal batteries of the EVO or is this what you mean?

                      Comment

                      • DBMandrake
                        Automated Home Legend
                        • Sep 2014
                        • 2361

                        #12
                        Charging the batteries probably generates additional heat but if the unit is left on the charger the batteries are rarely charging as they sit fully charged most of the time.

                        The always on display backlight, cpu and various other chips on the board all generate heat continuously. Not much but enough to cause a temperature offset on the built in sensor, however you can adjust the calibration in 0.1C steps so if you have another thermometer to compare it against you can calibrate it quite well.

                        Comment

                        • rs1987
                          Automated Home Jr Member
                          • Nov 2019
                          • 44

                          #13
                          While we can deal with the extra heat generated by the components by using the offset, the thing that is really inconvenient is how slow the system is to return to the correct temp once it's been handled. It literally takes over an hour to stabilize. So I went ahead and bought another TR87 and I'll be using this instead of the touch panel.

                          Comment

                          Working...
                          X