Please give me some pointers
Have had evohome set up for a week or so now. It's failing to reach my desired temperature in some zones.
Please could you give me some pointers on what to check.
Example
Living room is supposed to be 19C between 6-7am
This morning it was 18C at 5.45am, radiator cold, and didn't warm up anytime before 7am. Boiler was firing as other (non HR92 rads) were hot.
Why am I not reaching my desired settings?
Why was the living room radiator not hot when it was below target.
I have optimum start and stop enabled (with regards to the latter, i assume it wouldnt 'stop' until the target temperature had been reached)
The photos below show the status of the living room, together with the desired set points, at 5.45am - 15 mins before temperature is meant to be 18.
IMG_3567.jpgIMG_3568.jpgIMG_3569.jpg
Thank you
I suspect that it probably has something to do with the optimum off and it may need more than an hour so it kicks in. (ie opt off is kicking in before the target temp is reached. I think also that the system probably needs a bit more time to 'settle'.
Have you looked on the logs in the SmartThings app. I have found this very useful as you can get a good idea as to how each rad is heating up and down.?
We haven't all got access to the SmartThings beta @Mavis!!!!!
(Grrrrrr)
Good idea!, although those logs only provide minimum info. I'll let it go through the cycle tomorrow morning and then take a look at the log.Originally Posted by Mavis
Shame we don't have access to this info directly via evohome and app.
It's been observed in other threads on this forum that optimum start does not act if the temperature shortfall is only 1 degree or less.
So in your case because the target at 6am was 19 degrees and it was already 18 degrees before that time it will not come on early. However it will still change to a set point of 19 degrees at 6am. Had the room been 17 degrees before 6am optimum start would have turned it on early.
Because it was not below target. In your photo taken at 5:45 the current temperature was 18 degrees but the set point is still shown as 16 degrees - from the previous set point, therefore the valve will be closed as the room is hotter than the current set point by two degrees.Why was the living room radiator not hot when it was below target.
At 6am the set point will have changed to 19 degrees and the valve would partially open - partially because only 1 degree needs to be made up, from 18 degrees to 19, which does not require the valve to fully open.
All optimum start does is change the set point to the next scheduled temperature earlier than the scheduled time - and tries to calculate how much earlier it should do so based upon the temperature difference (the greater the difference to make up, the earlier it will increase the set point) and how quickly it has learnt that the room warms up.
Thanks for this, it certainly helps my understanding.
Back to the tweaking board then.
My main concern about this system is that the temperatures we set seem to be purely arbitrary numbers and don't actually reflect 'actual temperatures' in any useful way. (I do understand however that this same observation could be be applied to any room thermostat - it's only measuring the air temperature at that specific location)
I'm also trying to get my head around the temperature offset. In my bedroom this is a necessity as the HR92 tends to record a temperature at least 2 degrees (when the radiators are at their hottest) lower than where the bed is. (and it's only an average size bedroom). So I've set the temperature offset to -2.
What effect does this have on the operation of the system? - will it now continue heating for an 'extra 2 degrees' (so i leave my 'desired' temperature as it was) or does the offset just adjust the reported temperatures?
Drat, I have blown my cover - it was supposed to be top secret (I think that your comment on the ST forum might be closer to the truth.)
Have a read of my comments in the following thread, for a bit of explanation of what is going on:
http://www.automatedhome.co.uk/vbull...ng-temperature
I think your understanding of the calibration offset is back to front.I'm also trying to get my head around the temperature offset. In my bedroom this is a necessity as the HR92 tends to record a temperature at least 2 degrees (when the radiators are at their hottest) lower than where the bed is. (and it's only an average size bedroom). So I've set the temperature offset to -2.
What effect does this have on the operation of the system? - will it now continue heating for an 'extra 2 degrees' (so i leave my 'desired' temperature as it was) or does the offset just adjust the reported temperatures?
When you measure the temperature at the radiator as the HR92 does, it normally reads too high when the radiator is hot, due to the influence of the hot radiator right next to it. To compensate for this you would use a negative calibration value - I use -1 on mine. This makes the displayed temperature 1 degree lower than the "measured" temperature, and in theory closer to the actual room temperature away from the radiator.
This will also have the effect of heating the room 1 degree hotter than before you applied the calibration adjustment - the "corrected" value that is displayed is also the value that is used to decide whether the valve needs opening or not, so it is not merely a visual change, it will change the temperature that the room settles on.
(But see my posting in the other thread as to why there is no one correct temperature in a room - if the radiator is hot there will be temperature gradients through the room) If your HR92 reads too low you would actually want to set a positive offset, not negative.
If your HR92 is reading too low (before you added your -2 correction, which is the wrong way around) then it is most likely due to the influence of a window above it letting in cold air. I see similar effects to you. Let me paint a picture of our bedroom.
The head of the bed is against the north facing wall, the door is on the south facing wall, and the window with the radiator below it is on the east facing wall, closer to the south end of the room.
If the window and door are both closed then after the room warms up the HR92 reads about 1 degree too high, thus I set calibration to -1 to correct this. So long as the room is closed and nothing is on the radiator it is reasonably accurate.
However if I open the door to a cold landing or cover the radiator with a towel, the room temperature measured by a thermometer at the head of the bed can be as much as 3-4 degrees below the temperature reported by the HR92, even with the -1 correction in place. The HR92 is not "wrong", it is simply the case that the temperature at the radiator valve is a lot higher than the temperature at the head of the bed under these conditions. It is impossible to accurately gauge the temperature on the far side of the room from the radiator as measured at the radiator under these adverse conditions.
Conversely, if I remove any towels from the radiator, close the door, but open the window above the radiator (even an inch or so) the cold air coming down through the window passes the HR92 cooling it down, causing it to attempt to compensate by heating the room further - in this condition the room as measured at the head of the bed is actually 1-2 degrees hotter than the HR92 indicates. But again this is because the part of the room where the HR92 is and the head of the bed are different temperatures. It's not an error in the measurement, just a limitation of the measurement location that would apply to any TRV located there.
Because the offset varies depending on the state of the doors/windows in the room and anything covering the radiator it's impossible to "calibrate" this error out. The only real solution if the temperature at the head of the bed is important to you is to install a DTS92 or similar at the head of the bed and bind that as the temperature sensor for the room.
That solves all the problems in one fell swoop - there is no longer any localised heating from the radiator affecting the reading, (so no need for a -1 correction) cold air coming in the window past the radiator no longer fools the HR92 into thinking the room is colder than it really is, and above all, it is now the temperature at the head of the bed that it is trying to regulate. So it doesn't matter whether the window or door is slightly open or closed (within reason) or whether there are towels or clothes on the radiator - it will always settle on the correct temperature as measured at the head of the bed, and do what ever is necessary to achieve this.
I have tested this theory by setting up the controller with its internal temperature sensor at the head of the bed for a week and temperature regulation is excellent under all conditions, so I plan to install a DTS 92 on the wall at the head of the bed when funds are available for it....(I want to finish zoning with HR92's first as I still have quite a few manual TRV's)
Last edited by DBMandrake; 11th February 2016 at 04:06 PM.
Thanks once again for a very comprehensive response!
Actually my understanding of the temperature offset is not back to front because I have just realised my post had a 'typo' (reflecting the fact my head was hurting trying to understand!). When I said the HR92 was recording a lower temperature than at my bed, I meant higher. So having read your information, I am now satisfied that I was correct to offset by a negative value.
I agree that the way to go is to get the room stats. Shame they are so expensive. I much prefer the look of the Y87's (I assume these would be suitable) but they are £30 more than the DTS92's.
The square box is more functional than the pretty round one, and it can also be table mounted. So apart from looks, the DT92 wins every time.
P.
Posting Permissions
Reply With Quote