Page 1 of 3 123 LastLast
Results 1 to 10 of 24

Thread: Why is the setpoint way over the set temperature?

  1. #1
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default Why is the setpoint way over the set temperature?

    As part of me catching up with the latest Evohome I have a question.

    Ever since installing the new Evohome controller I notice that one of my zones gets a higher setpoint than necessary. This zone is a hallway, with a small radiator, that is set for 18C during the day and evening, and 15C during the night.

    Every morning during warming-up the setpoint of this zone is set to 23C. Temperature then runs up quickly and the setpoint is adjusted to 18C. The temperature itself overshoots just a little, but I don't understand why Evohome selects a 23C setpoint when it only needs 18C. None of my other zones show this behavior, even when they have to warm up 2C they set the setpoint at the desired temperature.

    I already reinstalled the zone (deleted the zone, cleared the HR92 and re-learned it) but that did not change anything.

    Is this normal?

    setpoint overshoot.jpg

  2. #2
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    2,254

    Default

    A few questions -

    1) What firmware version is the controller running ? (Press settings then hold down Device Info for 20 seconds)

    2) How are you graphing the temperatures and set points ?

    3) Are you using Optimal Start ? (I'm assuming so)

    4) Have you changed the maximum zone temperature for that zone in the zone configuration to 23C ? (default is 35C)

  3. #3
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default

    Quote Originally Posted by DBMandrake View Post
    A few questions -
    1) AS: 02.00.19.31, WiFi: 02.00.17.00

    2) I'm graphing the temp data with a home automation system, Domoticz. This basically gets these temperatures every couple of minutes from Honeywell via my account and the API. The values are the same as the values displayed on the controller: temperature measured and temperature set per zone. Domoticz (running on an always-on RPi) charts this data per day, month and year which gives me a nice insight in the performance of Evohome.

    3) Yes, both optimal start and optimal stop are switched on

    4) Yes, Tmax is dialed down on all my zones to 23C. Which made me suspicious of seeing precisely that value being used by my hallway zone....


    Paul

  4. #4
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default

    Ah, seems I found an answer for this problem elsewhere in this forum. For some reason my brand-******** new Evohome controller has a beta FW version that exhibits this problem. Should be repaired in 19.33 so now I have to see how to lay my hands on that version. I read something about mailing someone to get that done. I'll look that up again and see if I can get this fixed.

  5. #5
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    2,254

    Default

    Yep, known problem with the 19.31 beta, fixed in 19.33.

    A lot of people who didn't sign up for the beta test seem to have ended up with the first (quite buggy) beta on their system - some sort of error on Honeywell's part.

    Good news is that the second beta, 19.33 works very well, I've been very happy with it since my system updated to it, and apart from one minor niggle it feels very stable and polished.

  6. #6
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default

    I requested the beta with Honeywell. Let's see if that updates it to 19.33.

    Oh, and by the way, this forum seems to be more prudish than my grandmother; the intensifier I used to emphasize how new my Evohome is was replaced by a lot of asterisks :-))

  7. #7
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default

    19.33 was installed yesterday and the hallway zone is now behaving as expected; the setpoint only rises to the set temperature. Nice!

    One question about the 'advanced load scaling' option: does that also work with an on-off interface (bdr91) or only with an OT interface to the boiler?

  8. #8
    Automated Home Legend
    Join Date
    Jul 2014
    Posts
    1,375

    Default

    Actually only works with the BDR91, doesn't work with the OT as that is deemed to be an adaptive control anyway

  9. #9
    Automated Home Jr Member
    Join Date
    Feb 2015
    Posts
    32

    Default

    So scaling does not do away with the 'ask for 100% as soon as a setpoint differs more than 1.5C' using an R8810? That is a pity as that might solve the problem I raised elsewhere on this site. One would expect this also to work with OT which does not seem very adaptive, at least not when when used with a Remeha boiler.

  10. #10
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    2,254

    Default

    Yes it only works with a BDR91 - I find it works great, however I have mine set to "Partial" which we're lead to believe means that it doesn't take the outside temperature reading into account whereas Normal will scale the load from a zone down even more when the weather is warmer than when it is colder, which in theory is a good idea.

    I found Normal too "aggressive" at reducing the heat demand to the point where some of my zones were struggling to get up to their set points. Partial works well for my system.

    When you change the Advanced Scaling Mode setting you need to leave it for at least 2-3 days for the HR92's to adapt to the new behaviour, so you might find your zones initially overshoot (or more likely undershoot) a bit for a day or two until the system adapts to the change as it fundamentally alters the heat demand sent to the boiler.

    The point of load scaling is to allow the system to adjust more gracefully to schedule changes and minimise overshoots. For example without load scaling if you have 5 zones running and up to equilibrium temperature then a 6th zone in a cold room - say a bedroom - is scheduled to come on and the zone was previously <1.5C below the set point, the boiler heat demand would go instantly to 100% and get the boiler up to maximum flow temperature very very quickly. The old algorithm just sent the highest heat demand from any zone to the boiler, so any one zone calling for 100% would cause the boiler to go to 100%.

    This would usually have the effect of other zones that were at a nice equilibrium overshooting because those HR92's were counting on the low average flow temperature they were calling for until the 6th zone upset that nice balance.

    Load scaling reduces the heat demand from each zone before sending it to the boiler, so a single zone calling for 100% does not in fact call for 100% from the boiler - it may get scaled down to 40% for example. Also the heat demand from zones is semi-cumulative. So an individual zone calling for 100% can't cause the boiler to go to 100%, but multiple zones calling for 100% can. The scaling ratio also adapts over time so for example the ratio between requested heat from the zone and the demand set to the boiler is different for each zone, learnt over time.

    The end result is sudden changes to the set point caused by the schedule are "smoothed out" and the temperature changes a bit more gradually with much less impact on other zones. If you graph your zone temperatures you'll probably set the set point targets being hit and maintained much closer with much less overshoot than before. That's what I see anyway.

    There is one exclusion with load scaling though - if you increase the temperature in a zone manually as a timed override, and the override is sufficiently large load scaling for that zone is cancelled. So you might still see an overshoot in other rooms when making a manual override which you wouldn't if it was just a normal schedule change.

    This was a compromise introduced in 19.33 as without this (19.31) manual overrides to for example turn up a bathroom radiator could take too long until the radiator actually started to heat up. I think it's a good compromise though because the system stays stable with minimal overshoot when following the schedule but still allows you to get quick heat in any zone with a manual override.
    Last edited by DBMandrake; 10th November 2020 at 01:55 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •