Page 1 of 16 12345611 ... LastLast
Results 1 to 10 of 154

Thread: Opentherm + EvoHome + Viessmann 111-W overshoots

  1. #1
    Automated Home Guru
    Join Date
    Feb 2017
    Posts
    102

    Default Opentherm + EvoHome + Viessmann 111-W overshoots

    Hi,

    I have had EvoHome for a little over a year now with no real issues, utilising a BRD91 relay.

    I have been reading this forum with interest around Opentherm and have had the BDR91 replaced with the R8810 Opentherm Bridge yesterday.

    This was straightforward....however, where I had very few overshoots that I can remember on the old (reliable) BDR91 relay, I'm seeing very large overshoots on multiple rooms even after 24hours of temp set point changes.

    One thing that struck me straight away when the Gas Safe engineer first installed and binded the R8810 Opentherm Bridge was that when calling for heat, and where all zones calling for heat were only 1.5 degrees C below set point, it fires the boiler up to full throttle and reaches 81/82 degrees C flow temp according to the Viessmann 111-W Combi boiler (35KW) digital display.

    The other problem, and a contributing factor to the overshoots no doubt, are that the radiators get far too hot at this temperature - and there doesn't appear to be a way on this boiler (that I know of) to reduce the max CH flow temp (to say 65 degrees c).

    Whilst the burner eventually cuts out, and then let's the flow temp fall down, this happens about 5 minutes or so *after* the EvoHome display shows all zones have reached or even passed beyond the set point in other words, extremely laggy.

    Once rooms have fallen back down to set point, it does seem to modulate the flow temp down as expected - but the issues reaching set point as described just don't seem right to me.

    Hopefully the attached screenshots of the datalogging shows what I mean here.

    Is there any way of reducing the flow temp on a Viessmann 1xx boiler? I think I'll have to go back to the BDR91 relay if not due to the risk of scalding (and excessive gas use whilst no where near condensing temps!!)

    I have 11 zones with 1 HR92 in each zone for further context, this has been working really well to date.

    I only have a large towel rail (no TRV) and 2 fairly small rads (with TRVs) that are not dynamically controlled.
    Attached Images Attached Images
    Last edited by StephenC; 8th March 2017 at 10:34 AM. Reason: Spelling

  2. #2
    Automated Home Legend paulockenden's Avatar
    Join Date
    Apr 2015
    Location
    South Coast
    Posts
    1,601

    Default

    Maybe changing the control method has re-started the learning process? See if it starts to request more sensible temperatures after a couple of weeks.

  3. #3
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    1,841

    Default

    Agreed. It won't learn how to modulate the boiler in only 24 hours, that learning process can take a couple of weeks.

    While the flow temperature remains excessive for the demands you will see overshoots or oscillation in some rooms.

    I'm very surprised your boiler doesn't allow you to limit the flow temperature though - it was my understanding (not having used Opentherm personally) that whilst the Evohome can request a higher temperature, the maximum allowed flow temperature is still set at the boiler using the normal flow temperature control - this acts as a "cap" on the maximum flow temperature that the Evohome can effectively call for. If you have large efficient radiators you probably don't want flow temperatures anywhere near 80 degrees. (Even on my old system with less than ideal radiators I don't set it above 75 in the middle of winter)

    Have you tried contacting Viessmann to find out how to manually limit the maximum flow temperature ? Allowing the Evohome to have access to the maximum temperature that the boiler can put out if it is not necessary for your radiators will only result in overshoots, and while it will eventually learn to moderate this you won't get as good a temperature control as you will if you can set a sensible cap on the maximum flow temperature.

    (For example a room scheduled to come on to a few degrees above its current temperature while other rooms are already at their set point can cause maximum flow temperature to be called for which will disrupt the other rooms causing them to all overshoot temporarily. This effect is much less if the maximum allowed flow temperature that can be called for is moderated to a reasonable temperature by adjusting the setting on the boiler - otherwise the Evohome always calls for maximum available flow temperature when there is a large demand, even just for one zone)
    Last edited by DBMandrake; 8th March 2017 at 10:16 AM.

  4. #4
    Automated Home Guru
    Join Date
    Feb 2017
    Posts
    102

    Default

    Thank you both for your thoughts on this.

    I will give it more time to settle down and report the result back. In the mean time, I have reduced (a degree or two) the set point targets to allow for overshoot and to avoid uncomfortable rooms. I will see if it can handle meeting this before moving the set points up slightly over time.


    Re: reducing max CH flow temp of the boiler, the Gas Safe engineer did ring Viessmann to see about any way of reducing CH flow temp, but they were fairly (but not completely!) sure you cannot, as the analogue CH flow temp dial on the front of the boiler is disabled when controlled via Opentherm.

    The KW output for CH has been reduced down a bit as per last paragraph on pages 27 and then page 28 of the installation manual below, but all this seems to do is just take longer to reach 81/82 degrees C! Doesn't seem to affect the flow temp.


    http://www.viessmann.com/http-svc/we...e%20Manual.pdf

    It would be useful if EvoHome could have a config in the settings as well for situations like this to config a max CH flow temp. I think Opentherm protocol has this attribute capability.

    DWH has been unaffected for info which is good (and expected), but just thought I'd share as not many people seem to have Opentherm or combi boilers on here.

    One other point that I have seen debated is that Viessmann stated on the same phone call that Weather comp was superior to Opentherm.....highly controversial from a logical and common sense approach, but explains why Opentherm is not pushed in this country with that info being pushed down maybe?

    My Gas Safe engineer has suggested if the learning algorithm doesn't work, then possibly adding weather comp to the Opentherm/boiler setup to see if it can reduce the flow temp down from 81/82 degrees C in all but extreme cold weather. My assumption (dangerous I know) is that weather comp will only affect the top end CH flow temp, and then Opentherm will attempt to lower the flow whenever possible. This is another fallback option if this logic is correct I guess.
    Last edited by StephenC; 8th March 2017 at 10:33 AM. Reason: Spelling

  5. #5
    Automated Home Legend paulockenden's Avatar
    Join Date
    Apr 2015
    Location
    South Coast
    Posts
    1,601

    Default

    Quote Originally Posted by Stephen Crown View Post
    Viessmann stated on the same phone call that Weather comp was superior to Opentherm
    How would weather compensation know that your house was pretty much up to temperature and so the boiler just needed to trickle a bit of heat in to keep it there?

    Weather compensation is perhaps an 'equivalent' to optimisation in Evohome. Comparing it to Opentherm is confusing the function with the control mechanism.

    P.

  6. #6
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    1,841

    Default

    Quote Originally Posted by Stephen Crown View Post
    Thank you both for your thoughts on this.

    I will give it more time to settle down and report the result back. In the mean time, I have reduced (a degree or two) the set point targets to allow for overshoot and to avoid uncomfortable rooms. I will see if it can handle meeting this before moving the set points up slightly over time.
    If you can't reduce the maximum flow temperature some way then I don't think you'll ever get a fully satisfactory result. What did you have your flow temperature set to manually before the change over ?
    Re: reducing max CH flow temp of the boiler, the Gas Safe engineer did ring Viessmann to see about any way of reducing CH flow temp, but they were fairly (but not completely!) sure you cannot, as the analogue CH flow temp dial on the front of the boiler is disabled when controlled via Opentherm.
    That is pretty ridiculous - hopefully some of the OpenTherm users on this forum will chime in with their experiences of whether their boiler lets them set a maximum flow temperature while under OpenTherm control. I thought all did, but apparently not if yours is anything to go by!
    The KW output for CH has been reduced down a bit as per last paragraph on pages 27 and then page 28 of the installation manual below, but all this seems to do is just take longer to reach 81/82 degrees C! Doesn't seem to affect the flow temp.
    Correct. This won't help the problem and will just cause the system to take longer to heat up when first turned on.
    I searched through that manual - only one match for OpenTherm and no information on how it works whatsoever. Also not a single mention of how to set the flow temperature that I could find! Typical that the installation manual basically says nothing about OpenTherm. Is there another users manual that goes into more details about controls, does it say anything at all about OpenTherm ?
    It would be useful if EvoHome could have a config in the settings as well for situations like this to config a max CH flow temp. I think Opentherm protocol has this attribute capability.
    Yes a maximum flow temperature setting in the Evohome itself in OpenTherm mode in theory could be implemented very easily, the Evohome asks for the flow temperature it wants so it could easily limit itself to asking for a lower temperature.

    Unfortunately Honeywell seem to be glacially slow when it comes to introducing any new features to the Evohome controller - in the 1 1/2 years that the firmware upgradable Wifi model came out there has only been one update, and the only new feature of note was the local override display for HR92's.

    I would contact Honeywell and express your desire for such a feature though - if nobody asks for it it won't happen!
    One other point that I have seen debated is that Viessmann stated on the same phone call that Weather comp was superior to Opentherm.....highly controversial from a logical and common sense approach, but explains why Opentherm is not pushed in this country with that info being pushed down maybe?

    My Gas Safe engineer has suggested if the learning algorithm doesn't work, then possibly adding weather comp to the Opentherm/boiler setup to see if it can reduce the flow temp down from 81/82 degrees C in all but extreme cold weather. My assumption (dangerous I know) is that weather comp will only affect the top end CH flow temp, and then Opentherm will attempt to lower the flow whenever possible. This is another fallback option if this logic is correct I guess.
    As Paul says, OpenTherm and Weather Compensation whilst both controlling flow temperature are somewhat orthogonal to each other and can actually even be used together.

    OpenTherm allows the flow temperature to be modulated based on heating demand - during initial heat up a high flow temperature can be called for to get the house warmed up quickly but when rooms approach their set temperatures the flow temperature can be lowered to be just high enough to maintain the status quo. This gives smoother room temperature control and lets the boiler work in condensing mode even in the middle of winter, once the house has warmed up.

    Weather compensation only adjusts the flow temperature based on outside temperature - if its freezing outside it would give you a high flow temperature regardless of whether the demand in the house was high or low. What weather compensation gives you is reduced flow temperature during warm weather - this means the boiler will work in condensing mode and it also helps prevent overshoots that would occur if you didn't manually turn down your flow temperature in warm weather.

    Depending on your boiler, as your engineer suggests, it may let you use OpenTherm and Weather Compensation at the same time - with OpenTherm modulating based on demand, and Weather Compensation setting a flow temperature "cap" based on outside temperature. In winter weather compensation would set a high "cap" and OpenTherm would modulate within that range. In warmer weather weather compensation would set a lower flow temperature cap and thus restrict the range that OpenTherm could modulate across. That would be the ideal situation.

    Maybe if your boiler supports the two working together you could indeed get around the problem by adding the Weather compensation unit ? I would be wanting an assurance from the manufacturer first that both systems WILL work together properly though, otherwise it would be pointless.

  7. #7
    Automated Home Legend
    Join Date
    Jul 2014
    Posts
    1,002

    Default

    Suddenly my Vaillant boiler sounds so much better. I too don't like the amount of time the OT kicks in at max 90C but atleast I can vary that max and it's set to 70C. And 82C flow through the rads is going to toast people who touch them by mistake.

  8. #8
    Automated Home Guru
    Join Date
    Feb 2017
    Posts
    102

    Default

    Indeed, that's my main worry so far: the risk of scalding.

    I can understand learning needs to occur, but surely it should start at a safer/lower start temp and only increase if the algorithm detects the set point is not reached by the time needed.


    Re: my CH flow temps manually set prior to the Opentherm installation, I'd say:

    Winter: ~55-60 degrees C
    Spring/Autumn: ~40-50 degrees C

    For info, I had been trialling manual changes to the flow temp by adjusting the flow temp dial on the front when rooms were nearly up to temperature, and was surprised at how low I could set the dial (and therefore CH flow temp) whilst it still maintained set points in the rooms. Down to 35 degrees or so was sufficient in a lot of circumstances - and we are in an 1880's farmhouse with solid walls and some minimal internal insulation. The problem was of course at this low flow temp, rooms were never hitting set point if I didn't readjust back up the CH flow temp.

    It was reading up about weather comp and Opentherm - and then this forum that made me look at this manual experiment due to the potential gas savings (which there have been), hence now installing what I thought would be the "proper" automatic and dynamic control! Let's see how it goes and whether it actually does "learn"...

  9. #9
    Automated Home Guru
    Join Date
    Feb 2017
    Posts
    102

    Default

    @ bruce_miranda:

    Did you find massive overshoots initially upon Opentherm installation?

    If so, did the EvoHome controller learn the ch flow temps were too high for it to manage effectively and actually reduce them on its own?

    Does your Opentherm setup actually modulate the ch flow temp down when getting close to the set point? (Mine seems to ignore that - and go full blast for minutes after set point reached)

  10. #10
    Automated Home Jr Member
    Join Date
    Jun 2016
    Posts
    24

    Default

    Hey Stephen,

    I've got a 100W installed ( 19kw ) with Opentherm module , 9 Rads and stored HW.
    System was installed in june 2016 , so the learning process is over .
    Same behaviour here : the boiler will fire up to 81 if multiple rads are below the 1.5 degree set point and request heat.
    I also see overshoots on some rads -
    overshoot.jpg
    overshoot2.jpg
    I've set the max water temp to 60 on the boiler but this is ignored. I suspect that the EvoHome and Opentherm is overriding this ( would love to find a way to confirm / debug this )
    Overshoots get worse when the hot water kicks in : I find that the valves are slow to respond and close , so I'm running a script on my Pi / Domoticz which checks if HW is on : it then sets all the rads to 15 for x minutes. ( not a reliable method as some of the temperature changes don't go through - suspect the Honeywell API throttles the number of calls you can make within a period of time )

    Vic.

Posting Permissions

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