The Efficient Home Forum

Yesterday I was looking around the evohome menus and I discovered a menu that I did not stumble upon before. It shows what I assume to be the boiler load (opentherm in my case), and valve positions of the zones.
And I was just wondering - how does the boiler load always correspond to the valve position of the zones. In my case it seems to be equal with the largest valve position of the zones. See picture for an example (foreign language alert ).
It seems a bit strange to me. I would, in the situation shown in the picture, expect the valves in the zone requesting the most heat to opened fully, and then the evohome controller could regulate the temperature by water temperature.
That would allow the boiler to run cooler (under dew point), losses in the pipes and radiators without evohome valves would be less, the flow pump might consume less power.

Is there a reason for this approach?

Also, is it known how the opentherm percentage corresponds to a temperature?

evohome will always use the highest demand as its setpoint for the BDR91 or R8810... Failure to do this may mean the zone with the highest calculated percentage may never achieve the correct temperature. OpenTherm has zero control over pump speeds and since ErP this is directly burner linked.

The very nature of OpenTherm control means that as the % demand decreases, so too will the flow temperature of the boiler. From my observations so far based on a boiler with DT20 and the temperature of the appliance not temperature limited (my Intergas test boiler is limited to 70 deg C), the boiler doesn't go out of condensing mode until the % demand is over 60% (which is about 74 deg C flow temp).

Thank you for confirming the way for operation.
I still convinced that there is no reason to not open valves completely and regulate zone temperature by controlling water temperature if that zone is the only one requesting heat (or even opened 100% for the zone requesting most heat, if more zones request heating).

My boiler is the Viessmann model that does not limit flow temperature (mentioned somewhere on this forum) so in my case the statement as found on the evohome site "By reducing the flow temperature to a minimum as it leaves the boiler, the return temperature is kept below the dew point (55oC) whenever possible, this allows the boiler to operate in condensing mode more often" is debatable.
Unless somehow the return temperature drops below 55 even if the the exit temperature is 75 - 80 degrees, the efficiency of my installation is not maximized. I'll try to measure that.

Is there any tehnical documentation available more than user / installer guides? I'm an embedded systems software engineer and out of curiosity I would like to know how the system works.

Bart wrote:Thank you for confirming the way for operation.
I still convinced that there is no reason to not open valves completely and regulate zone temperature by controlling water temperature if that zone is the only one requesting heat (or even opened 100% for the zone requesting most heat, if more zones request heating).

My boiler is the Viessmann model that does not limit flow temperature (mentioned somewhere on this forum) so in my case the statement as found on the evohome site "By reducing the flow temperature to a minimum as it leaves the boiler, the return temperature is kept below the dew point (55oC) whenever possible, this allows the boiler to operate in condensing mode more often" is debatable.
Unless somehow the return temperature drops below 55 even if the the exit temperature is 75 - 80 degrees, the efficiency of my installation is not maximized. I'll try to measure that.

Is there any tehnical documentation available more than user / installer guides? I'm an embedded systems software engineer and out of curiosity I would like to know how the system works.

Of course there is always improvements that can be made to any system, but what is clear is that Honeywell is happy with how evohome does things.

I think the important point to note is that the OpenTherm phrase Honeywell wrote states 'whenever possible'... From all my testing and understanding this is a true statement.

I run an Intergas boiler here for my testing, which allows the maximum flow temperature to be set to the correct design limits of the heating system irrespective of what evohome calls for with OpenTherm. Ideally you want no more than 75 deg C flow temp on a boiler with DT20 to maximise condensing effect.

With a Viessmann boiler the boiler requires the OpenTherm controller to set the maximum temperature... OpenTherm controls don't have to do this and evohome currently doesn't. Hopefully this can be implemented into evohome's software in the future and this would eliminate some of the issues that are being caused like this.

As for Honeywell technical documentation on OpenTherm, there is none. Everything I know I have been told, found online or worked out myself.

Bart wrote:Thank you for confirming the way for operation.
I still convinced that there is no reason to not open valves completely and regulate zone temperature by controlling water temperature if that zone is the only one requesting heat (or even opened 100% for the zone requesting most heat, if more zones request heating).

I posted this image over on a thread in automated home recently where I measured the relationship between the HR92 valve pin position percentage (measured by option 10 in the HR92 menu) and the heat demand sent to the controller:



Heat demand sent to the controller is always proportional to the radiator valve pin position although as shown in my graph it is not a straight line. Below 30% pin position no heat demand is sent - this is because most valve bodies only start to open and allow water to flow at about 30% pin travel, because for the first 30% of the pin travel the rubber seal in the valve body is decompressing but still sealing. (This percentage can differ on some makes of valve though, which can cause some minor issues)

This makes sense because you don't want to call for heat when the valve is not open enough to allow any water to flow. Similarly, most valve bodies are "fully flowing" by about an indicated 70% pin position, with further pin movement not significantly increasing the flow of water. Only from about 30-70% pin travel is there progressive, variable control of the water flow, outside this range it is either not flowing at all or fully flowing.

You'll notice its a straight shallow line between 30% and 70% and a much steeper line from 70% to 100%. The reason this is probably done is so that in the 30-70% valve position range where the valve has the most direct control over water flow and therefore radiator temperature, the heat demand sent to the controller remains relatively low - only reaching 30% heat demand (30% TPI duty cycle for a BDR91) when the valve pin is at 70% - fully flowing.

So if the amount of heat required is small the heat demand sent to the boiler remains relatively low and the water flow is controlled to precisely control the temperature. However if the temperature can't be maintained with the "full flow" of a 70% open valve, the valve opens further - this doesn't result in any significant increase in flow, however the steeper line on the graph means that it calls for a lot more heat from the boiler resulting in a larger increase in flow temperature.

So the way the system is designed is that it does in fact try to get as high a radiator panel temperature as it can from a low heat demand (<30%) and mostly opens the valve before it starts calling for a higher heat demand.

The problem with your suggestion of fully opening the valve and keeping the heat demand really low when only one zone is active and requires a small amount of heat is that as soon as another zone comes online and causes a large (perhaps 100%) heat demand the flow temperature sky rockets and the valve that was previously fully open now causes that radiator to quickly heat up and the room temperature would massively overshoot.

In your picture you show a heat demand of 29% - which means the radiator valve pin position is at about 70%, so that radiator valve is almost fully flowing, however because it is at 70% and not 100% pin position if another room comes online and causes the flow temperature to increase and results in a small overshoot the valve will quickly close down towards 30% to minimise overshoot, and because it is within the 30-70% range where it has precise control of the water flow it can do so. If it was at 100% to begin with the first 30% of closing the valve would be "lost travel" where no reduction in water flow occurred and there would be a big room temperature overshoot. By the time it closed the valve enough it would be too late to avoid the overshoot.

Truth is that my assumption was that 29% reported for the valve was related to the position of the valve. I will check this of course.
Overshoot - true but since the radio connection is bi directional, the thermostat could inform the valves about water temperature so that they could anticipate an increase or decrease.

Bart wrote:Truth is that my assumption was that 29% reported for the valve was related to the position of the valve. I will check this of course.

It is related to the position of the valve but it is not a 1:1 relationship, the relationship is as I've shown in the graph, and is based on actual measurements of several of my own HR92's.

Overshoot - true but since the radio connection is bi directional, the thermostat could inform the valves about water temperature so that they could anticipate an increase or decrease.

Unfortunately it doesn't work that way. "Bi-directional" doesn't mean information about flow temperature changes is sent to the HR92's.

The HR92 sends temperature sensor readings, heat demand and user initiated set point changes to the controller, and the controller can send remote temperature sensor readings and set point changes to the HR92 as well as configuration data like disabling local overrides or window open detection. Flow temperature is not something that is communicated.

Also, if you use a BDR91 the Evohome doesn't even know what the flow temperature is to tell an HR92. All it can do is control the duty cycle of the boiler relay, but the flow temperature you end up with for a given duty cycle depends on the maximum flow temperature setting on the boiler, the maximum power output of the boiler on full burn and the heat load of the radiators that are currently flowing.

So a 30% duty cycle with all radiators flowing could be a low flow temperature say 40 degrees, while a 30% duty cycle with only one radiator flowing could be a much higher flow temperature, say 70 degrees. So the resultant flow temperature depends as much on the load presented by the radiators as the duty cycle or boiler power output.

OpenTherm allows a specific flow temperature to be requested so from that point of view what you suggest would be possible however the comms to the HR92's don't support passing this information.