Opentherm + EvoHome + Viessmann 111-W overshoots

[HenGus]

This is sad. When I asked, some months ago, Viessmann Tech Support confirmed that the 100 series of boilers were 100% Opentherm compatible. I would have gone for one had it not been for the fact that my two local Viessmann installers were not keen on installing something that they know nothing about.

[Don10]

Having just bought a Viessmann 100W to work with Evohome through an Opentherm bridge I am now rather depressed having read this thread. It has just been installed so I am committed. Having read the thread I would be grateful for your advice.
While I am sure one could debate for ages whether Viessmann's implementation of Opentherm is correct it seems to me that Evohome is failing to do what a control should do by demanding a 100% output in all conditions rather than a flow temperature appropriate for the sensed load. It seems to me to be very idle programming by Honeywell and that there are several easy ways for it to find an appropriate starting demand from the information that should be available to it.
What is likely to happen next? I am new to Evohome so I do not know whether Honeywell pay any attention to issues like this and if they do whether it is possible to update the software in the Evohome or bridge to fix it.
In the mean time what do members recommend that I should do? Fit the Opentherm bridge in the expectation of an update or ditch that idea in favour of weather compensation and a straightforward BDR91? I would be very interested in any views.

[bruce_miranda]

Weather and BDR91 gets my vote, especially since the BDr91 is already in the box.

[paulockenden]

it seems to me that Evohome is failing to do what a control should do by demanding a 100% output in all conditions rather than a flow temperature appropriate for the sensed load.

It does exactly that. 100% boiler demand until the temperature approaches the target temp (the 'proportional zone'), then it starts to reduce the demand.

[StephenC]

Having just bought a Viessmann 100W to work with Evohome through an Opentherm bridge I am now rather depressed having read this thread. It has just been installed so I am committed. Having read the thread I would be grateful for your advice.
While I am sure one could debate for ages whether Viessmann's implementation of Opentherm is correct it seems to me that Evohome is failing to do what a control should do by demanding a 100% output in all conditions rather than a flow temperature appropriate for the sensed load. It seems to me to be very idle programming by Honeywell and that there are several easy ways for it to find an appropriate starting demand from the information that should be available to it.
What is likely to happen next? I am new to Evohome so I do not know whether Honeywell pay any attention to issues like this and if they do whether it is possible to update the software in the Evohome or bridge to fix it.
In the mean time what do members recommend that I should do? Fit the Opentherm bridge in the expectation of an update or ditch that idea in favour of weather compensation and a straightforward BDR91? I would be very interested in any views.

Hi Don10.

Yes, it is not great that they don't play well together. I had read a lot about Opentherm ramping down the flow temp as it approached the set point, but in reality, I saw on my setup it wasn't until the set point had been breached by 0.5 degrees did it appear to scale back the temps. With radiators at 80 odd degrees by then, it is too late especially in Spring months. If there was any kind of temp reduction - it wasnt much to be visible and I had no way of telling anyway as there is no logging that we can see. That was consistent and did not change with "learning" over a few weeks in place as is mentioned a lot here.


Hopefully with the info on this forum and very helpful people here you can start ahead of us with more knowledge and see where you get to.

Your house may leak heat like a sieve, and have few (not enough) small radiators to account for heat loss, and therfore high flow temps might not affect you as much.

Don't fret too much, the standard EvoHome with bundled/free BDR91 works really well. I think we probably are fretting about Opentherm for that better control (theoretical for me still), and gas savings - but in reality the BDR91 works well enough.

To be clear I have a Viessmann 111-W Storage combi that has a 46 litre integrated tank as I could not switch from combi to system at the time. This is as far as I can tell the same as the standard and more prevalent 100-w combi (my father has one and looks to be the same) but with much better domestic hot water flow rates (to a point...).


Is yours a 100-w combi or system boiler out of interest? I only ask as with a system setup, domestic hot water (DWH) configuration may affect some settings and operations with Opentherm that may mean you get a different result.

What ch flow temps did you regularly use during summer/spring/autumn/winter months? As long as no other changes are being made, this may give us an idea of how Opentherm with no flow temp limit will affect you.

As for BDR91 with Weather compensation:

1) the Viessmann 100 series boilers have analogue weather compensation with fairly limited control (heating curve *preset curve* adjustment only, not heating curve *slope gradient* adjustment). The 200 series boilers have digital control of both the preset curves and slope gradients to match the house as close as possible. I really need a different gradient than available but it is not possible on 100 series.

EDIT: See page 4 for a better explanation. I can say with experience that I could do with the 200 series Weather comp. flexibility:

http://www.viessmann.co.uk/content/dam/vi-brands/UK/PDFs/Brochures/Weather%20Compensation.pdf/_jcr_content/renditions/original.media_file.download_attachment.file/Weather%20Compensation.pdf

2) I am finding that with EvoHome I set back rooms when not in use/overnight to low temps - say 8 degrees, and now rooms take forever to heat back up to 19 when it is 16 degrees outside, but say 12-13 degrees in one room. Of course I can manually select a higher curve preset , but then that kind of misses the point as the low flow temp objkective will be lost....Weather compensation then is pretty "dumb" in my opinion and only slightly better than fixed temp for me (a person that regularly sets back temps).

Remember most of the energy savings to be had with EvoHome I would say is in setting back rooms to reduce heat loss (which proportionally increases as the delta between inside and outside increases).
I've saved loads without Opentherm to be fair already. I'm chasing the last ~10% savings from low flow temps, increased comfort and faster warmup times with Opentherm to be clear (although not got there yet)


Weather compensation and Opentherm are compared regularly, but in my experience of both I can now see they are completely different and suited to different use cases:

1) Weather compensation is suited to people who runs their heating 24/7/365 heating of their rooms. By definition it is adjusting flow temp to account for heat loss only. Not suited to needing to warm rooms back up again from set back/cold position, as with the correct compensation curve, only limited heat will be available, so long warm up times.

2) Opentherm is much much (vastly) better suited for people who regularly set back rooms or house temps and want a fast warm up time, coupled with low flow temps for the rest of the set point maintenance duration. (If only the max ch flow temp was adjustable on Viessmann boilers)



Maybe a complaint to Honeywell and Viessmann from you as well will get them thinking.

My advice is get the BDR91 connected and take photos of the connections. (I didn't do this before I switched to Opentherm).

Viessmann Weather Compensation and Opentherm both require an adapter cable (circa £3) (the same cable supports connections for both for some reason even though WC works only if OT is disconnected). Both WC and OT are low voltage polarity insensitive connections that can easily be made by you afterwards to test Opentherm.

EDIT: See here for the adaptor cable needed and referred to above that supports both Weather Compensation and also Opentherm at the same time for a 100 series Viessmann Vitodens boiler:

http://viessmanndirect.co.uk/Catalogue/Spares/Vitodens/Vitodens-100-WB1B/Outdoor-Temperature-Sensor-Connection-Lead-7296465

Hopefully this helps, and for what it's worth, I still have my Opentherm Bridge and am keen to see this year if a solution (that is supportable) can be found.

My plan is to connect the R8810 Opentherm bridge back in, but have an inline switch on one of the cables to allow easy switching between BDR91 and Opentherm control in the future and for troubleshooting purposes. Changing binding between the two is also simple and quick.

[DBMandrake]

As for BDR91 with Weather compensation:

1) the Viessmann 100 series boilers have analogue weather compensation with fairly limited control (heating curve *preset curve* adjustment only, not heating curve *slope gradient* adjustment). The 200 series boilers have digital control of both the preset curves and slope gradients to match the house as close as possible. I really need a different gradient than available but it is not possible on 100 series.

EDIT: See page 4 for a better explanation. I can say with experience that I could do with the 200 series Weather comp. flexibility:

http://www.viessmann.co.uk/content/dam/vi-brands/UK/PDFs/Brochures/Weather%20Compensation.pdf/_jcr_content/renditions/original.media_file.download_attachment.file/Weather%20Compensation.pdf

2) I am finding that with EvoHome I set back rooms when not in use/overnight to low temps - say 8 degrees, and now rooms take forever to heat back up to 19 when it is 16 degrees outside, but say 12-13 degrees in one room. Of course I can manually select a higher curve preset , but then that kind of misses the point as the low flow temp objkective will be lost....Weather compensation then is pretty "dumb" in my opinion and only slightly better than fixed temp for me (a person that regularly sets back temps).

The problem you're having is lack of slope adjustment for the weather compensation to match the characteristics of the house. Unless you're really lucky basic weather compensation with no slope adjustment can be more or less useless, as you've unfortunately discovered. The goal should be to adjust the flow temperature with outdoor temperature in the same way that you would do so manually - but to achieve that you need both slope and offset adjustments.

My boiler doesn't support weather compensation and I've actually thought about designing my own basic weather compensation system for it - as I already use an external digital flow temperature controller I would just need to replace that with one of my own devising. There never seems to be enough time to work on projects like that though...

Remember most of the energy savings to be had with EvoHome I would say is in setting back rooms to reduce heat loss (which exponentially increases as the delta between inside and outside increases).

Not sure where you get exponentially from - heat loss is linearly proportional to the difference in temperature on either side of each wall.

The formula for heat loss rate is: Rate = k•A•(T1 - T2)/d

Where k is thermal conductivity of the insulator (wall), A is the surface area of the wall, T1 is inside temperature, T2 is outside temperature and d is the thickness of the wall. As you can see its the difference in temperature that it is proportional to, no exponents.

Weather compensation and Opentherm are compared regularly, but in my experience of both I can now see they are completely different and suited to different use cases:

1) Weather compensation is suited to people who runs their heating 24/7/365 heating of their rooms. By definition it is adjusting flow temp to account for heat loss only. Not suited to needing to warm rooms back up again from set back/cold position, as with the correct compensation curve, only limited heat will be available, so long warm up times.

I'm not sure that I agree with that conclusion - that all depends on setting the slope correctly. If the offset and slope are set correctly the warm-up time should be constant, year round. The change in the flow temperature should perfectly balance the changes in heat loss rate through the walls. Going for a really fast warm-up time in warmer weather by keeping the flow temperature high just leads to overshoots and discomfort due to the excessive surface temperature of the radiators.

2) Opentherm is much much (vastly) better suited for people who regularly set back rooms or house temps and want a fast warm up time, coupled with low flow temps for the rest of the set point maintenance duration. (If only the max ch flow temp was adjustable on Viessmann boilers)

Indeed. Weather compensation and OpenTherm control are almost orthogonal things - they both modify the flow temperature but for different reasons.

Weather compensation reduces the maximum flow temperature when the outside temperature is higher to reduce the heat output of the radiators to match the reduced heat loss through the walls. This can lead to condensing operation even during the warm-up phase in warm weather. (Which OpenTherm on its own can't do) It also helps prevent overshoots, and maintains an overall similar system response time.

OpenTherm adjusts the flow temperature based on individual room demands - if any rooms have a large shortfall in temperature the flow temperature can be increased to "maximum" (whether that is boiler maximum, manually set maximum, or weather compensation set maximum) to get the rooms warmed up quickly (but not too quick) and then when heat demand is reduced when all rooms are maintaining a set point, OpenTherm can throttle back the flow temperature to give condensing operation even in winter. (Which weather compensation on its own can't do)

So the ideal system is clearly a melding of the two - but for it to work properly you need slope adjustment on the weather compensation that has been properly set to match the house, and you need an OpenTherm implementation which respects the maximum flow temperature selected by the Weather compensation - and on your system neither appear to be the case!

[paulockenden]

I notice in this house that there's a significant difference (for the same outside temp) when there's a north wind. Even to the point that I have to raise Evohome temps by a degree or two because it makes the house FEEL colder, even with the same zone setpoints. I'm not sure weather compensation would ever be able to cope with effects like that.

[StephenC]

Not sure where you get exponentially from - heat loss is linearly proportional to the difference in temperature on either side of each wall.

Well spotted and fixed.


Regarding weather compensation and room warm up times from cold/set back, I can assure you it's not ideal having tested the theory for real hence my comments.


For context, traditionally at this time of year I have manually set the ch flow temp to approx 50 degrees c via the analogue dial.


Today was around 13-14 degrees here outside.

I set the ch flow dial such that at 20 degrees, the rooms could be maintained but not increase in temp. The boiler was reporting around 32 degrees c needed (fluctuating due to cycling but most of the time around this figure).

Note at this setting on the dial the recent frost we had (0 degrees c) gave approx 63 degrees c ch flow temp so it is working albeit crudely.


I had a largish room with a 3 metre long double radiator that was setback overnight and half the day and was showing 13.5 degrees c (doors closed).

At the old rate of 50 degrees c I would expect a warm up rate to 20 degrees (HR92 reported) in a little over an hour in this weather. Acceptable in my book.

With weather compensation corrected as far as I can see and the ch flow temp at approx 32 degrees, 3 hours later it still wasn't up to temp (20) only reaching 17.5. Not acceptable nor usable in my book.


So as I said, WC when corrected to match heat loss at a given/desired room temp shows the method is suited (constant low flow temps) to maintain room temps, but not cold starts or setback usage on moderate to high heat loss rooms.


I've gone to the other extreme now!

[StephenC]

People may be interested to know that I have reported Viessmann to the Opentherm Association for standard non-compliance with their 100 series boiler range and had the useful (and backing Honeywell's stance) reply:

I will once again follow this up with Viessmann and see where I get.


Quote:


"Dear Mr. Crown,

I’m chairman of the Technical Committee of the OpenTherm Association. I do agree with you statement that the maximum supply set point is an option for a master device (room control).
The boiler should also work properly with a OpenTherm controller without the possibility to set the maximum supply temperature.
I propose you contact Viessman to solve this issue in a for you acceptable way.

Regards,
Twan Smets"

Unquote


With this information, does anyone think there may be a case with trading standards or similar to force action?

Edit: or maybe a warranty claim (I'm 2.5 years into a 5 year warranty) maybe stating there is now a evidenced bug (intentional or otherwise) in their Opentherm control firmware?

[DBMandrake]

"Dear Mr. Crown,

I’m chairman of the Technical Committee of the OpenTherm Association. I do agree with you statement that the maximum supply set point is an option for a master device (room control).
The boiler should also work properly with a OpenTherm controller without the possibility to set the maximum supply temperature.
I propose you contact Viessman to solve this issue in a for you acceptable way.

Regards,
Twan Smets"

Unquote


With this information, does anyone think there may be a case with trading standards or similar to force action?

I'm not entirely sure that the quote above helps. If the ability to set a flow temperature from the master device (Evohome) is optional, Viessman could argue that in such circumstances when the option is not supported the flow temperature knob on the front will in fact function.

However with the Evohome the flow temperature option is supported, meaning that Viessman's inability to override the flow temperature when the option is supported by the master device is not necessarily contrary to the spec or even covered by the spec.

A bit of reading of the spec would be necessary to see whether override behaviour of the boiler is defined in the case of a master that supports the flow temperature option. Without having read it my suspicion is that this would not be specified as part of the spec and be an implementation specific detail, and that the spec will only say if the flow temperature option is supported by the master the slave (boiler) must honour it. Which Viessman are, a little to literally...

Anyone read carefully through the spec to see what it says about this ?

[StephenC]

I'm not entirely sure that the quote above helps. If the ability to set a flow temperature from the master device (Evohome) is optional, Viessman could argue that in such circumstances when the option is not supported the flow temperature knob on the front will in fact function.

However with the Evohome the flow temperature option is supported, meaning that Viessman's inability to override the flow temperature when the option is supported by the master device is not necessarily contrary to the spec or even covered by the spec.

A bit of reading of the spec would be necessary to see whether override behaviour of the boiler is defined in the case of a master that supports the flow temperature option. Without having read it my suspicion is that this would not be specified as part of the spec and be an implementation specific detail, and that the spec will only say if the flow temperature option is supported by the master the slave (boiler) must honour it. Which Viessman are, a little to literally...

Anyone read carefully through the spec to see what it says about this ?

Two things:

1) Viessmann boilers don't have the ability to set the max ch flow temp when in Opentherm mode.

2) EvoHome doesn't have the ability supported or not to set the max ch flow temp.


I'm not sure you've read this right, and after all, the above two points (related to max ch flow temp) are the issues that need resolving (either one to be honest), and what this thread is about.

[HenGus]

Doesn't The Opentherm Association bear some responsibility if a product doesn't meet the Opentherm protocol requirements?

Quote: To receive a licence for an OpenTherm product, a prototype of the product has to be tested for conformity. These tests are carried out by an independent testing body.

The OpenTherm Association has appointed Gastec Certification as a specialised and independent body to carry out conformity assessments for each new product type. These tests are made according to a specification drawn up by Gastec Certification in consultation with the association. The same test specification is made available to association members Unquote

[StephenC]

Doesn't The Opentherm Association bear some responsibility if a product doesn't meet the Opentherm protocol requirements?

Quote: To receive a licence for an OpenTherm product, a prototype of the product has to be tested for conformity. These tests are carried out by an independent testing body.

The OpenTherm Association has appointed Gastec Certification as a specialised and independent body to carry out conformity assessments for each new product type. These tests are made according to a specification drawn up by Gastec Certification in consultation with the association. The same test specification is made available to association members Unquote

Possibly. I guess it depends on whether what was submitted for testing/was tested meets the spec, and then what was sold was different (seen that before). If it did meet it, then Viessmann are at fault, if it didn't and this got through/wasn't tested/spotted, then that is as you say also a failing of the Opentherm Association.

That is where I think (hope) I'm closer to forcing Viessmann to do something about it - by the looks of it against their wishes as the product (boiler) doesn't meet the specifications they have marketed and sold their actual products as meeting.

I have followed up the email from Opentherm Association with questions around how are Viessmann certified if it is that they are designing outside the specification, and also what they are going to do to protect future/other customers who buy for the Opentherm compatibility - and certification.


At this point, I'm trying to get this sorted for myself, but also importantly make people (e.g. this forum, Viessmann, Opentherm Association, local Gas Safe engineers) aware of these issues so they can make informed decisions based on actual results. Maybe Viessmann will get this properly sorted for a next-gen boiler product/release as it can't be that hard (surely!).

[paulockenden]

Didn't someone mention that Ideal are the same? Obviously some serious issues with OpenTherm certification.

[bruce_miranda]

Do either of these two boilers bear the OT logo on them?