Honeywell CM927 holiday mode problem

[Gywndy]

During a recent 2 week holiday I set my CM927 to holiday mode with a control temperature of 14C. Given the ambient weather conditions whilst I was away gas consumption should have been minimal but was actually at least 50% higher than under normal control conditions at this time of year!!

I am also concerned that weekly gas usage is considerably higher than 'normal' based on some 4 years weekly data.

I have checked all the CM927 settings and all seem OK. I have also checked the boiler gas consumption when firing and this is about right. In normal operation the thermostat is controlling at set point - checked by a separate thermometer alongside the CM927.

Does anyone have any ideas what is going wrong please?

 

[D_Hailsham]

During a recent 2 week holiday I set my CM927 to holiday mode with a control temperature of 14C. Given the ambient weather conditions whilst I was away gas consumption should have been minimal but was actually at least 50% higher than under normal control conditions at this time of year!!

When you are at home, what do you set as the 'off' temperature, i.e. overnight or when you are all out?

I am also concerned that weekly gas usage is considerably higher than 'normal' based on some 4 years weekly data.

How long have you had the CM927? Which make and model boiler do you have?

 

[Gywndy]

Thanks for your reply.

I've had the CM927 for about 3 years and it has always worked well. I have a gas fired Halstead boiler (~28 Kw heat input) which is about 12 years old - not a condensing boiler.

The overnight temperature on the CM927 is set at 15C. Experience shows that this is about right to avoid the house getting too cold but low enough so that heating does not come on unless outside temperature is very low. The house is well insulated and holds heat very well.

I realised today that I had set the boiler stat back during the summer so I have put this back to 'normal' setting. (I'm working on the principle that higher flow temperature will result in more rapid room heating and therefore reduced boiler firing time. There's a logic there somewhere!!)

I also noticed that there was frequent short cycling of the system with the boiler firing for a minute or so. The firing cycle was so short that there was no significant heat input to the radiators. Looking at other posts / advice about the CM927 I have now set the cycles per hour (2:Cr) to 3, and the minimum boiler on time (1:Ot) to 5 minutes.

I am now monitoring gas usage daily to see what happens.

 

[D_Hailsham]

I've had the CM927 for about 3 years and it has always worked well. I have a gas fired Halstead boiler (~28 Kw heat input) which is about 12 years old - not a condensing boiler.

3 years is long enough to get meaningful info. 28kW is a big boiler - unless you have a combi.

The overnight temperature on the CM927 is set at 15C.

That's shot down one theory - that 14C was above the normal night temperature, so the boiler would be on longer. However 14C is fairly high for an empty house, all you are doing is preventing it freezing. I set mine to 5C whenever I go away.

The house is well insulated and holds heat very well.

In that case you should not even need it set to 15C overnight.

My house is also well insulated and the overnight temperature is set to 5C. The overnight temperature dropped the other day from 21C to 16C, when it was about -1 outside.

I realised today that I had set the boiler stat back during the summer so I have put this back to 'normal' setting. (I'm working on the principle that higher flow temperature will result in more rapid room heating and therefore reduced boiler firing time. There's a logic there somewhere!!)

I reckon you have found the real cause of your 'problem'!

The low flow temperature will reduce radiator output, so the house will take longer to heat up and might not reach the required temperature if it is very cold outside.

As an example, a '1kw' radiator (measured at 75C/65C) will only produce 620W when running at 60C/50C.

I also noticed that there was frequent short cycling of the system with the boiler firing for a minute or so. The firing cycle was so short that there was no significant heat input to the radiators.

How do you know there was no significant heat input? These thermostats maintain the room temperature within very close limits (±0.5C). If, for example, you need 21kW to heat the house from -1C to 20C you will only need 0.5kW to raise the temperature from 19.5C to 20C.

I have now set the cycles per hour (2:Cr) to 3, and the minimum boiler on time (1:Ot) to 5 minutes.

If your boiler is a high water content/cast iron heat exchanger 3 cph may be OK.

Do you have motorized valves (obviously not if you have a combi)? These can cause problems due to the time they take to move from open to closed. I have my min on time set to 2 minutes because of this.

Is the bandwidth still set to 1.5C? As your house is well insulated you could try increasing it and see what effect it has.

 

[Gywndy]

Thank you for your very extensive and considered reply - very helpful and thought provoking.

To answer your points / comments -

1. The boiler is not a combi - pumped CH and gravity HW with Honeywell V4043 valves on each circuit. The valves open and close (spring return) rapidly and the boiler / pump start is initiated by the valve(s) being fully open.

2. The house has 11 radiators including two towel rails. The boiler was replaced 12 years ago and was sized to allow for some expansion of the system if necessary.

3. I may well have the overnight / empty house temperature set too high but was erring on the side of caution as I don't have a frost stat on the system. Even so, I still cannot understand why the gas usage at a constant low temperature during a two week holiday period was so high compared to a 'normal' heating period. This just doesn't make any sense to me!

4. My comment about there being no significant heat input to the system when it is 'short cycling' is based on observation. For example, I had to get up during last night (for obvious reasons) and heard the boiler / pump start and run for approximately one minute but the radiators were not even lukewarm - i.e. the CM927 had called for heat and a minute or so later had decided that it was satisfied.

5. The prop band is still set at 1.5C. I will try altering this in due course but I think that I should try your other suggestion of reducing the overnight temperature first. (I am still scarred by memories of control systems on large high temperature industrial kilns where prop band was just one of the variables that had to be set!!)

6. I am puzzled by the minimum boiler on time and cycles per hour functions. Even with the min boiler on time set to 5 mins the system will still run for just a minute or so, and I'm sure that it is cycling more than 3 times per hour! Am I missing something?

Thanks again for your help. I will post further observations in due course.

 

[D_Hailsham]

1. The boiler is not a combi - pumped CH and gravity HW with Honeywell V4043 valves on each circuit. The valves open and close (spring return) rapidly and the boiler / pump start is initiated by the valve(s) being fully open.

So the boiler water is pumped round the rads and through the heating coil of the HW cylinder and there is a cold water tank feeding the HW cylinder?

That's called a fully pumped system!

The house has 11 radiators including two towel rails. The boiler was replaced 12 years ago and was sized to allow for some expansion of the system if necessary.

Still sounds oversized to me. Use the Whole House Boiler Size Calculator to see what size boiler you really need; it may be an eye opener!

I may well have the overnight / empty house temperature set too high but was erring on the side of caution as I don't have a frost stat on the system. Even so, I still cannot understand why the gas usage at a constant low temperature during a two week holiday period was so high compared to a 'normal' heating period. This just doesn't make any sense to me!

The CM927 has a built in frost stat, set to 5C, which operates when the stat is set to 'off'. The high usage could have been because the boiler temperature was turned down, so it was having to run longer, even though the required temperature was only 14C.

I ... heard the boiler / pump start and run for approximately one minute but the radiators were not even lukewarm - i.e. the CM927 had called for heat and a minute or so later had decided that it was satisfied.

The time the boiler is 'on' is determined by the deviation of actual temperature from the required temperature, which is related to the 'Proportional Bandwidth. When the deviation is greater than the PB the boiler is on all the time; within the PB the 'on' time is directly proportional to the deviation; e.g at a 0.75C deviation the boiler will be on for 50% of the cycle, i.e. 5 minutes in every 10 minutes for a 6 cycles/hour setting. The fact that the rads were only lukewarm shows that the system is working correctly. They only need to be at full temperature when the house is stone cold and it's freezing cold outside.

I am puzzled by the minimum boiler on time and cycles per hour functions. Even with the min boiler on time set to 5 mins the system will still run for just a minute or so, and I'm sure that it is cycling more than 3 times per hour! Am I missing something?

You're not missing anything; it's just that Honeywell don't tell the whole story. The TPI algorithm is able to modify the cycle rate. The parameters you set are just a starting point.

 

[Gywndy]

So the boiler water is pumped round the rads and through the heating coil of the HW cylinder and there is a cold water tank feeding the HW cylinder?

That's called a fully pumped system!

No - it is pumped CH and gravity feed to the HW coil. See earlier post.

Still sounds oversized to me. Use the Whole House Boiler Size Calculator to see what size boiler you really need; it may be an eye opener!

I have and the result seems ridiculously low! I accept that my boiler is probably oversized. I will address that if and when I replace it.
EDIT: I have just checked the boiler installation manual and I see that it can be adjusted to reduce its rating. As supplied and installed it is on maximum rating. The reduction is marginal (~3.7Kw heat input) but may be worth doing. I will contact the installer that I use and see what he says.

The high usage could have been because the boiler temperature was turned down, so it was having to run longer, even though the required temperature was only 14C.

Agreed! And the set point was probably too high.

The time the boiler is 'on' is determined by the deviation of actual temperature from the required temperature, which is related to the 'Proportional Bandwidth..........

Thanks - useful explanation. (But during the night the radiators were cold even after the boiler had done it's short cycles.) In the great scheme of things the short cycling is not a big issue as the gas usage during these periods is not very significant.

The TPI algorithm is able to modify the cycle rate. The parameters you set are just a starting point.

Again - very useful explanation. It would help if Honeywell added a little more detail about these parameters in the Installation Guide!

Thanks again for your help. I think I'm getting somewhere at last!

 

[D_Hailsham]

So the boiler water is pumped round the rads and through the heating coil of the HW cylinder and there is a cold water tank feeding the HW cylinder?

That's called a fully pumped system!

No - it is pumped CH and gravity feed to the HW coil. See earlier post.

I stand corrected. So many people refer to the standard cold water tank in the roof as a 'gravity hot water' system, although the coil is fully pumped that I assumed you were one of those.

Do you have a motorized valve on the coil circuit (Honeywell C Plan)?

Still sounds oversized to me. Use the Whole House Boiler Size Calculator to see what size boiler you really need; it may be an eye opener!

I have and the result seems ridiculously low!

If your house has had the insulation improved after the boiler was installed it will be a lower output. The figures do appear low, but that's due to the way installers, particularly the old 'Gas Board' used to size systems. They had a monopoly in installing boilers when gas was cheap and, as a selling point, they would guarantee 'minimum temperatures'. The easy way of doing this was to oversize the system! This has been passed down to the independent installers, many of whom served their apprenticeship with the Gas Board.

Fortunately installers are beginning to learn the error of their ways and are using the Whole House system for sizing a boiler. Of course you can do a detailed room-by-room heat loss, but the result will be very close to the Whole House method. I've done both.

 

[Gywndy]

Do you have a motorized valve on the coil circuit (Honeywell C Plan)?

Yes. I had this installed some 26 years ago as there was no control at all on the HW circuit. It works well.

When the boiler was replaced ~12 years ago it was done on a 'size for size' basis. If I remember correctly the old Potterton boiler was 80,000Btu/h so a replacement of the same capacity was installed. No calculation of heating requirement was done. Insulation has been progressively upgraded over the time that I have lived in the house - cavity wall insulation, double glazing, and loft insulation increased to >250mm.

So I'm sure that you are right - the boiler is way over spec for the heating needs of the property. I tried your 'whole house' calculation link and the result came out at about 8kW. I found another calculator which was more detailed and this gave me an answer of around 15kW which seems more likely.

Anyway, hopefully my heating engineer friend is coming tomorrow to check the boiler gas pressure and wind it back to the minimum setting. This will reduce the heat input by around 3.7kW so should have a noticeable effect on gas usage.

Thanks again for your help - I'll post again in a day or two when the boiler has been checked.

 

[Gywndy]

Update!

This post has moved a long way from my original query but the alterations that I have made based on D_Hailsham's suggestions may be useful to others.

The gas pressure to the boiler has been reduced to the design minimum which theoretically reduces the heat input by around 3Kw. Based on the time per cu ft of gas usage I think that the reduction has actually been greater than that and that the boiler output is now somewhere near the 'whole house' calculation of heat requirement.

The boiler stat has been turned back to its original setting (which has worked fine for many years) and gives a maximum measured flow temperature of ~77C.

No sooner had we done this than the temperature plummeted so making any meaningful comparisons of gas consumption were impossible. However the weather during the last 24 hours has been about the same as before we altered the boiler settings and based on this one day my gas usage is down by ~12%. Obviously this is far too short a time for proper comparisons but I am confidently optimistic that I am moving in the right direction.

One unexpected but very welcome result of reducing the heat input to the boiler is that all the kettling, crashing, and banging has completely stopped. (I'm guessing that there was some localised overheating of the water jacket before.)

None of this has anything at all to do with my original question as to why I had used such a lot of gas with my CM927 set in holiday mode for two weeks. I guess that will have to remain as one of life's little mysteries!!