Gas Boiler Modulation

[tuftythesquirrel]

Hello,

Sorry this is a bit long, but I though folks would have a bit more time over the holiday period 😁

 

Could someone explain how gas boiler modulation works in my case. I understand the principle, for example, if you have an Opentherm interface/thermostat and you have a room setpoint of 21C, if the room temperature is 15C, the boiler burner might be operating at 100%, whereas when the room temperature has risen to 19C the boiler burner might have reduced to 10%. I appreciate there is a limit to how much modulation can be achieved e.g. 10:1.

 

I want to implement priority DHW and low temperature heating and also use weather compensation, as recommended by the following nice chaps, Adam Chapman (Heat Geeks), Nathan Gambling (BetaTalk), Kimbo (Heating Academy Northampton), Richard Burrows (Mid Wales Plumbing & Heating Supplies) and others. The aim being to reduce the boiler return flow temperature to maximise the condensing effect.

 

I have a home automation system, that will be controlling an Alpha E-Tec 20S gas system boiler (24kW). The house is all UFH, via 4 manifolds. I have a separate CH buffer tank and DHW cylinder (both unvented). I want to apply separate weather compensation to both the tank temperatures. That is, in winter they will be hotter than in summer, though at different temperatures. I will need to experiment with the amount of compensation required. I have temperature/humidity/motion/light sensors fitted to all rooms and outside the building as part of the automation system. It also has temperature sensors fitted to the CH and DWH tanks.

 

I have 2 relays connected to the boiler to select the CH and PDHW modes. The boiler parameters have separate CH and DHW parameters. e.g. the boiler output for CH can be set much lower than DHW.

The boiler has a weather compensation feature using a NTC thermistor, but this changes the flow temperature. So this doesn’t help me in applying different setpoint temperatures to the tanks.

 

In CH mode, the boiler output can be set to 25% of the rated 27kW. I can limit the CH flow temperature. I can also run the boiler pump in delta T mode to keep the flow and return temperature differential constant. So these features are already causing boiler modulation by turning the output down and will aid boiler condensing, but…

 

Am I correct in saying that because I am using relays and not Opentherm that the boiler won’t fully modulate?

 

PS Piping and electrical drawings attached.

Visio-Piping Diagrams v0.59.pdf 2172_029.pdf

[JohnMo]

I think implementing DHW cylinder weather compensation is somewhat over thinking things and may not be easy to implement and give little or no benefit. If you haven't purchased your DHW cylinder, purchase one suitable for a heat pump with a 3m2 coil, then you can heat it at much lower temps and get a very low return temp.  Trying to heat via a normal coil at low temps gives a small DT and isn't efficient.

 

If you are doing weather comp on the heating, why do you need external control system, your boiler has all this built in, add the external sensor and the Apha diverter valve - job done.  Simple over heat thermostats are all that's needed, set a couple of degrees over ideal room temp, balance system to get ideal room temps.

 

Set your WC curve, balance your loops and the system then looks after itself.  Adding third party equipment just makes a simple system, expensive and messes with things in an unhelpful way.  Efficiency is likely to be low as a result.

 

Based on loads of testing on my system, heating the buffer on a thermostat, will yield an efficiency (gas to UFH heat at the manifold) of around 95%. Leaving the buffer to float on boiler flow/return temps will give somewhere close to 105 to 110%, due to lower flow temps.  My previous complicated system (before simplification and efficiency improvements) used 30 to 50% more gas, depending on the weather.  My last 30 hours of operation (average outside temp -1) gave a system efficiency of 113%.

[Temp]

My understanding is that gas boilers modulate the burner to keep the flow temperature to a set/target value. So if the flow rate changes due to TRV closing the boiler output power reduces automatically. In that case you would just use weather compensation to control the set/target temperature.

[tuftythesquirrel]

2 minutes ago, Temp said:

My understanding is that gas boilers modulate the burner to keep the flow temperature to a set/target value. So if the flow rate changes due to TRV closing the boiler output power reduces automatically. In that case you would just use weather compensation to control the set/target temperature.

@TempI thought that the Delta T function for the pump did that i.e. pump speed varies to keep the flow and return temperature difference the constant. No TRV's on our UFH but I think the principle still applies. Also I thought the modulation meant at a large setpoint - actual temperature difference the boiler would be producing more power than with a small one. Maybe both of these things control the burner.

[tuftythesquirrel]

20 minutes ago, JohnMo said:

I think implementing DHW cylinder weather compensation is somewhat over thinking things and may not be easy to implement and give little or no benefit. If you haven't purchased your DHW cylinder, purchase one suitable for a heat pump with a 3m2 coil, then you can heat it at much lower temps and get a very low return temp.  Trying to heat via a normal coil at low temps gives a small DT and isn't efficient.

I think that might be the case. We already have a Gledhill 300L unvented cylinder with a 25kW coil. I wasn't going to very low temps but just trying to prove that when I do go to ASHP we can keep the equipment we have.

[tuftythesquirrel]

27 minutes ago, JohnMo said:

If you are doing weather comp on the heating, why do you need external control system, your boiler has all this built in, add the external sensor and the Apha diverter valve - job done.  Simple over heat thermostats are all that's needed, set a couple of degrees over ideal room temp, balance system to get ideal room temps.

Unfortunately that boat has sailed and I've already embarked on a house automation system. In each room I can put a temperature/humidity/light/motion sensor (for £80) which is a lot cheaper than a smart thermostat. That will control heating, lighting, humidity and security, so easy to detect lack of occupancy and apply setback temperatures and so on. 24/7 logging of all data and events etc. 

 

Therefore, everything can be controlled from a central location with all functionality implemented in software. Having a few valves and sensors hardwired to a boiler and means I would have idea what's happening or has happened and why is not the way I wanted to go. Not everyone's choice I admit and sorry but my background is in factory automation and I find it difficult to justify any other way.

[Temp]

41 minutes ago, tuftythesquirrel said:

@TempI thought that the Delta T function for the pump did that i.e. pump speed varies to keep the flow and return temperature difference the constant. No TRV's on our UFH but I think the principle still applies. Also I thought the modulation meant at a large setpoint - actual temperature difference the boiler would be producing more power than with a small one. Maybe both of these things control the burner.

 

I think the pump controls delta T but modulation is used to reduce the actual boiler power. 

 

If the boiler output power (size of flame) is constant changing the delta T doesn't reduce the power going to the floor/rads. 

 

Edited December 27, 2022 by Temp

[JohnMo]

Good luck, trouble with ufh is the reaction time, mine is quite extreme at 6 to 8 hours for a change to be seen. So no point relying on thermostats or anything smart.  Been there tried and failed.

 

My night setback starts at 4pm and ends at 1am.

 

My thermostats fitted cost £5, they basically do nothing (being set above the room temp), could remove tomorrow and it would make little or no difference to the heating.

 

Read Jeremy Harris's threads on home automation and trying for a couple of years to get really smart with ufh control.  In the end it was all dumped and a simple +/- 0.1 hysteresis thermostat installed.

 

On my boiler there are three setting that affect modulation, temperature setpoint, max temperature setpoint and gradient.  The delta T is a moving target, the hotter the flow temp the wider the delta T, this defined by the boiler controller.  A 30 deg flow temp will have a 26 degree return temp, as delta T decreases, supply temp is increased, this will increase until return temp is equal to set point or max temperature set point is reached which ever comes first.  The gradient limits how quickly the temperature ramps up (range of 0-15).  Set to O and the gradient is removed, so boiler ramps direct to set point as fast as possible, at low temps boiler trips a few seconds later, this is great for 80 deg flow temps.  Gradient 1 it applies 1 degree per min max, this can cause return temp and delta T to go out limits.  Mine is set to 2 and it lets the boiler run for about 10 to 15 minutes. At a gradient of 4 the boiler stops after about 3 to 4 minutes, not great.

 

I think most boilers have a gradient that can be changed, but the setting is well hidden.  Nothing is written in my installer manual, except a small reference to it, many hours of searching later, I found out what is was and how to change it.

[tuftythesquirrel]

4 hours ago, Temp said:

 

If the boiler output power (size of flame) is constant changing the delta T doesn't reduce the power going to the floor/rads. 

Thanks, good point regarding delta T equating to constant power.

[Adsibob]

6 hours ago, tuftythesquirrel said:

In each room I can put a temperature/humidity/light/motion sensor (for £80)

Can you provide a link to the sensor please, this sounds interesting

[Gus Potter]

Thanks folk for all you comments and time spent writing.

12 hours ago, tuftythesquirrel said:

Sorry this is a bit long,

 

Trying to digest it all at my end and thought I would respond in a way longer than you..

 

I have an ex council house, extended with a hybrid system. Thus new solid concrete slabs and retro fitted timber suspended floors

 

Rads on the top floor  I oversized them to keep the flow / return temperature down to promote the boiler to condensate. Ground floor has a combination of retrofitted UF on the old timber suspended floor with and extenion that has a concrete slab.

 

It sounds dog rough but I made my own manifolds and it is really simple. Few electronics.

 

But it is user friendly and simple to maintain. The house is much bigger now and we use less energy to heat it even though the house is much bigger.

 

If things break I can fix it in an hour roughly.. don't know as have had no problems.

 

Look folks.. this home automation.. have you factored in how often you are going to have to replace the different parts? Also who other than you is going to be able to fix it? You could be doing something that detracts from the house value?

 

I'm all for technology, I'm a pretty innovative designer, love working with new materials and learning about their behavoir.. but I also as a designer I need to give best advise to my Clients and that includes ongoing potential maintenance and often environmental cost. Relpacemnt of parts is also an environmental cost.. intuitively the less parts the lower the environmental cost?  I'm just asking you to think about this, I'm not slagging the concept off.

 

I'm not kidding you.. I got a  guy in to service my boiler.. he has a PHD.. but became a heating enginneer, he had the cheek to say to me you are asking me tecky questions and you are asking me to give away my trade secrets, my response.. hey donkey if you want play that game.. and this is from a guy that works with a regular contractor that I do design work for. I am paying him to be techy with me and want sensible answers.. ground my gears.

 

The point is .. keep it simple.. yes if you want to make your house high tech.. fine but I would not buy your house... unless you can produce a full and verified O & M manual.. otherwise you could buy a pup! High tech can be an appreciation of simplicity and how you get the best out of it.

 

Folks UFH is supposed to be simple that is the elegance of the principle. I'l say agin UFH is supposed to be simple and environmentally friendly! It works great when you design the rest of the house to suit. It's holistic design.

 

I  designed my first  UFH some 25 years ago. The Scandinavians have done it for much longer. Lets just look at the cost of all the controls, flow meters, zoning, automation. It will probalbly be fine at first installation but just wait for two or three years.. then when things start to fail, getting replacemet parts.. good luck to you. Yes weather compensation works but when you look at the lifetime of the buildings and they way folk use it.. I'm not sure about how friendly it all is.

 

The house I have at the moment has hybrid floors, all sorts of nuances but as a kick off I just built a simple manifold DIY.. it's working great, to set it up I just put my hand on the flow and returns pipes and crancked down the really hot ones. yes folk say I should not use gate valves but I did this 25 years ago.. if one leakes.. a couple of quid to fix it. 

 

To say again..UFH is easy to design, the controls can be very simple. There are some basic conservative rules. For 15mm pipe keep the loops under 75m, DO NOT over bend the pipe. MAKE sure you get plenty flow rate.. it can be turned down. It is really simple with common sense. If you are using a pug mix my own view is that you should pressurise the UF pipes to say 4 bar then pour in the concrete.

 

Reasoning is this.. the pipes need to last for 50 years. The heating system may run at about 1.5 Bar but will expand and contract over the length of the runs. They have bends that are in the pug mix.. thus abrasion will occur. The pressure relief valve on a typical system is set at 3.0 bar.. in the round the 4.0 bar pressure at time of laying the pug mix will be just enough to compensats for the screed shrinkage and movement in the sreed plus say movement in the structural floor.. my gut feeling is that this will allow the pipe to last 50 years.

 

To put this into context. I did my first UF system some 25 years ago, have been keeping an eye on others and had hand in others for some 30 years.. have not seen signs of abrasion in the UF pipes in a pug mix to date. To be honest when these pipes were installed they were sold as barrier pipes.. to stop oxygen ingress, I though it was a bit bollocks at the time but the boilers have held up so maybe it was not sales bollocks after all. Nice to be proved wrong.

 

Yes if the weather suddenly warms up there is a lag time.. but all that happens is we say.. it's warm outside and the house is a bit warmer.. but that can happen on a sunny day too as we get a lot of solar gain. The thing is that in the UK UFH is great but from time to time the house may get a little warm.. it is rare.. just open a window or the door! There will probably be folk going in and out the house anyway so any heat model you make it likely a bit rough.

 

Save your money folks and go for the most simple system you can.. have a look at the Sputnik that the Russians sent to the moon. .it had gate valves (look like Screwfix today)  brave boys right enough. Think.. how much is my home automation going to cost vs a Guses "Fintstone" design..

 

If there are any takers I'll be happy to defend my design approach as I think this is the best environmental choice and most cost effective in the long run. There is one caveat.,. you need to make sure we are debating on a reasonable insulated external house envelope.

 

There you go have stuck my head above the parapet.  In other words.. I'm saying that for UFH go for the simple stupid, low maintenance solution. Save yourmoney and spend it on the insulation envelope.. or if you have large areas of glazing then spend you money on sinking the solar gain into an internal wall to be released later.

 

I know folks on BH debate thermal mass.. but that is not what I'm talking about.. more the concept of how a Tromb wall works and how you start with the paint finish and store the solar gained heat from that.

 

 

@tuftythesquirrel

 

When I say simple DIY UF manifold have attached the prototype I talk about above. Yes I know it is probaly a million miles away from what you envisage. But it works, is cheep maintainable and reliable.

 

 

 

 

 

 

 

 

 

 

[SteamyTea]

I am with @Gus Potter

I dislike anything that is over complicated.

I have very basic E7 storage heaters, they have two controls, energy in, power out.  Sometimes I adjust them.

Below is a chart that shows this year's energy usage and temperatures.

This year has been useful as we have had extreme (for UK) heat, cold and dryness (more solar gain).

At the beginning of the year, I was away for a few days at a time a lot, I would turn all the power off and let the house do its things.  This used less energy, but larger swings in temperature.  Heating was totally off by the 14/03/2022 and did not come back on till 15/11/2022, so 119 days of heating and 246 days without.  A quick look at the data shows that at the start of the year, the mean external temperature was 7.5°C and the house temperature was 19.1°C. End of the year mean external temperature was 6.3° (1.2°C lower) and the house temperature was 19.3°C (0.2°C higher).  Energy usage was similar at less than 0.8 kWh/day difference, with the beginning of the year using less.

When looking at the variation in temperatures, at the beginning of the year, the min and max external temperatures were 0.1°C and 15.9°C, with the house swinging between 5.2°C and 23.3°C (this was when I was turning everything off).  The maximum power draw was 10.6 kW.

So far this winter, the min and max external temperatures have been between -2.7°C and 13°C, with the house temperatures between 16.1°C and 22.6°C, maximum power draw has been 8.3 kW.

Two proper temperature distribution would show what is happening better rather than the extremes.

 

[tuftythesquirrel]

9 hours ago, Adsibob said:

Can you provide a link to the sensor please, this sounds interesting

http://www.idratek.com/products/catalogue/plh-001/

 

[tuftythesquirrel]

@Gus Potter Hi Gus, Thanks for the reply and you’ve made some excellent points. You’re not the first person to say keep it simple. @SteamyTea And I do spend part of my working life saying to others (usually younger graduate engineers), this is a great multilayer communications strategy, but do you think the plant electricians are going to be able to understand this?

 

I do think this software solution is simple, but let me explain.

 

I’ve been contemplating which way to go regarding the automation for some time. My background is in industrial automation. Back in the 80’s people (engineers, factory managers etc.) decided that having a massive electrical cabinet with 1000’s of interconnected devices coupled with a mass wiring, out of date, inaccurate drawings was a bad thing and that a software solution was a good thing. So all the devices became 2 wire (L/N or +/-) and everything was interconnected in software. It became easier to maintain and make modifications. It took a few years to get the electricians on board, but that’s how went. It’s kinda gone the same way in the automotive world. My Morris 1000 had very few, if any ECU’s. A new car has (probably) dozens. And yes, you now need a diagnostics computer instead of a screwdriver.

 

From my own perspective, the moment that I started on a scheme where the thermostat is connected to the zone valve, then the zone valve open limit is connected to call the boiler, I would have time travelled back to the 80’s. I appreciate that is where your average home owner currently and maybe heating engineer sits. However, once you start to say, oops, not one thermostat, actually twelve. And they all have to be hardwired back to the plant room 15m away. Oh and I need domestic hot water to take priority over the central heating. Oh and I need setback temperatures if the rooms are unoccupied. Oops, so need to fit motion sensors to the rooms to do this. Ah, with the motion sensors I can also turn the lights off – but which lights and when. How do I connect four lights to one motion sensor without adding a relay or two. Ha, wouldn’t it be great if I could turn on some ambient lighting in the evening. But then I need light sensors. Then the Light sensors need to work with the motion sensors connected to multiple lights. I also somehow have to adjust the light sensor’s sensitivity on a crappy little potentiometer up in the ceiling. The whole thing spirals out of control. Hopefully you might see where I am coming from. I think the home automation software solution could be simpler.

 

Great point regarding who is going to fix it. I thought that myself. Our neighbour has an ASHP. Installed by persons nameless. No pipework diagrams. No electrical drawings. No operating instructions. Three companies have tried to fix it, failed and walked away. So no home automation system but still problematic. Similarly in my world, any house we have ever lived in, when investigating any problem the plumber usually says “I wonder where that pipe goes” and the electrician “I wonder where that cable goes”. Mostly due to a total lack of information.

 

So to make your good self consider buying my home, I did this….

 

I have piping diagrams that show every component (valve, motor, sensor etc.) and the components are identified with tags. I’ve got electrical schematic drawings (30 x A3 sheets) that show every switch, valve, light, motor etc. all labelled, all connected to the automation system. So for example, if you had an issue with a light in the bedroom, it might be labelled B1H02 (Bedroom 1, light 2). In the automation software you just go CTRL+F (Find) “B1H02”. It will then locate it the light. You can then look for the “connections” to the light and see it is connected to switches “B1S02” and “B1S03”. On the electrical schematic and in the software, the switch will be identified “B1S03 – Bedside Switch” and so on. So then each switch can be tested in turn to see where the issue is. A non-electrical person can change the switch since it is ultra low voltage. I’m hoping my wife might be able to do this, without calling an electrician. Though to be fair, I haven’t asked her yet.

 

I’ve also got a construction document (in Google Docs.) where the components are identified. For example if you want an automatic air vent, the Google Doc will have a hyperlink to the supplier I bought it from. I’ve also got a Software User Manual. It shows the homeowner how the basic features are configured and how to operate the system. Mostley, heating and lighting. And a few basic fault finding tips.

 

…. would that convince you, Gus?

 

We’ve tried to fit as much insulation as is physically possible – on a very, very neglected 1950 hall.

Roof - 0.12 W/m²K, 150mm of Kingspan Quadcore composite sheet
Walls - 0.145 W/m²K, 200mm EPS external wall insulation system (EWI Store)
Floor – 0.09 W/m²K, 200mm PIR/Celotex insulation, UFH and 60mm Cemfloor screed (as the finished floor)
Heat load at -2, 4.5kW, Area 350m²

 

I’ve seen your self-made manifolds before and fair do’s amazing. KISS in a nutshell. Flushed and pressurised our manifolds using mains water to 3 bar before screeding, so tick for me.

 

I am well aware of the story of the Americans spending millions to get a pen to write in space – the Russian’s used a pencil.

 

@JohnMo pointed out @Jeremy Harris Jeremy Harris's UFH journey where all the clever UFH stuff got binned. Two guys far more capable than me have come to the same conclusion. I really still do think I may do the same. However, when I make that decision, I won’t have to rewire a single item – I can just change the control software.

 

I don’t think we are a million miles apart – we just have different ideas of simple!

 

A bit like being down the pub, I seem to have meandered off topic a little and forgotten completely about boiler modulation 😆

[MikeSharp01]

7 hours ago, Gus Potter said:

 

 

Tidy enough job.

[SteamyTea]

7 minutes ago, tuftythesquirrel said:

I won’t have to rewire a single item – I can just change the control software.

I use basic Raspberry Pi Zeros for as much as I can.

They change the OS and Python version.  Not a problem at the moment, but may well be in a decade or two.

 

I did some work at Bletchley Park before Dr Sue Black took over.  The guys there were all retired programmers (and very dull and unimaginative).  I remember one telling me that he made a good retirement living doing work for the banks as he was there when they started to work with FORTRAN and they were still using it.

 

All a heating system has to do is keep a building within set temperature limits, if you make those limits to tight, you will spend way too much effort on the control side.

 

KISS

[JohnMo]

 

24 minutes ago, SteamyTea said:

a heating system has to do is keep a building within set temperature limits,

Plus UFH is a slow low temp system, linking to a PIR is point less, the room will still be cold when the person has up and gone elsewhere.

 

40 minutes ago, tuftythesquirrel said:

So for example, if you had an issue with a light in the bedroom, it might be labelled B1H02 (Bedroom 1, light 2). In the automation software you just go CTRL+F (Find) “B1H02”. It will then locate it the light. You can then look for the “connections” to the light and see it is connected to switches “B1S02” and “B1S03”. On the electrical schematic and in the software, the switch will be identified “B1S03 – Bedside Switch” and so on. So then each switch can be tested in turn to see where the issue is. A non-electrical person can change the switch since it is ultra low voltage. I’m hoping my wife might be able to do this, without calling an electrician. Though to be fair, I haven’t asked her yet.

If I walk into a room switch the light switch and my light doesn't come on, I change the light bulb, sorted.  When I go out I switch the light off.

 

Living in houses for last 59 years, never seen a normal light switch need to be changed.  Plenty of bulbs have been changed though.

[SteamyTea]

7 minutes ago, JohnMo said:

If I walk into a room switch the light switch and my light doesn't come on, I change the light bulb, sorted.

In the scheme of things, I have spent very little time replacing lightbulbs and switched.  I changed two single switched to one double one once.  Think that is it.

53 minutes ago, tuftythesquirrel said:

you now need a diagnostics computer instead of a screwdriver

You probably need a screwdriver as well as a computer.

[JohnMo]

Strange how the first diagnosis thought process is the tech has failed, not check the bulb or the fuse.

[Sparrowhawk]

8 hours ago, tuftythesquirrel said:

I have piping diagrams that show every component (valve, motor, sensor etc.) and the components are identified with tags. I’ve got electrical schematic drawings (30 x A3 sheets) that show every switch, valve, light, motor etc. all labelled, all connected to the automation system. So for example, if you had an issue with a light in the bedroom, it might be labelled B1H02 (Bedroom 1, light 2). In the automation software you just go CTRL+F (Find) “B1H02”. It will then locate it the light. You can then look for the “connections” to the light and see it is connected to switches “B1S02” and “B1S03”. On the electrical schematic and in the software, the switch will be identified “B1S03 – Bedside Switch” and so on. So then each switch can be tested in turn to see where the issue is. A non-electrical person can change the switch since it is ultra low voltage. I’m hoping my wife might be able to do this, without calling an electrician. Though to be fair, I haven’t asked her yet.

 

[Software engineer here] Not wishing to derail the conversation, but what have you used to do the schematic drawings?

 

As I reverse engineer my 1930s-and-badly-extended house and discover where the pipes run (and which are disconnexted but still under the floors) and where the cables go I'm thinking that a model to record this information would be most helpful. In my mind I can see a 3D model of the house showing where the pipe runs are, annotated with details; I'm guessing this is done but no idea what that is called or how to do it. You've gone for schematics which is another way to do the same thing, though abstract enough that I find them hard to follow.

 

I've got a couple of floorboards up atm trying to chase down air leaks and came across electrical wiring I wasn't expecting. Be nice not to forget about that by the next time I lift the floor.

 

Do people do this?

[SteamyTea]

1 hour ago, Sparrowhawk said:

In my mind I can see a 3D model of the house showing where the pipe runs are, annotated with details; I'm guessing this is done but no idea what that is called or how to do it. You've gone for schematics which is another way to do the same thing, though abstract enough that I find them hard to follow.

I think there are BMS packages that integrate into the CAD software.

 

[tuftythesquirrel]

On 28/12/2022 at 20:45, Sparrowhawk said:

Software engineer here] Not wishing to derail the conversation, but what have you used to do the schematic drawings?

Sorry, only just saw this. Visio for the piping diagrams and Nanocad v5 (free) for electrical drawings. Never paid for any CAD software🤣