Smart TRV's

[Tilerdon]

When I moved into my bungalow 10 years ago the central heating was a nightmare. Hot water cylinder on the ground floor and presumably lack of general use of zone valves back in 1971 when the system was installed meant that 2 pumps had been installed. Firing of the boiler was controlled by a 4 pole contactor which was wired incorrectly so hoy water and heating functions were mixed and confused.
I was unable to persuade any local heating engineer to even come and look at it so I set to myself. I sorted the contactor wiring so that the right pump ran according to whether the call was for heat or hot water but a common return meant there was still some crossover between the systems. I sorted this by moving the cylinder from the bathroom into the roof space closer to the kitchen and fitted zone valves on the two systems. Functionally the system now worked properly. Some radiators already had TRV's fitted and I added them to radiators which did not have them, leaving one in the Hall where the Thermostat was located as manual .

We still had problems getting the system right in terms of room temperatures so I installed a Honeywell Sundial Programmer Thermostat but it didn't really solve the problem.

I'm still very technical so decided to look at Smart TRV's. We probably won't be here for years to come so I looked at a fairly low cost approach , ignoring Hive, Nest and others. I bought unbranded Valve heads which actually work quite well and are extremely quiet in operation. I thought that as it is possible to know when the set temperature has been reached in each room then it seems logical to turn the boiler off when no heat is required by any radiator and turn it on again when any radiator calls for heat - First question - Is this a valid assumption?

My problem is that the valves I have bought seem to have a mind of their own and partially open or close as the room temperature approaches the set temperature either rising or falling rather than opening when the temperature is below the set temperature and closing when the temperature. I'm pretty sure that this is the way a standard TRV would operate and the Smart TRV has no other variable user input other than the Set temperature. It is not able to account for other possible sources of heat in the room and therefore shouldn't make that assumption. Second question - is this a correct summary.

The real issue is `illustrated by an actual happening 2 days ago. The Lounge was set at 22° and the recorded room temperature 22.5° The valve was showing as open at 50%. My external control logic assumes anything less than 51% open is in fact closed and will turn the boiler off if no other valves are open. So the boiler and central heating pump were both off. I decided I wanted a shower. The Shower room was set at 15° and the actual temperature was 19° so the valve opening was 0%. I set the temperature to 22° and the valve opened to 100%, the boiler fired, the pump ran shower warmed up nicely but the Lounge radiator got a new heat hit because it was 50% open. The lounge suddenly became uncomfortably warm as the large radiator dissipated that heat.

Final question for now if you aren't totally bored - Is it reasonable to go back to the valve manufacturer (I'm led to believe that my supplier can do this) and request a modification to the software in the valve and suggest as it stands, it is not fit for purpose?

My boiler by the way is an Oil fired Boulter Classic which I believe to be the original installed in 1971

 

[Chuck]

I thought that as it is possible to know when the set temperature has been reached in each room then it seems logical to turn the boiler off when no heat is required by any radiator and turn it on again when any radiator calls for heat - First question - Is this a valid assumption?

Not really, because boilers have a minimum output power.

My problem is that the valves I have bought seem to have a mind of their own and partially open or close as the room temperature approaches the set temperature either rising or falling rather than opening when the temperature is below the set temperature and closing when the temperature. I'm pretty sure that this is the way a standard TRV would operate and the Smart TRV has no other variable user input other than the Set temperature. It is not able to account for other possible sources of heat in the room and therefore shouldn't make that assumption. Second question - is this a correct summary.

Not really. Broadly speaking, traditional TRVs use the expansion and contraction of a working substance (for example melting wax) to throttle the flow, i.e. 'proportional (P)' type control over a band of a few degrees between fully open and fully closed. Smart TRVs can employ more sophisticated control strategies and these vary between makers and models. What you have described could either be a form of 'proportional+differential (PD)' control or possibly a heuristic method that treats rising and falling temperatures differently perhaps as a way of introducing hysteresis to prevent 'hunting'.

The real issue is `illustrated by an actual happening 2 days ago. The Lounge was set at 22° and the recorded room temperature 22.5° The valve was showing as open at 50%. My external control logic assumes anything less than 51% open is in fact closed and will turn the boiler off if no other valves are open.

I can't see that working. The controller needs to shut down the boiler when the emitter system is dissipating less than the boiler's minimum output.

Final question for now if you aren't totally bored - Is it reasonable to go back to the valve manufacturer (I'm led to believe that my supplier can do this) and request a modification to the software in the valve and suggest as it stands, it is not fit for purpose?

There's no law against trying. Whether you are successful or not will probably depend on how many of the modified valves you want. Don't expect to get an answer unless you're placing an order for a very large number.

The conventional way to set up a domestic heating system is to balance the radiators with all TRVs fully open so that the house can be maintained at the correct temperature by the system thermostat, normally in a hallway, which turns the boiler on and off. The TRVs are there to trim out variable external influences such as solar gain reducing the heat required in south facing rooms on sunny days. If you want to use TRVs to make every room an independently controllable zone you really need a buffer tank to act as a thermal reservoir that removes the 'minimum output power' constraint. People who install 'smart home heating controls' often overlook this point and the system performs below par as a result.

 

[Tilerdon]

Thanks for that quick reply Chuck

I did think I should try and provide a little more detail about the boiler itself because I haven't really looked at it for a long time.

It is in fact a Boulter Classic 70/90 commissioned in 2001.

Not really, because boilers have a minimum output power.

The only external electrical connection is a 3 core 240V supply cable and the only internal control is a knob without any real markings on it which is presumably to set the actual primary circuit temperature and to set the output power within the range.

Surely it is better to turn the boiler off when none of the rooms being heated require heat than to control this from a central thermostat.

Not really. Broadly speaking, traditional TRVs use the expansion and contraction of a working substance (for example melting wax) to throttle the flow, i.e. 'proportional (P)' type control over a band of a few degr ees between fully open and fully closed. Smart TRVs can employ more sophisticated control strategies and these vary between makers and models. What you have described could either be a form of 'proportional+differential (PD)' control or possibly a heuristic method that treats rising and falling temperatures differently perhaps as a way of introducing hysteresis to prevent 'hunting'.



I can't see that working. The controller needs to shut down the boiler when the emitter system is dissipating less than the boiler's minimum output.

The valveheads I have bought do have very poor documentation but they do have one parameter that can be set which is defined as Control Type of Valve and the options of either PID or ON/OFF. I have set these to ON/OFF . I would expect the valves to operate in the way you describe with the PID option selected but would also expect them to be open or closed when the ON/OFF otion is selected - Does that seem unreasonable. As I said, the documentation is bad and doesn't define what the effect of choosing between the option actually does.

 

[Chuck]

The valveheads I have bought do have very poor documentation but they do have one parameter that can be set which is defined as Control Type of Valve and the options of either PID or ON/OFF. I have set these to ON/OFF . I would expect the valves to operate in the way you describe with the PID option selected but would also expect them to be open or closed when the ON/OFF otion is selected - Does that seem unreasonable.

Firstly, ON/OFF control requires hysteresis, i.e. ON a little below setpoint and OFF a little above setpoint, to avoid instability sometimes known as 'hunting'. Secondly, you don't really want TRV's to snap closed instaneously because that will result in a sudden rise in the return temperature, which when it arrives back at the boiler may cause lock-outs or other issues.

Obviously, the details of what will actually happen are specific to each system and boiler. I don't recall (if I ever knew) how your boiler would behave and I haven't tried to look it up.

 

[Ric2013]

I have no prior knowledge of your boiler. If it be old enough to work on a semi-gravity system and if it has a large internal heat exchanger you may find it would run quite happily without any minimum flow. Some old boilers eg the Housewarmer range, with no flow at all but a call for heat from the thermostat, would just warm up to set temperature and then stop firing until they started to cool down.

 

[Tilerdon]

I have no prior knowledge of your boiler. If it be old enough to work on a semi-gravity system and if it has a large internal heat exchanger you may find it would run quite happily without any minimum flow. Some old boilers eg the Housewarmer range, with no flow at all but a call for heat from the thermostat, would just warm up to set temperature and then stop firing until they started to cool down.

Hi Ric

Yes this is no measly wall mounted affair. The hunk of cast iron above the top of the burner is about 40cmx40cmx37cm and there is no other way to control it other than to switch the mains on and off

 

[Ric2013]

Controlled by a thermostat in a pocket on the heat exchanger itself as opposed to clipped onto the flow pipe leaving the lump? No pump over-run or overheat cut-out? If I'm guessing correctly then I would also guess the effect of being switched off suddenly will be a heat surge and possibly near-boiling inside the boiler and then everything settles and that's only if you're running the boiler at 80d egrees int he first place.
Even if I'm wrong and the boiler installtion instructions require a minimu flow or over-run this can normally be achieved by an ABV or a radiator with the TRV removed such that it is permanently on.
Your smart trv system does sound a bit weird in fairness to you.

 

[Tilerdon]

Controlled by a thermostat in a pocket on the heat exchanger itself as opposed to clipped onto the flow pipe leaving the lump?

Yes

No pump over-run or overheat cut-out?

There is a red reset button on the front of the burner. But other than the thermostat that is it

If I'm guessing correctly then I would also guess the effect of being switched off suddenly will be a heat surge and possibly near-boiling inside the boiler and then everything settles and that's only if you're running the boiler at 80d egrees int he first place.

Surely what I am doing is no different from what the wall Thermostat in the hall would do? I'm also treating a valve opening of 50% as being closed so all the radiators could be at 50% when the boiler was turned off but at least one would be. The boiler doesn't turn back on until at least one radiator is more than 50% open.

Even if I'm wrong and the boiler installtion instructions require a minimu flow or over-run this can normally be achieved by an ABV or a radiator with the TRV removed such that it is permanently on.

The very first radiator in the system which sits just above the pump has its lockshield turned down low and its TRV set at 35° . It heats the airing cupboard and acts as a by-pass.

It may sound odd because I agree it may be very non standard but I already appear to have more accurate control of my room temperatures and my boiler-on time is very short at the moment. We do supplement the central heating with a log-burner during the colder months and how the system reacts when that is brought into play will be the crux.

 

[Ric2013]

Not sure I'm making my point clear. What I meant was I can't really understand the logic of your fancy system (which I take it you would have designed rather differently). It doesn't really seem very well designed unless it has a way of learning the heating characteristics of the house and tweaking how it controls the valves. A bathroom 4°C below setpoint not calling for heat sounds frankly crazy.

I created a completely 'dumb' version of what you have many years ago using a number of zone valves and room thermostats that simply turned the boiler and pump on and distributed the heat to the rooms requiring heat. It was arguably inefficient in that it would run a 0.6kW radiator off a 13kW output boiler (which was cycling madly) rather than waiting for a few rooms to need heat and heat them simultaneously, but because the boiler was similar to yours it coped with the situation. I still think it used less gas than heating the whole house needlessly. It would still technically overheat as the radiators would be red-hot when the valved shut, but because the radiators were not oversized, this was a very minor symptom. Point is, a 'smart' system should be superior!

Chuck commented that your observation I thought that as it is possible to know when the set temperature has been reached in each room then it seems logical to turn the boiler off when no heat is required by any radiator and turn it on again when any radiator calls for heat - First question - Is this a valid assumption? was not necessarily valid as a boiler has a minimum output and I agree with him, as logic is now probably to prevent a modern boiler cycling by keeping above its minimum output and thus run the radiators on tickover and allow the boiler to do the same rather than overheat the rooms slightly, allow them to cool slightly and then repeat. As yours cannot control the flame except on/off, the boiler will have to cycle to keep to the required output. but it should still work. All I can assume is that the TRVs are assuming that the radiators are only large enough to keep the room warm and so will not overheat by providing a pretty decent flow at just below the set temperature. Which obviously won't be the case in the early heating season when your radiators will have excess capacity.

 

[siricosm]

As Chuck mentioned before, the simplest way to control individual rooms via TRVs is by using a buffer tank and a smart pump. Then you don't need any room thermostats or zone valves, and you can get rid of the bypass (maybe the tank would fit there?), and program each TRV for whatever temperature schedule you want in each room.