TRVs, flow switches and boiler cycling

[Mike Clarke]

> Is retrofitting a flow switch a practical option without modifying the

>> internal boiler wiring? > >Correct. You may find it works with an additional external pump set to >minimum, if the boiler doesn't mind the extra help! The flow switch output >would connect to the thermostat input on the combi. > >Whether this would work or not depends on the internal design on the combi.

That's an interesting possibility but I'm not sure if it would be possible with our boiler. After living in a house with ducted warm air heating for the last 35 years I'm seriously lacking in experience with radiator systems so any comments and advice will be welcome.

To summarize my original post, we've just bought a house where there is no room stat and only the bedrooms have TRVs. The "standard" approach of a room stat in the living room and TRVs everywhere else doesn't appeal since there will be times when we'll want to heat parts of the house without heating the living room. Separate zones would obviously be the best solution but I don't fancy the task of ripping up floors and modifying the pipework layout to achieve this so I was considering fitting TRVs to all the radiators and relying on the boiler's internal bypass.

I was back at the "new" house last week and picked up the manual for the boiler. It's a Worcester Bosch "350 Combi" (HC350.FSN), probably about 4 years old. The boiler has a single heat exchanger in the combustion chamber, this normally feeds the CH circuit unless there's demand for DHW when the CH pump stops and a separate pump starts up to divert the flow to a water to water heat exchanger to provide DHW. In case my description isn't clear I've put a copy of the boiler water flow diagram on . As I said I'm a newbie to radiator systems so I don't know if this is an unusual design or not.

It looks to me that an extra CH pump would have a detrimental impact on the DHW mode by stealing some of the flow intended for the heat exchanger unless it was connected across the flow and return lines with a motorized valve (or a couple of non return valves) and configured to cut in only when the flow valve sensed that all the TRVs had closed. Since the extra pump would always be pumping into a "blocked" line I suppose it would also need it's own bypass valve. I suspect I'd end up with an overly complex setup just to avoid a bit of boiler cycling.

The boiler manual claims that "the electronic controls prevent rapid cycling of the appliance in the central heating mode" but fails to go into any detail. I suspect this is achieved by an imposed minimum "off" cycle of 3 minutes which is mentioned elsewhere in the manual.

I suppose it all boils down to whether the cost of the fuel wasted in cycling the boiler is sufficient to justify the cost of installing a more complex control system.

[Malcolm Reeves]

That won't work and probably is against building regs which now cover heating systems. If you had a modulating boiler it could work but you would need a bypass or an alpha pump since when all the TRVs closed the pressure would rocket and most likely the TRVs make an awful racket.

Have you thought about a radio thermostat? Not cheap but you could take it from room to room. Then in the room that you want to keep constant turn the TRVs to full and in the old thermostat room down to some level.

Are you sure the CH stops? Usually both pumps run and the flow is shared. It's the same as one pump + 3 port valve only with 2 pumps instead. Boiler drives DHW, CH or CH+DHW.

Don't forget that the entire house is a system. Consider what happens when the thermostat switches on. The water in the rads is at X C. This is pumped through the boiler raising its temperature, heating the rads up, heating the room air up. Eventually room stat clicks off. Water in rads cools, air cools, room stat clicks on. Cycle repeats. So the cycle time is a function of boiler output, amount of water in the rads, amount of air in the room, etc.

If the boiler flow water exceeds its set temp before the room stat clicks off then the boiler short cycles. But as you can see from above this is function of the boiler output, rads etc. You can't really stop it with controls unless you vary something above (i.e. increase rad size etc.).

Basically the boiler is stuffing out X kW. It has to go somewhere, the water, the house fabric, etc. If the boiler output is too high then it will exceed temperature before the room has warmed up and the stat turned off. The solution is to increase the output in the stat room so that the stat turns off first (or increase the power sinking, more rads say). Clearly balancing the system so each rad gets its share of the power is a must.

[Mike Clarke]

It is a modulating boiler and has an internal bypass so it works OK but I am conscious of the energy wasted when all the TRVs are shut as the boiler cuts in from time to time to keep the water circulating inside itself up to temperature. Since this is an existing system it doesn't have to be modified to comply with the more recent regs. OTOH I do need to make some improvements over the current situation where 7 of the 12 radiators have no thermostatic control at all.

That looks like a possible option to go on the list.

Well, the manual states that demand for DHW will override the CH requirement and when DHW is no longer required the burner will extinguish for a waiting period of about 3 minutes before returning to CH state. I can certainly confirm that the burners always shut down for a few minutes after drawing hot water.

[Snip - explanation of room stat operation]

Yes I appreciate that the boiler will cycle in this way when it can't modulate down low enough. This form of cycling is OK, it's saving energy. I was more concerned about the wasteful cycling when all the rad valves are shut and it's just circulating through the bypass. Perhaps I'm just being paranoid and the total heat wasted will be very low compared to the heat used in warming the house at other times.

[Andy Hall]

Given that situation, there is not a lot that you need to do other than to fit TRVs on all of the remaining radiators bar one and then arrange for the thermostat to be in the area of the non-TRV one.

That's fairly typical behaviour.

It is very little energy but if you can avoid it it is better to do so. You will probably find that the boiler adopts a different on/off pattern at the bottom of its range to avoid rapid cycling. It knows from how it has controlled the burner that the requirement is low, and it also knows that this is because there is little or no demand in the heating circuit.

The room thermostat hookup would eliminate it as long as you set the temperature of it and the flow in the radiator appropriately.

.andy

To email, substitute .nospam with .gl

[Christian McArdle]

One method that could be used is to separate the heating circuit from the combi boiler primary, using a plate heat exchanger. Then, the new circuit can have a full time alpha pump, flow switch or whatever, with all TRVs. The flow switch, which detects flow through any radiator is used connected to the combi as the call for heat. The system would work and be efficient, but viability depends on obtaining a cheap plate heat exchanger.

+--------+ PHE | >-------------++---->>---+-FS-----> to rads | Boiler | || BP | > = external pump, run continuously BP = bypass (p. not required if pump is alpha) FS = flow switch, demands heat from boiler PHE = plate heat exchanger

Christian.

[Andy Hall]

Cost is in the £80-100 range......

.andy

To email, substitute .nospam with .gl

[Malcolm Reeves]

I keep forgetting that modulating boilers are common on gas. I don't have gas so I'm oil centric :-), and modulating oil burners are very rare (and expensive).

[Mike Clarke]

Thanks, this looks a much more elegant approach than my idea for an extra pump and valves. Either way it's going to be quite a lot more expensive than just adding a flow switch if that had been feasible.

I'm reassured by Andy's earlier comment about the losses from cycling when the TRVs are all off being low, so I'll probably just go ahead with adding TRVs to the rest of the rads and see how things go. I'll try to get a feel for the losses by monitoring the gas meter for a couple of hours or so with all the rads shut down.

[Christian McArdle]

Remember that such a system would not comply with building regulations, though. If you really do want to continue with such a system, it is essential that the entire primary circuit is insulated to reduce losses.

Christian.

[Mike Clarke]

Agreed, but it will be a big improvement on the existing situation where almost half of the radiators have no thermostatic control at all. Since the system was installed about 4 years ago I assume it doesn't have to comply with the latest regs though I do want to make it reasonably economical to run, providing the cost of any improvements can be recouped from the energy savings.

The pipework does seem to be quite well insulated but I don't suppose this is relevant to the situation where all TRVs are shut and the only flow is through the internal bypass in the boiler.

[Christian McArdle]

The problem comes when the radiators are mostly shut off. In this situation, you would probably want the flow switch to cut the power, as the boiler will be well below modulating minimum and will short cycle. Better to cut totally and wait until the rooms require real levels of heat.

Without the interlock, enough water may leak past the TRVs to keep the primary circuit hot at all times, with the short cycling boiler, so you really need the entire circuit insulated.

Christian.

[IMM]

Your gif is too fuzzy for me to see, so I can't see if there is an internal by-pass. If not, then install one between the flow and return. On the outlet side of the by-pass, bwteen the two tees, fit a flow switch which switches on-off the boiler. When the by-pass opens there will be flow, when it opens further the flowswitch cuts in and cuts out the boiler.

[IMM]

All so complicated.

[IMM]

Ah! Simple. Just fit a flowswitch on the return. When all TVRs are closed the boiler will protect itself by opening its own by-pass and the flow switch will simply turn the boiler off as there is no flow.

What is all the fuss on this thread. My, oh, my!

[IMM]

Just re-read what I wrote. To expand. Use an external by-pass and flowswitch and a timer to switch back on to check if there is still flow (TVRs opening), if no flow then if immediately switches off. If flow it stay on until the flowswitch cuts in again.

[Andy Hall]

Are you sure that's the reason? I can see it all perfectly easily.

You may want to check for hairs on the palms of your hands.......

[IMM]

Will they make the screen go clear?

[Andy Hall]

Doubtful. You'd have to clean it off first.

[Mike Clarke]

But it shouldn't switch off immediately, we'll need a few seconds for the flow to build up. So we're talking about a bit of custom timer logic here or a couple of cascaded timeout switches, probably getting as complicated as previous ideas.