My central heating was installed by the previous owner, 20-odd years ago. Works fine, but has the minor irritation that I can't have heating only - I must have the hot water on as well.
However, it is fully pumped and controlled by a three-port valve: to my mind these are the essentials to allow CH independent of the DHW.
Prompted by the controller's decision to pack up, I had a poke around the wiring. The only difference I can see between my current setup and the wiring diagram supplied with the new controller is that I have no link from the DHW Off terminal to the wiring centre (where it should connect to the grey lead to the valve).
Is enabling independent control as simple as running this additional connection? Or am I likely to have to replace the valve (currently a Honeywell F5-9940) with a more modern variant?
On another note, I intend to replace the boiler and cylinder soon, keeping the radiator circuits (but zoning), adding underfloor and possibly a solar input. However, I don't have any real understanding of the design principles I should be following. I've read the FAQs and followed lots of links from here, but can anyone recommend a good resource? I have a mental image of a nice fat book called something like 'Central Heating Design - the basic principles' - does it exist?
That's pretty much all you have to do, and you shouldn't need to replace the valve as long as it's a mid-position valve and not just a diverter valve.
The other thing you need to check is that the cylinder stat (you *have* got one, haven't you?) has a changeover switch rather than just an on/off switch. Just as a HW-off output is needed from the programmer, a HW-satisfied output is required from the cylinder stat. These both supply power to the valve's grey wire under different circumstances - without which the boiler won't fire when there is only a CH demand.
As a matter of interest, what happened in the old setup if HW was on at the programmer but satisfied at the cylinder stat?
All should be clear if you study the Y-Plan wiring diagram in
Don't waste time fiddling with your existing system if you intend to subzone, and perhaps, add a heat bank for the solar. When you convert your system, you will probably have an entirely different layout.
In all probability you will have a traditional heatbank (something like a DPS Pandora) directly heated from a condensing heating boiler (direct heating saves 10C differential and gives lower return temps).
Then, for any actual heating zones coming off the heatbank you have the choice of treating the heatbank like an S-Plan-Plus boiler and using a single pump and a zone valve for each zone, or (probably cheaper and more reliable), coming out to a manifold and just having a pump for each zone and no valves at all (or possibly an anti-gravity circulation flap valve, if the zone is of higher altitude than the heatbank).
Grand - it's always a comfort to have some reassurance ...
Probably a daft question, but how do I tell the difference? Google hasn't heard of the model number (fair enough, it's years old), and I can't see any movement in the wee lever when the timer kicks in. Both outlet pipes do get hot simultaneously though.
Er, yes, I didn't mention the fact that I had a connection from the cylinder stat to the grey valve.
All seems to work as it should. Hot water on, stat satisfied, boiler cuts out. CH on, HW on, cyl stat satisfied, boiler fires.
I haven't worked out the radiator piping yet, but I'd have thought it's
going to be a lot easier to cut into the existing circuits to zone than it is to take up floors throughout the house and repipe. (I do have excellent underfloor access downstairs, hence also thinking of underfloor heating there.)
Aye. Sounds about right. But I'm still hazy on the theory, and I'm not quite sure how a directly heated heatbank/thermal store would work with other heat sources, eg a solar coil. Would I end up essentially preheating water that would then go through the boiler?
I'm sure I remember posts a little while ago discussing the merits of running CH from the heatbank rather than directly. ISTR commentary that running
from the heatbank meant bursts of full power from the boiler, while direct heating meant longer stretches of modulated burn, which would be more efficient and get into the condensing zone. Is this right?
Presumably if I came out to a manifold I could essentially make every room a subzone if I bought enough pumps? This would be overkill, but I can (for instance) see separation of front and back rooms being useful. Or am I best not bothering with too many subzones and just relying on TRVs?
Following up to my own post, presumably rather than connecting from the HW Off terminal at the programmer back to the wiring centre, I could connect from this terminal to the Satisfied terminal of my cylinder stat?
It's just that that makes cabling a *lot* easier (straight through a wall, rather than down into the cellar, across the house, up the other side to the boiler).
Yes, I didn't mean the pipework layouts. These would be identical to what you have now, except that the zones must either terminate near the heatbank, or, alternatively, you must have an unzoned backbone with local zone control (probably with zone valves, rather than pumps to prevent parasitic circulation with shared backbone pipework).
I converted my existing single zone heating system with kitchen boiler to be S-Plan-Plus to a boiler in the loft. This just required a new backbone to the understairs cupboard from the loft. The existing connection between upstairs and downstairs went through this cupboard, so I mounted the zone valves here, with an additional unzoned takeoff to run the kitchen heating and conservatory zones, each with their own local zone valves.
If I was designing a system from scratch, I would run the zones off a manifold with the zoning devices (pumps or valves) all mounted in one place with 15mm or 10mm plastic from there to the rads. However, it is usually easier on a retrofit to have local zone valves and just run new cable, rather than new pipework!
The heat bank acts as a zero pressure differential point. You can have as many heat sources and heat sinks as you can fit pipes. Each heat source or sink is completely independent of the others and is not affected by any pumping or gravity circulation from another appliance.
The success of running a multiple source system is in setting up the thermostats right.
You want the solar to be constant background heating whenever energy is available. Some systems might require a reliable gravity circulated heat dump radiator, too, to avoid boiling when excess power is available.
Any dino fuelled appliance, though, you want to only come on when absolutely necessary. This is achieved by mounting the boiler's thermostat aggressively high and ensuring the heat bank is still large enough to ensure that you can get a bath load and a bit out of the space remaining above the boiler stat position.
Almost, but not quite. The main issue running condensing boilers is getting the return temperature as low as possible. With a dino only system, you may find it gets a lower return temperature running the radiators directly, compared to going via the heat bank. However, there are several methods of allieviating this. Firstly, running directly will reduce the temperature requirements. Secondly, being aggressive on the thermostat position means that the bottom of the cylinder is frequently quite low. Thirdly, the benefits of solar completely outweigh the slight disadvantage of higher return temperatures.
Therefore, if you intend solar, heat bank direct heating on the gas side is strongly recommended. If you have no alternative heat source, then there is a case for running the rads straight off the boiler, although the benefits are marginal.
Indeed you can, although it can be a hefty retrofit to an existing system. You can also fit local zone valves to every room running off an unzoned submain. It can be difficult to hide the zone valves for this, although there are some electronically operated radiator valves that can be used for this purpose. You would still need to run an official zone for the heating in order to provide pump interlock (boiler interlock if run directly).
A good compromise is to zone rooms requiring similar timing together. In my house, I have 5 zones (one not actually operational yet). These are selected for both ease of installation and commonality of use.
Zone A - downstairs reception rooms and hall Zone B - 1st floor bedrooms Zone C - loft bedroom Zone D - kitchen Zone E - conservatory
I find this works very well. I don't find I heat rooms unnecessarily at all, whilst ensuring that all rooms actually in use at the time are very comfortable.
On Fri, 3 Nov 2006 16:19:48 -0000 someone who may be "Christian McArdle" wrote this:-
Drain-back systems avoid this problem altogether.
For vacuum tubes the more advanced controllers have the ability to operate a heat dump system for those occasions when this is necessary (such as possibly a holiday). It pays to have this sized appropriately for the input power and that may mean a pumped heat dump system, with all the considerations of ensuring there is electricity for this. The stage before this is to use the solar panel overnight to cool the thermal store. Some controllers can do this when necessary and this minimises the time any heat dump system may be needed the next day.
A diverter valve gives you either HW or CH - but not both at the same time, since it has no mid position. The actuator would almost certainly only have
3 wires going to it rather than the 5 of the mid-position valve. If you look at W-Plan in the Honeywell reference I gave earlier, you'll see what a diverter valve setup looks like.
So in fact, you *could* have CH without HW in the old setup - by having HW on at the programmer but turning the cylinder stat to minimum - so that it would be 'satisfied' even when the water in the cylinder was cold.
Yes - if the programmer and cylinder are close - but the wiring centre a long way away - that's fine, and achieves the same thing.
Or, as I suggested in my other post, you could get by without the HW-off connection if you left HW switched on at the programmer and turned the cylinder stat right down. If you're planning to totally re-jig the system in the near future, that would be an acceptable short-term solution.
Right, that's a nice easy task to kick off the weekend with then.
But surely that wouldn't actually give me hot water if the stat was permanently satisfied. I have an objection to cold showers ...
T'other issue is that when I talked of rejigging the system 'soon' that means it's within the current five year plan, not the second or the third ones ...
Christian McArdle wrote: [snip and some stuff moved further down]
I'm unclear on 'direct heating' for the heatbank. Is this circulating boiler water through a coil in the heatbank, or is this circulating the heatbank water through the boiler? I was interpreting it as the latter.
[snip]
This means physically high? IOW, stratification means the boiler fires only when the bulk of the heatbank is cool, meaning it results in a long burn? How high do you consider 'aggressive'? Two thirds of the way up?
So assuming a large bath (the planned bathroom refit is going to involve this, I'm probably looking at keeping 100 or 150 litres of water above the stat?
I presume that a largish heatbank (300-500 litres, say), maximises the use gained from the solar input. Is this enough to provide a decent reserve for CH, assuming that this is run from the heatbank? (Victorian semi, currently spacious 3 bed, scope for probable loft conversion and improbable cellar conversion.)
One thing I'm not clear on here is control of each zone. Presumably each one needs its own programmer (or a fancy one that can control multiple zones), or do I just leave the heating on 24/7 and control each zone using a programmable room stat?
[moved from top]
So, I have a CH output from the heatbank. This feeds a manifold. Each takeoff from the manifold has its own pump, and feeds a zone. The pump is controlled to ensure heating for that zone.
Alternatively, the CH output feeds a backbone (I assume this is essentially a pipe loop). This backbone is pumped. The zones are fed from the backbone in parallel, and are controlled by valves. Are the zones themselves pumped?
Hmm. Boiler (and heatbank) location is another can of worms. It's likely to be loft or cellar. Heatbank, especially if large, is probably better in the cellar - shorter pipe runs to most loads, and no weight issues. Instinct says the boiler should be close to the heatbank to minimise pipe losses, but fluing would be *much* easier straight up through the roof or out through the gable. [end of moved bit]
[snip]
I can see zones working well where there is predictable use. Presumably
though they are less helpful where there are more random patterns. Where a room is only used sporadically, am I as well just keeping the TRV turned right down, then turning it up when I want heat?
Trying to think this through, I can see possible zones including future expansion as: a) Front room and hall b) Back reception c) Dining room and kitchen d) Bedrooms and bathroom e) (currently unused) Loft f) (currently unused) Cellar
B is currently only used sporadically but times of use are likely to coincide with A. Possibly easier/cheaper just to abuse the TRV to keep it cool unless occupied?
Allowing for E&F would presumably just mean buying a six port manifold not a four port at the moment.
Hmm. This is getting rather long ... Any tips on other resources for research on zoning etc?
It would have the same effect as turning HW off at the programmer - but leaving CH on - which you seem to want to be able to do. Admittedly, to turn it on again you'd have to turn the stat back up rather than just flicking a switch on the programmer.
If you simply want to stop heating the HW when it gets up to its stat setting, but continue to run the CH, your system does that already!
The only time that the CH-OFF connection on the programmer will have any effect is if you want to run the HW and CH at completely different times (requiring two independent timers) or if you set CH to constant but HW to timed. Under those circumstances the CH wouldn't come on without the extra connection at times when HW was off.
Aye, and it's a touch easier to press a button on the programmer (or rather let the timer do its thing) than to fish out a screwdriver and fiddle with the stat setting.
Yes, but I'm sure I've been unnecessarily keeping it up to temperature after the tank has cooled, when there's no prospect of any DHW demand until the following morning.
And this is exactly what the programmer allows - and what the wiring mods I've just done now permit as well.
Your interpretation is correct, although not all condensing boilers can use direct heating, as it requires an open vented system.
It depends on the size of the heat bank. As you probably want to reliably take a bath, you will need about 100 litres above the stat, so if you are only using a 200l cylinder, you can't have it more than half way up. The higher it goes, the more energy efficiency (you can collect more solar, whilst the lower it goes, the greater reliability you have.
Yes, although CH will soon deplete a heat bank. In the height of winter, you will be largely running off gas. The real benefit is during autumn and spring, where the solar is providing useful input, but you'll probably still use some gas.
Bingo.
Yes.
No, in this case, you would pump the entire heating before the manifold and control with local zone valves. Although theoretically you could zone with pumps, this could be unreliable due to pressure effects over the long backbone.
You just go for smaller zones, right down to a zone per room if necessary.
There are several problems with this approach.
You have to bend down.
You can forget to turn them off again (with a prog stat, you'll probably hit a 2 hour boost).
If the rest of the house is up to temp, the zone stat will be off and turning on the TRV might not be effective for some time.
So presumably to enable direct heating of the heatbank, I need an input from the boiler going in somewhere near the top, and an outlet coming out to the boiler very near the bottom of the tank. This keeps the return temp down and maximises condensing potential.
So ... I'd be looking at a heatbank with direct heating from the boiler as above, triggered by a stat at fairly high level. Solar input provided by a coil at very low level. Radiator circuits taken from a coil (or a heat exchanger?) about the same level as the boiler stat. Mains pressure DHW provided by a coil or heat exchanger with a high level takeoff, pretty much at the top of the tank.
Underfloor heating circuit taken either from the rad supply with a mixing valve, or from a coil at lowish level.
Presumably I'm looking to keep the top of the heatbank as hot as can plausibly be managed, while maintaining the need to condense and maximising solar
input. Does this imply boiler control that fires the boiler when the (high-placed) stat falls to (say) 60C, and ends the burn when the same point is significantly higher (85C?)? Or a second, lower, stat, to end the burn?
And if I have a programmable stat for each zone, where the zones are pretty much down to a room level, I have no need for TRVs?
Presumably if I end up with both rads and underfloor in the same room then I don't want a TRV as it will end up fighting the other heat source, and both can be controlled from the same room stat?
Hmm. Lots of things to think about and try to grasp. Thanks very much for your input!
Yes. You also want to keep the boiler pump speed down to the minimum possible whilst still using its maximum power. This provides low return temps and helps prevent disturbing the stratification.
Solar can also run direct, see what the solar panel system provider recommends and go with that.
I wouldn't use a coil. In fact, I would probably run the rads direct. You could run the rads off a plate heat exchanger if you liked, as it would provide sealed pressurised operation and easy draining down/refilling. However, it would be more expensive.
No, I would take the rads off somewhere near the top. It certainly has to be significantly above the boiler stat, or you wouldn't be able to run the rads off gas reliably.
Yes, via a plate exchanger.
Must be above the gas thermostat, or it won't run off gas. You should probably just run it off the rad manifold. You'll need a blending valve and dedicated pump to run it. You can't expect the heat bank to provide ticely tepid water directly for you.
Differential positioning provides better hysteresis than wide temperature ratings, although both will work just fine.
Yes. You should have no TRVs in a room with a zone stat. Ergo, zone per room = no TRVs. I actually failed on this count. My kitchen zone has a small TRVed interloper from the reception zone for historical reasons. However, the kitchen zone itself runs an extremely powerful fan convector, and barely notices the tiny rad (~400W) at the end of the room. The only side effect is that the kitchen doesn't get quite so cold as it would when the kitchen zone is off.
Yes. I would run both rad and underfloor off the same stat. I wonder if someone markets a programmable stat with two outputs so it can only use the rads when necessary (i.e. initial warmup and uber-cold), whilst using the underfloor to maintain temp?
A stab in the dark I know but some programmers have a jumper link in the back to remove if you have a basic system with gravity hot water and minimal control so check if this is missing. Trevor Smith
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.