One for Chris - Heating

[Paul_B]

Chris,

I tried originally to send this to you as a private message but it looks like you've exceeded your quota limit.

Do you have schematics, pics and description of the way you have setup your heating system? I'm not overly happy with the setup I have at the moment using a thermal store and boiler configuration so wanted to see what others are doing.

Paul

[chris_j_hunter]

Hi Paul -

have freed-up my quota ... give me a day or two, will tidy up a diagram ...

Chris

[Paul_B]

Appreciated Chris, many thanks

[chris_j_hunter]

Paul -

herewith ... copy of our status-board ...

of course, what turned-out wasn't quite what we intended, and there were a few additions to fix issues (as per notes) ...

HTH

Chris

[Paul_B]

Thanks Chris although I might need some help in interpreting the details. I'm specifically interested in how you control the heat input to the main thermal store from the heat sources (I don't see a boiler?) and how this is then used to provide heat to DHW and heating in an efficient manor.

Paul

[chris_j_hunter]

Hi Paul -

just to say am putting together some notes ...didn't get time enough today to finish them, but hope-to tomorrow ...

Chris

[chris_j_hunter]

Hi Paul -

OK, as promised, some notes on the why's & wherefore's, effectiveness, efficiency, etc ...

no boiler - for us (new-build), our sums pointed to gas or oil bring more expensive, to install & to run, so we went all-electric ...

our heat-sources are :

- an ASHP (Sanyo CO2 - works at high-pressure, more effective than other types when it's cold outside) ...

- solar-thermal vacuum tubes (NEG Riomay, because twice as effective as other types, on our roof) ...

- plenty of immersion heaters for back-up ...

the ASHP produces up to about 4.5kW, and the Riomay's up to about 6kW (peak Summer sunshine) ... together, that’s between 10 & 30% of what a boiler would produce, so efficiency & control & managing practical issues well become more important than usual ...

for effectiveness / efficiency, basic thermal & thermodynamic considerations pointed to :

- keeping temperatures low (to minimise unwanted heat flows / losses) ...

- maximising insulation (ditto) ...

- smoothing supply & demand (to minimise peaks & troughs, and to avoid playing catch-up) ...

- using counter-flow heat-exchangers ...

- looking after latent heat ...

- minimising temperature differences between inputs & outputs, for every heat engine (eg: heat pumps) ...

so we :

- built walls & floors in concrete (high thermal mass, good smoothing) ...

- run thermal stores as cool as we can ...

- run the ASHP during the day more than at night ...

- use counter-flow HEx where we can ...

- went for UFH (large working area, so can run relatively cool) under tiled floors (good heat transfer, so can run even cooler) and with maximum exposure (we ran it everywhere we could, and we’re keeping furniture & carpets off the floor as much as we can, so it can be run even cooler) ...

- run the heating 365/24/7 (smooth & no catch-up) ...

the ventilation units (Vanvex & Genvex) each include a smaller ASHP, too - in the latter case, partly to be sure the air supplied to the house is warm-enough in winter (so people won't over-compensate with the UFH), and in both cases to recover otherwise-wasted latent heat (in the vapour given-off by people & cooking & the pool) ...

sources & 'stores :

- the vacuum tubes are used to heat the DHW 'store, with a variable speed pump controlled both directly by a Solar Logic and (in a supporting role) by Cortex - using a Thermal Store object & several gates of General Logic to pulse the pump at intervals (as a function of outside light level & temperature, to improve responsiveness / enhance solar gains / avoid boiling & freezing) and also to manage heat-dump ...

- the ASHP is used to heat another thermal store, from which we heat the small pool - again it has its own controller, again it's supported by Cortex (this time using an HVAC object together with some General Logic) to better manage the immersion heaters & to look after heat transfer to the pool (via a Bowman HEx, plumbed for counter-flow) ...

- the Vanvex (cross-flow HEx + ASHP) keeps heat & humidity under control in the basement pool-room, and puts heat extracted into another thermal store, again with its own controller, again with Cortex support (several HVAC objects & some more General Logic, to monitor & intervene when necessary, trigger boost, manage heat transfers to UFH & Genvex) ...

- the Genvex (counter-flow HEx + ASHP) has no thermal store of its own, but again has its own controller supported by Cortex (HVAC objects & some more General Logic, via another HEx, to monitor & intervene if necessary, trigger boost & manage heat input) ...

Cortex manages heat transfers between 'stores (more HVAC objects & General Logic) to spread excesses & make-up short-falls, and arrange heat-dump when needed (first to other ‘stores, then to the pool) ...

Cortex will also manage the instant HW loops (more HVAC objects etc), and already makes sure things don’t lock-up or blow-up (using Thresholds & General Logic & Macros, arranged to respect priorities & avoid conflicts, ensure valves & pumps are switched in appropriate sequences, not-least to ensure partial vacuums aren’t created as things cool-down) ...

using multiple HVAC objects allows better focus & optimisation ...

data & sums :

- we measure flow rates & temperatures in & out of all tanks & coils (pulse-meters wired to Digital Inputs, clip-on thermistors wired to QAIs) ... and this has been essential in getting things working right & with good optimisation - Cortex is a great data recorder & handler, and we'd be acting blind without it all (having both flow-rates & temperatures is essential to really know about heat transfers, gains & losses) ...

- doing the sums holistically has been important, too - things need to work together & making the most of synergies can help a lot ...

- in deciding on investments, we generally went for better than five-year pay-backs ...

heat-exchangers :

- the HEx in the Genvex is counter-flow, the one in the Vanvex is not, the Bowman is plumbed for counter-flow, as are the coils at the bottom of both the Vanvex & the ACV, and the full-height coil in the Sanyo ‘store can be pumped either way, depending on which way heat is to be transferred ...

'sizes & temperatures :

- in deciding settings for the direct controllers & thresholds for the Cortex objects, the aim has been to keep the pool (10,000 litres) at 29-30degC and the house (three-storey, 270m2) at 20-21degC (target profile varies a little during the day, for comfort) while at the same time minimising temperature differences within the system & avoiding transient problems (freezing, boiling, thinking ahead to avoid any need for catch-up, etc) ...

- the DHW ‘store (ACV tank-in-tank, 800 litres) has a TMV set to give water at around 40degC and, when the tank itself goes over 60degC (from solar gain), heat is transferred to the other 'stores, until it falls below 53degC ...

- the Sanyo CO2 ASHP controller is set to keep its ‘store (235 litres) at just 45degC ...

- and the Vanvex ‘store (285 litres) is kept above just 35degC, by its own action & by ‘transfers from the other 'stores, with heat transfer to the other 'stores initiated (by Cortex) should it get much higher ...

we're still trying ideas & optimising - so temperature targets could still change ...

keeping temperatures low works in three ways :

- lower losses, even with good insulation ...

- more solar-gain (cooler the tank, longer it's worth pumping from the 'tubes) ...

- better thermodynamic efficiency - eg: heat pump efficiency depends directly on the difference between input-output (source & supply) temperatures, at every stage - eg: in practice, the main ASHP has a CoP that varies from about 2.0 in cold-winter to as much as 4.0 in what goes for a heat-wave ‘round here ...

NB: haven't been able to measure the CoP of the smaller ASHP (in the Vanvex & Genvex), but it will be around 3.0 - given the temperatures involved (BTW, they are conventional, not CO2) ...

other points :

- when the house is Unoccupied or Vacant, we do let temperatures fall-back a little, house to around 19degC, pool to around 23degC, ‘though we have to careful about this, because the high thermal masses & relatively small heat sources mean it can take a few days to get things up again (we need to find a way to deal with this - some currently unexplored HVAC functionality plus some General Logic gates will likely do it, ‘though we’d still need to find a way to conveniently input & update holiday plans & visitor schedules, perhaps with the help of Cortex Mobile functionality) ...

- we pre-heat (winter) or cool (summer) ventilation system air intake by taking it in via a long-ish big-diameter underground pipe (big diameter, so slow speed, so plenty of opportunity for heat-transfer) ...

- was v.tempted to go for a multi-fuel stove, but the cost was too high (stove, hearth, chimney, fuel-store, draught-avoidance, ash, maintenance, need for back-ups due to basic incompatibility with water-heating & the impossibility of being able to keep it going unattended for more than a few hours, plus basic inefficiency) ... despite all that, still fancied it - but SWMBO used her veto, because of the mess !

performance :

- everything's been running for a while, but we haven't quite finished building, fitting-out, getting thing set-up, meaning jury's still out ...

- but measurements so-far do seem to show it's working pretty well & the bills will be appropriately down on those we suffered with our previous house ...

- one weak-point, has been heat losses on the transfer loops (between sources & ‘stores, and between ‘stores, all about 10m long, each way) - we thought we'd insulated them well, but we should have done more (losses are just a degree or two, per 10m, but it adds-up over time, and consequential temperature increases & heat-gain reductions perhaps more than double the effect) ...

HTH

Chris

[Paul_B]

Chris,

Thank you for the very detailed information. I'm sure it will help me and others when trying to optimise our own systems. I have listed quite a few things from your post to go and explore more.

My own system is retro-fit into a standard built home from 1998. The insulation is ok but nowhere near as much as if I would be building myself. We've invested in a thermal store to act as a buffer for multiple fuel sources over the coming years. At the moment it is only fed from a high-tech Viessmann boiler with weather compensation. The thermal store holds 750l. So far Cortex is only monitoring the four temperatures at the bottom, 1/4, 3/4 and top level take-offs. It does show the thermal store is achieving good stratification, the top recording 55 deg whilst the bottom is at 35 deg. This is great for showers and we are enjoying hot water at mains pressure (the house was originally built using a vented system). However, when the heating kicks in this disrupts the stratifying effect and pulls the hot water temperature down at the top of the store and can lead to cooler water for showers (not a huge problem, but the third shower in a row can suffer a bit).

Central heating is where I need to concentrate my research a bit more. When the boiler was first installed without the thermal store it worked well and was pretty clever at heating the house with the weather compensation. Initially the radiator temperature was high to heat the rooms but soon reduced to a point where it was modulating only circulating water at around 30 or so degrees to the radiators to maintain the heat rather than on and off spike / troughs. The introduction of the thermal store has added complexity, with no real benefit at the moment, I'm not sure this is overly efficient. I have found a novel approach to heating, though, from an Italian company called Emmeti, they offer a combined manifold setup that can run hot and warm. The warm circuits are designed for UFH and the hot traditional radiators, but driven from a manifold. As this is retrofit it seems a nice solution. I'm just trying to work out the best compromise now and in the future as I add solar thermal and a wood burning stove (the whole reason for the thermal store).

Your comment about heat loss in the pipework struck a cord. Our thermal store is in a integrated garage, although the store itself is insulated with 100mm of PUR the garage is now the warmest room in the house. I'm steadily insulating the pipework by either boxing in with spacetherm and / or Kooltherm pipe insulation.

Paul

[chris_j_hunter]

Hi Paul -

>thermal-store ...

interesting ... similar size to ours, too ... our ACV came ready-insulated & we find it loses around 4degC per-day, which seems a lot ...

rather than being in the garage, ours sits in the adjoining plant-room, which is under the stairs & quite small, plus it's filled with lots of other stuff (Genvex & Vanvex, pumps & valves, manifolds, etc) and it's closed-off with access by removable panel, so it's difficult to imagine convection currents of air cooling it down ...

OTOH, plant-room temperatures are just a bit warmer than the house generally, which seems to point to heat going somewhere - we haven't insulated the ventilation ducts yet, so the ones carrying the intake from outside & exhaust to outside are possible culprits, 'though the parts of them that run through the plant-room are only about a metre long (but 150mm diameter) ...

the tank sits straight on the tiled concrete floor, but below the concrete there's the usual foam insulation, so losses in that direction shouldn't be too bad ...

we also wondered about thermal syphoning, but the runs to water outlets are one-way & the secondary flow loops have solenoids on them (open only when the secondary flow pump runs), and the readings from the solar loop don't seem to show any going-on there, either ...

NB: we did find thermal syphoning an occasional problem between the ASHP 'store & the Bowman HEx - the loop was not much more than a metre long, each way, but was largely vertical - adding a two-port motorised valve seems to have fixed it ...

NB: we also had to put two-port 'valves on the UFH loops - each floor is separately pumped. and when just one was on, we had reverse flow through the other ...

the solar loop is pumped at intervals, but (overnight) only for ten seconds in every hour, so losses from that should be v.small ...

>transfer-loops ...

our boxings-in of the loops between ASHP & 'store (horizontal & well-stiffed with fibreglass insulation) & between solar 'tubes & 'store (mostly vertical) are (we thought) well insulated ... but losses are a degree or two end to end (10 & 15m respectively, each way) and the out & back pipes exchange some heat, too, which cuts the gains on both counts (all based on readings from clip-on sensors etc) - it's difficult to imagine the insulation has moved, but in the solar loop we do wonder about convection currents, because we insulated the pipes rather than stuffing, on account of the verticality ... opening both of them up would be a right pain, but we might have to do it ... 'though maybe we should first do our sums again ...

>pumps ...

BTW, all our pumps are variable-speed (0-10V) - in the hope of maximising heat-transfer (slower the better for transfer, but not too-slow for losses) & minimising power consumption (since they're running quite a lot) ... so far, though, only the solar one is being varied - we've not yet been brave enough to try out our QAOs & we need to set-up some logic, too ...

Chris

[JonS]

Most people I've spoken to think my aims of per room zoning is complex but I am blown away by the sophistication and thought that has gone into your systems. Very impressive and a testament to Cortex that it is controlling it all effectively. Interesting info on the practical use of thermal stores I don't have one but imagined they would make heating much more efficient. On the radiator front I am suprised that circulating water@ 30C makes much difference - is it enough to get convection working in reality? Is that the return temp or flow? I made big savings last year reducing the return temp to 50C from the plumbers default of IIRC 65C or 70C. I guess one has to trade the constant running of a boiler and pumps vs occasional running in efficient bursty (once warmed up). Over the very cold nights last year our heating came on for about 20mins every 90 to keep the temp set point in the bed rooms. We have a mix of well and very poorly insulated rooms upstairs. During the day it runs much longer on the UFH. but one smaller room with smaller free floor area is very slow to respond due to impact of furniture blocking the heating element of floor. Our boiler modulates but only 18-24kW, whereas it would be good if that was lower as room loadings vary between ~<1 and 6kW, so I think a smaller boiler would be better to run efficiently, and lower capital cost.
Thanks for posting with such detail
JonS

[chris_j_hunter]

Hi Jon -

for sure, we've found Cortex more than up to the job (its very disciplined object-oriented approach makes for a very manageable set-up that gives good visibility & can meet most challenges) and its data-logging is first-rate, too (which helps a lot in getting things optimised ...

radiators / 30degC - that was Paul's Viessmann boiler, backing-down to 30degC once the house was up to-target ...

with its large surface area, flows to our UFH can be kept quite low - in the range 21 to 28degC, depending on this & that - we've tried to avoid convection, the focus being on getting a comfortable radiant environment, leaving the MVHR arrangements to look-after air quality ...

interesting to hear of your savings through reduced temperatures ... our experience has been that having that as well as even-ness (space- & time) can help in terms of both comfort & economy ...

Chris