HVAC configuration

[JonS]

I have my heating controlled by Cortex with an upstairs radiator zone and individual zones down stair heated by UFH. The control is way better than achieved by the standard controls but I was wondering how other people had theirs set-up.
For the 2 upstairs floors I have a single HVAC zone at the moment with the temperature averaged between the room sensors. Control is "on/off" with a minimum duration of 20 minutes.
In the UFH zones it’s a 30 minute minimum period (the floors take a long time to react).
All the HVAC objects are connected to delayed occupancy set at 5 minutes. This is generally fine but as there are no doors between kitchen and family room an active parent in the kitchen can suppress heating demand in the family room if the kids a in front of the TV - I like this feature!
How water is timed as I do not have a tank thermometer.

One thing I have observed is that a traditional room-stat setting is really an average with large over and under shoots. The HVAC object's temp is the minimum, but with much less (approx 1C) overshoot.

I am happy but wondering how to improve further. What is other people's experience and with so many options to tweak what have others found most successful optimisations?

Thanks
JonS

[jpdw]

I've had a setup for a about 4 years which is not dis-similar, though I was fortunate to get the radiators zoned by floor - ie zone 3 zone valve for ground first & second (loft) floor.

I've mainly concentrated on using the ability to tune the HVAC demand temperatures to the fraction of a degree per hour per day, gradually tweaking demand down when its not needed (so Cortex is helping my energy miser efforts). Only in a couple of rooms did I add presence to switch between occupied/unoccupied.

We're just nearing the end of an extension/refurb of our kitchen which now has UFH on it's own zone separated from the rest of the ground floor. My intention is to use a couple of temp sensors around the room but not presence sensors for the UFH as advice on here was that the slow heating time of UFH mean waiting for occupancy doesn't really work. I had thought about using an 'aggregate' of bedroom presence to decide upon target temp for the kitchen UFH in the morning, but not got that far.

[chris_j_hunter]

we're still installing (most of the plumbing & half the UFH now pressure-tested, half the electrics also, MVHR installed but too much dust about to run it yet, something over half our 'modules installed, Cortex about to be loaded), so our thinking remains theoretical atm, but ...

with a largely open-plan all-electric layout, we've gone for a fairly liberal spread of peripherally-placed PLH modules augmented by some centrally-placed black-bulb sensors (working into QAIs) & several clip-on pipe temperature sensors (also working into QAIs) that we'll use to modulate the speed of the UFH pumps (one per floor), to switch ventilation & secondary HW flows according to presence & calendar / diary, and move heat between thermal stores (solar / HW & heat-pump accumulators) ...

we've also included some pulse flow-meters, working into DIs, to ensure we can measure heat, not just temperature ...

initially, the plan is to have the pumps & fans on automatic & use just time & the black-bulb sensor readings to simply switch them on & off, keep an eye on the data-logs for all sensors, look for clues, and move to variable speed & sophistication as seems sensible !

Chris

[Paul_B]

Hi Chris,

Any chance you could list and link the products you purchased from the above e.g. black bulb, thermistors, pulse flow sensors?

Many thanks

Paul

[chris_j_hunter]

Hi Paul -

yep ... hopefully later / late today ...

Chris

[chris_j_hunter]

Hi Paul -

OK, as promised :

A. temperature & humidity sensors ...

- generally PLHs - by windows (light-level sensor determines location), and by mirrors & elsewhere (motion sensor determines location) ...

some MFPs ...

and an LTH-X01 (outside) ...

all work well ...

plus individual temperature sensors wired into QAIs ...

- these clipped onto pipes :

http://www.rapidonline.com/Electroni...-sensors-76422

- these inserted into pockets & laid in tile-adhesive & tied to ducts :

http://www.rapidonline.com/Electroni...th-Lead-122892

- and these for room comfort sensing :

http://www.titanproducts.com/categor...lack_Bulb.html

QAI resolution is a bit coarse, but it's OK ...

B. rate meters ...

- water-flow meters :

http://www.jetrange.co.uk/files/broc...r%20Y98043.pdf

ours are ten-litres per pulse - we wanted one-litre per pulse, but they weren't to be had ...

these weren't available when we bought, but look interesting :

http://www.rapidonline.com/Electroni...6-c90b647a6176

but they look to generate too many pulses (our circuits have flow-rates in the range one to ten litres per minutes)

- power meters :

http://www.swissnox.de/bestseller/sw...-35mm-lcd.html

http://www.eltako.com/fileadmin/down..._energy_gb.pdf

C: water-level sensors :

http://www.rapidonline.com/Electroni...-Gentech-30818

use more than one for more detail of level & rate-of-change ...

http://www.titanproducts.com/categor...e_relays_.html

the Titan has a rain-sensor option, but we'd already used one of these :

http://www.hunterindustries.com/irri...ors/mini-clik#

all connected to digital-inputs ...

water-valves :

regular two & three-port by Honeywell :

http://www.bes.co.uk/products/112.asp

E: water-pumps :

http://www.wilo.co.uk/home/products-.../#.UY-whnCVj8s

we used Wilo-Stratos ECO-Z pumps with BMS option for UFH & heat transfer between 'stores, the 0-10V coming from Cortex via QAOs, and a Wilo-Star-Z NOVA for DHW secondary flow - augmented by solenoids, with both pump & solenoids controlled by Cortex via QRI relays ... with three UFH zones on the top floor (driven by available pipe lengths), and four UFH zones on the middle-floor (two plus two half-lengths, again driven by pipe-length availability), all being cast within anhydrous screed with ceramic tile laid-over ... plus a loop for post-heating of the MVHR airflow ...

so-far, we're controlling things via simple Cortex Macros, keeping an eye on how things go & experimenting with various strategies ... we've read-through a good many Cortex Help screens, among them those for HVAC objects, but have yet to choose how to do it ...

lots of sensors are helpful, around the house & on pipes, etc, we found, to be able to understand what's going-on - else it's just guess-work - very precarious !

NB: we've found Rapid v'good to deal with - very fast & reliable, as also BES for all manner of plumbing bits & pieces, Titan were fine, if a bit slow ... SwissNox was via Amazon, also good, and the rest we sourced via local people, from whom we got reasonable prices, with some waiting ...

HTH

Chris

[Paul_B]

Chris,

Many thanks for putting this post together. I know what I will be doing tonight when I get home after work! Did you install solar thermal, solar PV or both during your build? Is the house now complete and do you have any pics?

Paul

[chris_j_hunter]

Paul -

solar-PV - too marginal, even with FiTs, for us anyway - as an add-on (rather than being fully integrated) the cost-benefit seemed about quits if nothing went wrong ...

solar-thermal - yep, vacuum tubes, for which good data here :

http://www.solarenergy.ch/index.php?...L=6&no_cache=1

lots to choose from, and quite a lot of variation in design & installation details ...

quite popular (in the UK) are the Navitrons (SPF-Nrs 781 & 782), but we went for the Riomays (SPF-Nr 939) - about twice as effective in our situation, half from basic design, half from ability to install at optimum angles (Navitrons go down the roof, Riomays go along & can be individually rotated for best elevation) ...

we aligned ours to maximise Winter gains & ease Summer excesses ... and they feed the solar coil of our DHW thermal store (800 litres, ACV tank-in-tank) ... we find them as effective as expected (SPF data) ...

big tank was sized for maximum demand, when family descends, but its size is useful at other times, in that lower temperatures mean more opportunity for solar-gain ...

we bought anti-freeze, but found it a both a nightmare & unnecessary (the tubes being good at capturing heat & not losing it, it takes not a lot of heat on frosty nights to keep them from freezing) ...

keeping them from overheating is an issue, though - due to temperature sensors lagging water temperature rise ...

control is by a IntaEco SolarLogic controller, which varies pump speed to maximise ‘gain, with Cortex keeping an eye & able (by thresholds & a DRB relay) to by-pass it when trouble looms - works well !

we’ve an ASHP as well, and the ‘tubes do a good job at keeping the demand on that down (HW & UFH) - for us, the ASHP offered lower costs than gas (haven’t run it long enough to be sure but, if the Winter just gone is any guide, signs are promising), but it’s sized to match the long-term heat-load rather than short-term fast response (as is usually the case with gas), so things have to be managed carefully (well-insulated house with UFH & plenty of thermal storage - in tanks & structure - is a good match) ...

house - not quite, it’s probably now over 90% done ... the 10% includes quite a lot of work with Cortex (which is v.capable & quite brilliant), to get everything fully-working ...

Chris

[Paul_B]

B. rate meters ...

- water-flow meters :

http://www.jetrange.co.uk/files/broc...r%20Y98043.pdf

ours are ten-litres per pulse - we wanted one-litre per pulse, but they weren't to be had ...

Chris,

I had the same feeling as you that 10L per pulse was possibly a bit coarse for my requirement and after a lot of searching was coming to the same conclusion as you, but then came across the following:

http://www.meters.co.uk/tomiwatermeters.htm

I've got a quote from the guys of ~£56 and gone for a hot water unit with pulse output of 1 pulse per litre, it should arrive on Friday so will see if it fits the bill

Paul

[chris_j_hunter]

Paul -

well done - well worth a try ...

our heat transfer circuits run for hours rather than minutes, so 10 per pulse isn't the end of the world, but it does mean we have to manually interpret the graphs Cortex produces - whereas we were hoping to have Cortex present heat transfer data directly ...

some of our circuits run at precise speeds (speed control has, say, five settings) - so we might be able to add some logic to home-in on the right figures ...

but others are continuously variable, so not so easy - perhaps with some experimentation, we could work-out a calibration ...

still, the data they've produced has been invaluable - in assessing how things are actually working, and giving us the ability to optimise ...

we couldn't find one to work on the solar circuit - which can sometimes get quite hot - but the pump station came with a simple ring-pull device which we were able to calibrate (SolarLogic will run the pump at ten speeds, and the pump has three ranges, so some simple experiments gave us some graphs we could improve (smoothing-out the errors) and cross-correlate with the SPF data for the vacuum-tubes (with the addition of the clip-on temperature data) ...

BTW - temperature sensors ... we find clip-on / strap-on has its limitations - ie: the sensor isn't actually immersed in the fluid, so there are significant lags to be taken into account, when starting & stopping ...

BTW, too - your 0.5degC temperature drop for the tank with 100mm insulation might be optimistic - ours is not much bigger & similarly insulated, and losses are more like 5degC per day (without us using water) - the solar loop is partly responsible, but we've yet to fathom it out ...

Chris