Niall’s virtual diary archives – Thursday 18 November 2021

Thursday 18 November 2021: 22:47. Last post in my series on my future house build I described the house we have designed and submitted for preplanning feedback using Virtual Reality via the Oculus Quest 2, and I did some rough ‘back of envelope’ energy modelling to see if it would meet Passive House energy consumption standards. I concluded by looking forward to seeing if a PHPP model of the house (PHPP is the computer modelling software for Passive Houses) would match my crude calculations.

So yeah, I just went ahead and inputted the house design into PHPP v9.6 straight away and this post is about what it spat out, because it contains quite a few surprises. To remind folk so later makes sense, we intend to meet the Passive House Classic standard and the RIBA 2025 challenge for this house, with stretch goals for the Passive House Plus, the RIBA 2030 Challenge, and the Passive House Premium minus PV generation:

	Required: Irish 2019 NZEB BER A1 House	Required: German Passive House Classic	Required: RIBA 2025 Challenge		Stretch goal 1: German Passive House Plus	Stretch goal 2: RIBA 2030 Challenge	Stretch goal 3: German Passive House Premium
Heat & Ventilation Energy:	≤ 25 kWh/m2/yr	≤ 15 kWh/m2/yr (40% better)			≤ 15 kWh/m2/yr (40% better)		≤ 15 kWh/m2/yr (40% better)
Air Leakage:	≤ 3 m3/hr/m2	≤ 0.6 air changes per hour (72% better)			≤ 0.6 air changes per hour (72% better)		≤ 0.6 air changes per hour (72% better)
Wall & Floor Insulation:	≤ 0.18 W/m2K	≤ 0.15 W/m2K (17% better)			≤ 0.15 W/m2K (17% better)		≤ 0.15 W/m2K (17% better)
Window Insulation:	≤ 1.4 W/m2K	≤ 0.8 W/m2K (43% better)			≤ 0.8 W/m2K (43% better)		≤ 0.8 W/m2K (43% better)
Total Energy Use:	≤ 120 kWh/m2/yr	≤ 60 kWh/m2/yr (50% better)	≤ 60 kWh/m2/yr (50% better)		≤ 45 kWh/m2/yr (63% better)	≤ 35 kWh/m2/yr (71% better)	≤ 30 kWh/m2/yr (75% better)
Embodied lifecycle carbon emissions:	≤ 1200 kg CO2e/m2		≤ 800 kg CO2e/m2 (33% better)			≤ 625 kg CO2e/m2 (48% better)
Potable water consumption:	≤ 125 litres per occupant per day		≤ 95 litres per occupant per day (24% better)			≤ 75 litres per occupant per day (40% better)
Indoor CO2 levels:			≤ 900 ppm			≤ 900 ppm
Indoor VOC levels:			≤ 0.3 mg/m3			≤ 0.3 mg/m3

To clarify:

• Passive House Classic is a low energy house, approximately twice better than the current best rated house under the 2019 EU Near Zero Energy Building (NZEB) regulations (Passive House Classic long precedes EU NZEB, and indeed strongly influenced the design of EU NZEB which is why they look so similar).
• Passive House Plus is a net zero energy house, over a calendar year it generates about as much energy as it consumes by generating energy in summer and consuming an equivalent amount in winter.
• Passive House Premium is a net generating house in that it contributes much more to the grid than it consumes. Because in Ireland there is no feed-in tariff for solar PV, and when they introduce one next year it will pay at desultory rates and has a low annual cap for maximum payment, it is not economic in Ireland to generate much excess electricity on-site. One can still aim for a stretch goal of ≤ 30 kWh/m²/yr total energy use however, and to put that in perspective, what an Irish BER A2 rated house uses only for space heating is what a Passive House Premium uses for everything – cooking, showers, all electricity use etc.

How I configured my PHPP model

Firstly, I must stress the following:

• I have never taken the Passive House Designers Course.
• I basically just filled in the PHPP v9.6 spreadsheet using common sense, a fair bit of head scratching, and occasionally peeking inside the equations to figure out what the hell they mean and/or what the error messages were going on about.
• I did not invest time on inputting detailing such as thermal bridges nor exact shell layer composition.

What I modelled:

• Climate: PHPP defaults for Cork Airport in Ireland, albeit with altitude reduced to my site’s.
• U-Values: Three building assemblies of a single 0.5m thick shell giving a u-value of 0.147 W/m²K, each for ground floor, first floor, roof.
• Areas: Fairly close approximations of shell areas from plans.
• Ground: Fairly close approximations of shell areas from plans with u-value of 0.11 W/m²K. I left all other values at suggested defaults.
• Components and Windows: I approximated Munster Joinery’s Passive House range of windows and doors, Velux’s Passive House range of roof windows, the Zehnder ComfoAir 550 as the MVHR, and the ShowerSave 20 ltr/min waste water heat recovery unit. As per the design described in earlier posts, there is no heat pump employed here.
• Shading: I put full temporary sun protection on the south facing roof windows, as these will have external mechanical shutters. I put 70% shading on the south facing wall windows due to an external mechanical blind on the games room and the summer plants in the greenhouse.
• Ventilation: As my site is windy, I chose a moderately protected site with wind hitting it from all sides. I set the worst possible air tightness, 0.6 ACH, and made the shell air tightness layer somewhat elastic by 10 m² for the whole house. I configured 30 m³/hr of ventilation per person, and chose its suggested default of 180 m³/hr of ventilation for the house. I told it to run that constantly for 24 hours per day. I chose two metres of inlet and outlet air ducts, and a subsoil heat exchanger to represent the earth tube with efficiency of 60%. It reckons that with all that combinated, the total ventilation system should achieve 75.2% effective heat recovery, with the earth tube basically paying for the other losses around the 75% raw efficiency of the MVHR unit.
• Summer Ventilation: I configured an automatic bypass of the heat recovery on the MVHR, and set a 0.32 ACH for ventilation which is +50% of the winter ventilation, as suggested by the help box. Unfortunately the subsoil heat exchanger doesn’t appear to be factored in, so to account for it reducing the incoming air by maybe 60% of the difference to ground I increased that ventilation to 0.66 ACH to make the numbers come out about what look to me about right for a 0.32 ACH via a long deep earth tube.
• Domestic Hot Water + Distribution:
  • For Space heat distribution I configured 10 m of distribution pipe of 15 mm diameter with 24 mm 0.04 W/mK insulation @ 75 C temperature. Radiators are best run at 55 C to prevent them producing a scalded metal smell – or indeed burning people – but a thermal store sits at 75 C, so I have to choose 75 C here (in practice one would use a thermostatic mixer drawing from both the top and bottom of the tank to reduce the radiator temperature).
  • For DHW useful heat I left the defaults as was (16 ltrs/person/day for showers, 9 ltrs/person/day for hot water) but configured a shower drain water heat recovery unit as described earlier for a 16 ltr/min ten minute long average shower with heat recovery into the cold water feed for the thermal store’s DHW inner tank and the shower cold water feed. This gave a 55% whole system shower heat recovery.
  • For DHW distribution I left the flow temperature at 60 C, and configured 20 m of circulation pipe of 15 mm diameter with 24 mm 0.04 W/mK insulation. I added a further 10 m of individual pipes off the DHW circulation pipe with six tap openings per person per day every day of the year.
  • For Storage heat losses I chose a DHW and heating tank with 5000 litres capacity @ 75 C to represent the thermal store. Assuming a 2x2x3 metre airtight box insulated with Kingspan Sauna Satu with a u-value of 0.023 W/m²K = 20 m² that gives a heat loss rate of 0.46 W/K.
• Photovoltaics: I configured sixteen of my cheap 375w Trina honey solar panels on the south facing roof at a 35 degree slope under the assumption that more cheap panels is better than fewer expensive panels. This has a maximum generation capacity of 6 kW. PHPP lets you enter their detailed functionality such as crossover voltage and the inverter’s details, and it thinks this install in my climate and this house would generate 6023 kWh of electricity per year for a materials cost of €4.7k, which is equivalent to €0.04/kWh if amortised over twenty years.
• Electricity: I left the defaults for all the appliances as-is, except to add an always on computer server consuming 875 kWh of electricity per year.

My PHPP model results

Yeah, so PHPP is ridiculously detailed:

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Heating degree hours - External	10.9	9.7	9.9	8.6	7.0	4.9	3.7	3.7	5.0	7.2	8.9	10.5	90	kKh
Heating degree hours - Ground	6.0	5.7	6.3	6.0	5.3	4.7	4.4	4.1	3.9	4.7	4.9	5.5	61	kKh
Losses - Exterior	1428	1281	1307	1134	923	643	486	488	652	943	1173	1382	11840	kWh
Losses - Ground	197	185	207	196	180	162	156	149	142	165	168	186	2093	kWh
Solar gains - North	64	80	170	257	364	398	379	279	204	121	72	48	2436	kWh
Solar gains - East	8	9	19	32	39	41	37	28	23	14	8	6	264	kWh
Solar gains - South	355	335	535	743	726	687	640	602	636	473	395	295	6421	kWh
Solar gains - West	27	35	68	114	129	126	114	99	89	55	35	24	914	kWh
Solar gains - Horiz.	0	0	0	0	0	0	0	0	0	0	0	0	0	kWh
Solar gains - Opaque	52	63	124	203	240	248	226	185	154	95	62	42	1695	kWh
Internal heat gains	458	414	458	443	458	443	458	458	443	458	443	458	5396	kWh
Utilisation factor	99%	99%	92%	72%	56%	41%	35%	39%	51%	84%	97%	99%	66%
Heating demand	670	542	252	41	7	1	0	0	3	91	360	699	2667	kWh

Of the heat losses due to the shell, 929 kWh per annum is lost by ventilation (both intentional and unintentional). The southern glazing accumulates the most heat for obvious reasons, however due to the extensive raised high northern glazing it’s not a small accumulation there, which I found surprising. Internal heat gains (appliances, showers, cooking, people etc) are substantial, and you’ll note they are constant throughout the year. This is ‘baseline’ heating, and too much of it will cause overheating in summer.

Speaking of overheating, PHPP thinks it will be minimal, on four days per year (0.9%) in July might internal temperature exceed 25 C. If I add external blinds on the northern glazing that does fall to 0.5%, which probably isn’t cost beneficial.

Of the 5396 kWh per annum of internal heat gains, 4127 kWh is from electrical appliances and the remainder comes from the thermal store losing heat (222 kWh), domestic hot water (1055 kWh) and the remainder is distribution losses. In terms of monthly, that’s 344 kWh/month electrical appliances and 88 kWh/month domestic hot water.

Finally let’s look at the photovoltaic gains from the roof solar panels month by month, and note that the solar panels will generate more electricity than the house uses for everything for the months of April - September. This will be important, as you shall see later:

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec		Year
Solar radiation on tilted surface	44	51	95	147	160	158	145	127	116	75	52	35	kWh/(m²month)	1204.9	kWh/(m²a)
Ambient temperature	6	6	7	9	11	14	16	16	14	11	8	7	°C	10.5	°C
Total monthly yield	221	258	478	742	800	784	715	629	577	377	262	179	kWh/month	6023.2	kWh/a
Losses through shading situation	0	0	0	0	0	0	0	0	0	0	0	0	kWh/month	0.0	kWh/a

Does PHPP think this house design meets our goals and stretch goals?

PHPP thinks this house design has a Primary Energy Renewable (PER) demand of 38 kWh/m²/yr, which is well under 60 (Passive House Classic) and 45 (Passive House Plus). It is not low enough (though see later) for RIBA 2030 nor Passive House Premium.

Showers and sinks it thinks will consume 25 ltrs/day/person of hot water, which is about 30 ltrs/day/person once mixed with cold down to 40 C. Toilets are supplied by rainwater, so most of the remaining potable water use will come from washing dishes or clothes. Assuming the washing machine and dish washer are run once per day, that adds 15.5 and 4.5 ltrs/day/person, giving a total of 50 ltrs/day/person. Assuming that cooking and drinking water wouldn’t exceed 25 ltrs/day/person, I’d say the RIBA 2030 challenge for potable water use should be met.

The RIBA 2030 Challenge embodied lifecycle carbon emissions is mostly down to what building material you choose – concrete is very bad, wood is good. If you choose wood for everything, unless your builder does something really weird, one should come in easily under 625 kg CO₂e/m².

Finally, because 6kW of PV panel doesn’t generate enough electricity, we don’t meet the generation requirement for Passive House Plus (it requires generation ≥ 51 kWh/m²/yr if PER demand ≤ 38 kWh/m²/yr, which ours just about meets). We would need to fit +50% more solar panels to meet the generation requirement.

Correcting the Primary Energy Renewable (PER)

As I mentioned on a few occasions now, PHPP assumes that you feed your excess PV generated electricity back to the grid. It has no facility for indicating that you’d want to do anything else with that excess electricity e.g. heat the thermal store so nighttime electricity doesn’t have to. An added wrinkle for Ireland is that there will be a cap on the payment for your contributions to the grid, at the time of writing the details are not clear, but it seems that the rebate will be capped to 30% of your total including PV electricity consumption (how they are going to validate this is beyond me, but that’s the current thinking apparently).

We need therefore to correct PHPP so as little nighttime electricity to heat the thermal store is used as possible, which by definition will lower total direct electricity usage and therefore reduce PER demand.

My 5000 litre thermal store if completely full @ 75 C contains 322 kWh of space heating or domestic hot water. One would keep no less than 50 kWh of space heating hot each night, so that leaves 270 kWh to buffer sunny days across overcast days. I thought it would be interesting to also model a 3000 litre thermal store, which contains a maximum of 193 kWh and so its buffer is 143 kWh.

Annual figures from above:

• Total electrical appliance usage: 4128 kWh
• Total space + DHW heating: 3723 kWh
• Therefore total house electricity consumption: 7851 kWh (+ 841 kWh for pumps, ventilation etc)
• Estimated electrical appliance replaced by PV solar during daylight hours: 1296 kWh
• Estimated ‘spare’ PV solar electricity which normally would feed into grid: 4727 kWh

From this we can say that the feed in payments will be capped at 2355 kWh per year, which at €0.065/kWh is €153.

Estimated night time electricity savings for 3000 litre (193 kWh) thermal store

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Max electricity supplied by solar for electrical appliances during daylight	47	55	102	158	172	170	156	137	125	81	56	38
Solar electricity remaining	174	203	377	584	628	614	559	492	452	296	207	141
Domestic hot water heating	88	88	88	88	88	88	88	88	88	88	88	88
Space heating	670	542	252	41	7	1	0	0	3	91	360	699
Tank days between sunshine	6	7	13	30	30	30	30	30	30	24	10	5
Thermal tank contribution	-671	-528	37	454	533	526	471	404	361	117	-345	-717
Thermal store capacity	50	50	87	193	193	193	193	193	193	193	50	50
Mains electricity contribution	87	101	188	348	533	526	471	404	361	117	103	71
Mains electricity demand	-968	-818	-242	-186	-172	-174	-188	-207	-219	-263	-490	-1023

You might notice the ‘Tank days between sunshine’, this is the average number of days that the buffer in the tank can supply all space heating and domestic hot water for the average day in that month. So, if in December it were -6.8 C, the store would run out twice quicker as that is double the differential from the monthly average, whereas if were +13.3 C, the store would run out twice slower (this being half the differential from the monthly average). The 3000 litre tank gets you a bit less than a week in winter average i.e. sunny days, which tend to come in bursts in the Irish winter, need to occur at least once per week. From experience I find this rather too optimistic, so I’ve assumed that due to the limited spare capacity in the thermal store, in the winter it will only capture half of the sunny days and the rest goes back to the grid.

Annual for 3000 litre tank:

• Estimated mains electricity contribution as thermal store was full: 3310 kWh (45% reduction, due to cap earns €153)
• Estimated mains electricity demand from appliances or because thermal store was empty: 4951 kWh (37% reduction, or €288 of night rate electricity saved)

Estimated night time electricity savings for 5000 litre (322 kWh) thermal store

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Tank days between sunshine	11	13	24	30	30	30	30	30	30	30	18	10
Thermal tank contribution	-584	-427	37	454	533	526	471	404	361	117	-242	-646
Thermal store capacity	50	50	87	322	322	322	322	322	322	322	80	50
Mains electricity contribution	0	0	0	219	533	526	471	404	361	117	0	0
Mains electricity demand	-881	-716	-242	-186	-172	-174	-188	-207	-219	-263	-288	-922

‘Tank days between sunshine’ is nearly double for the 5000 litre tank, and I think it likely that a burst of sunny days would occur every ten days or so in winter, so the thermal store should be able to absorb all excess PV.

Annual for 5000 litre tank:

• Estimated mains electricity contribution as thermal store was full: 2631 kWh (44% reduction, due to cap earns €153)
• Estimated mains electricity demand from appliances or because thermal store was empty: 4459 kWh (43% reduction, or €339 of night rate electricity saved)

Something interesting here is that a 5000 litre tank is only about €600 more than a 3000 litre tank, so that is a payback time of twelve years or so given the electricity savings of €51/yr.

In any case, for either size of tank diverting excess solar to the thermal store makes a big difference to mains electricity consumed, primarily because all domestic hot water is covered by solar PV. If we adjust PHPP’s PER worksheet to add a ‘Thermal storage’ offset of mains electricity consumption, with 100% of DHW supply from the solar panels throughout the year and the remainder offsetting space heating:

• For 37% demand reduction (30% of space heating, 100% of DHW) of 3000 litre tank, PER demand drops to 26.9 kWh/m²/yr.
• For 43% demand reduction (48% of space heating, 100% of DHW) of 5000 litre tank, PER demand drops to 24.4 kWh/m²/yr.

Both of these are under Passive House Premium’s maximum energy consumption of 30 kWh/m²/yr, so in this sense both stretch goals 2 and 3 have been reached. However because the generation of electricity by the house is far below what it consumes (1627 - 2188 kWh), it does not meet the minimum generation requirements, and therefore Passive House Plus (let alone Premium) is not achievable.

As much as fitting +50% more solar panels looks tempting to reach Passive House Plus, that adds an additional capital cost of €2.2k or so, and because of the feed in cap, the effect of the added generation on our annual electricity bill of a mere €1000 or so is not going to be much. In other words, it’s not economic. Equally, for just a small added capital outlay one might fit an oversized solar inverter, and if the government in the future remove that cap, adding more solar panels later is easy.

Why did my crude model overestimate total energy consumption by so much?

Summary of my hand made crude model from the last post:

• Primary energy demand for the proposed house would be around 11,000 kWh.
• This is for 260 m² of living space, which gives a 41 kWh/m²/yr which is well below the limit of 60 kWh/m²/yr required for Passive House Classic, and is even below the 45 kWh/m²/yr limit required for Passive House Plus.

PHPP came in at 79% total energy use of my crude model. There were two main causes:

• My crude model assumed far more DHW goes on showers. This is because I forgot to scale the numbers for hot water consumption only (I used total flow), and also because I modelled five adults, whereas PHPP hardcodes modelling 3.2 adults.
• My crude model assumed more air leakage due to doors being opened and closed etc.
• I had basically put a smoothing estimator for the thermal store rather than anything more rigorous like the above, and whilst not a bad estimation for me effectively sticking my finger in the air, it was a bit off.

If I fixed those, it pretty closely matched PHPP which was reassuring!

#house #energy-modelling #PHPP