Chris Benson wrote to me (February 2008) having looked at this site, and we discussed what he might be able to do in his own home at ~55°N. Here follows an edited transcript of our emails.
Chris kicked off (after a flattering preamble) with:
A suggestion from a friend of mine who has just done it: If it it's not already been done, insulate any central heating pipes running in the ventilated space under the ground floor. My friend has just insulated about 35m of pipes to get much improved heating and a reported 20% decrease in gas usage. It's not very obvious or as sexy as thermal stores, but has a very good RoI [Return on Investment].
My '70s era CH [Central Heating] piping is also uninsulated (and starts by running through a concrete floor laid directly on earth) so my new super-efficient combi-boiler has been pretty effectively crippled. I'm looking for ways of insulating that avoid pulling up carpets/vinyl and floorboards.
I had to admit that my central heating and DHW [Domestic Hot Water] is so weird that I dare not touch it, which is why I went for solar PV (PhotoVolatics, ie electricity: 1.29kWp [kiloWatts peak] due to be installed 25th/26th Feb). I will however look again at lagging any hot pipes. At least one of our floors is solid (and containing a failed non-standard loop to a now-dead radiator). Plus the boiler has been moved leaving a bizarre re-routing of pipework.
Chris said:
Ah, commiserations on the CH/HW front: ours is not too bad apart from tunnelling though cold concrete, lack of insulation and the boiler being a 30m pipe-run from anywhere solar panels could be fitted.
But congratulations on the PV system: I'll be interested in reading your experiences. The people of AtoB magazine went PV in 2006 and wrote about it at http://www.atob.org.uk/Climate.html which was one of the triggers for my interest. But I need to get our electrical usage (14kWh/day annual average for 2 people) closer to your levels before it makes any sense at all. Of course being at 55N doesn't help with insolation.
We're about to replace our fridge/freezer as that will save well over 1kWh/day, though I'm still not sure about the embodied energy issue and the best thing to do with the old one, eg should I scrap the existing working unit to stop someone else using it, or can I assume that if (say) I freecycle it then it'll displace an even less efficient one? Questions, questions!
Chris gives a potted summary of how is house is now:
These are 3-story late-Victorian terraces, so (pros):(cons):
- solidly built
- relatively small surface area for the volume
- south-facing roof
- council planning dept. have given permission in principle for panels mounted on the S-facing (street-facing) roof.
- relatively small roof
- draughty (I've put in double-glazing throughout, a new front door, under-roof insulation)
- large rooms with high (>3m) ceilings
- solid walls
- v.little garden/yard space for e.g. a large thermal store which is a very attractive idea for me
- a damp-proof course has been injected: we keep soil away from the exterior walls
- built with v.little in the way of foundations: there's dirt a metre or so under the floorboards, which gets wet in wet weather so insulation on the underside of floorboards may also be in the future
- boiler is in the lean-to extension right at the back
My initial dream was to be independent of the grid for up to 48h even during winter -- even if that meant "only" keeping fridge, freezer, lighting and CH running. I didn't realise how ambitious that was!
But reality struck and I keep coming up against the amount of energy I can collect from my little bit of roof and get depressed. The dormer and its shadows limit me to about 6m^2 of unshaded collector. I could fit about 10m^2 with the limitation that the extra 4m^2 would be in shade either morning or late afternoon.
Using the insolation tables from some of the places you've linked to as well as from 'Photovoltaics Design and Installation Manual' http://www.amazon.com/Photovoltaics-Design-Installation-Manual-Sustainable/dp/0865715203/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1202594034&sr=8-1 I'd get about 1.2kWh/m^2/day in winter = 7.2kWh/day if I used all the well-lit area (summer might give 29kWh/day), but from our weekly meter readings, we use up to 20kWh/day in winter down to a minimum of 10kWh/day in summer. 6m^2 of panels would be hard to fund but at current usage (supplying about 1/3 of winter usage), without a good feed-in tariff or net metering I can't justify it at all.
(My wife dreams of retiring to the south coast of Crete where the economics (and insolation!) are considerably more favourable.)
Chris didn't have my minor fridge/freezer-disposal conundrum:
I'm glad I don't have to try to deal with that dilemma. We had a round of failures 4-5 years ago where the fridge, freezer and washing m/c died within months of each other, so I got the most efficient then available (I need to check how efficient they are now!) and didn't have to worry about someone else using them - the fridge and freezer were ~20 years old and scrap!
I've done the easy things: replaced lights that weren't already fluorescent, put 10-way switched power blocks in the rooms with lots of devices that don't need to be on all the time ... So I've been looking elsewhere for power suckers: I have been using meters to monitor individual devices so I know that e.g. it's 0.47kWh to bake a loaf in the breadmaker, that the (large) refrigerator uses ~0.25kWh/day. I really must record all these measurements in one place in one set of units.
I installed a inductive? meter between the meter and consumer unit a couple of months back and found that usage has never been below 350W since I connected it :-( I'm going to have to go through the house and try to account for every last Watt -- I have a horrible feeling there's an old door-bell transformer tucked under the floorboards somewhere :-)
But I've not yet taken the step of downsizing the computing infrastructure: I use a couple of fanless Mini-ITX machines for firewall and wife's workstation, and there're three other diskless machines around the house, and a Mac for watching DVDs, but the house server (usually 130W, but up to 170W when busy) with all the music/video/home-dirs/... is on all the time as are the SqueezeBoxen, so I'm afraid something's going to have to give there if I'm going to get my power usage in-line with what is sustainable.
I'm also starting to think about heat: important up here in the far North :-) Our gas usage is (in old units) between 3 and 5 x100ft^3 per week in summer when it's just DHW and the cooking hob, to 35-38 x100ft^3 per week in winter.
That's been a driver for double-glazing and roof insulation. I can do a bit more insulating in the roof voids, but I'm getting to the point of diminishing returns with that. Draughts through the eaves and under the floorboards seem the most productive areas for attention.
Your thermal reservoir idea is attractive because I don't see better way of storing that much energy so cheaply. I could see that working well if I could gather enough heat during the summer. But I don't have room for the reservoir! Someone living in Finland mentioned that a system of 'blowing hot air through rocks in the basement' during summer, then opening vents from the basement to the house in winter was, if not common, not unusual there. Perhaps I should fill under the floorboards with rubble and try that.
Your comments about vertical mounting of PV/solar panels has made me think about measuring the shadow of the houses opposite during winter to see whether it would be worth having PV and/or solar water panels on the front wall under the eaves to increase my collection area in the winter months ... but I've been prevaricating by trying to think of ways of automating the data-collection instead of just going out and taking a photo every hour!
I had an architect look at the place last year: his suggestion was a well-insulated extension in the back yard enclosing the coldest+darkest part of the plot to retain some of the heat lost through walls, windows and doors. But doing that properly would be even more expensive than filling the roof with PV and ancilliary equipment! Another day-dream is to enclose the whole front of the house from footpath to eaves in a sun-trap/greenhouse affair made of scaffolding poles and polycarbonate sheeting -- but it'll take a complete breakdown of the planning system before I'd be able to get away with it!
My current thoughts are that the best bang for the buck for me will come from draught-exclusion, insulation of external hot water pipes but probably from getting more of the computers off more of the time :-(
Then if/when I can get down to an annual average of 6kWh/day, PV makes sense ... with a feed-in tariff it might actually be profitable! Who knows?
My first thought was that his location might be very suitable for a vertical ground-source heat-pump [GSHP] collection run; especially if he has moving ground water to continue to keep the temperature steady/up. The boreholes can be inside and outside the house perimeter so far as I know...
As regards his potential for capturing solar energy, I'd see 6m^2 as about 1kWp of PV at best (eg using my preferred Sanyo HIT hybrid panels), which in winter will generate an average of ~1kWh/day. Chris says:
Yup. This is the bit that makes me depressed: we're used to such a high density of energy, having to collect it ourselves really brings that home.
[He would] probably capture much more energy as heat towards DHW and even space heating with solar thermal, and it might be possible to use the partially-shaded areas for solar thermal collectors. Chris says:
Yes, the PV Design and Installation Manual makes it clear that even with the fancy diodic(sp?) protection, you do not want to partially shade PV panels. So per my previous message a heat collection/store is of interest.
That 350W base load is horribly high: the council's not running a streetlight off your power, is it? Or maybe there's a secret MI5 unit under your house... In fact, I'd say that that is high enough to be dangerous simply in terms of heat that may be being released in a confined place. Have you tried turning off the house circuits one at a time until the load goes away? Chris says:
I suspect (hope) this is a combination of wired-in devices I can't measure with the portable meters: oven, boiler, burglar-alarm and an inaccurate meter. I've measured usage at each socket in each room with the portable meters and I've tried to measure the effect of individual circuit breakers on the fixed meter, but with the value bouncing so much it's not clear: I do need to shutdown the house and flip the Big Red Switch.Do all the insulation first: we're not going to get a feed-in tariff in the UK for the foreseeable future and thus exporting power on a domestic scale is NOT going to be profitable in its own right, IHMO, but avoiding importing power at retail rates is starting to look sensible. Chris says:
A[nother] thread at http://www.eurotrib.com/ had a German contributor giving details of their feed-in tariff: over 40c/KWh for some users (varying with later installation dates, size of installation and over time). PV really does make sense with that sort of incentive -- which is why Germany has added so much PV: farmer's barn roofs are supposed to be especially popular. Until he provided actual numbers I was wondering what planet he was from: he was talking about his parents generating all their own power[*] and making 2,000 euro/month from the sale of excess.
[*] net usage - using the grid as storage.So, insulation, draught exclusion (I saw mention somewhere of heat-recovery-ventilators which will become important if I as the place becomes air-tight. (Ha, who am I kidding? We've got 8 fireplaces with open chimneys and another chimney that would have been for a boiler in the old (single-walled, concrete on earth floored) wash room which is now a downstairs bathroom + utility room).
Get electrical usage down. An aside: I've been looking at PIR-controlled light switches, but all the ones I've seen so far need a minimum 40W load (and of course they have their own parasitic load). If/when I can get usage down enough that PV could cover fridge/freezer/lights, look at PV again.
Chris responded on the GSHP front:
I've been thinking about that, but not got as far as doing anything! The street slopes gently (1 brick every 2 houses) so while the foundations get wet in torrential rain, they dry again which suggests the water does/can move.
A friend living nearby (he of the underfloor pipe lagging) has gone as far as getting surveys and quotations for GSHP. His garden is the site of an old 'dew-pond' -- which we take to mean a surface-level well, so he had no concern about getting heat to the pipes.
But he's been put off by the 3kW to 4kW of pumps he's been advised he'd need. He's done the calculation on CO2 cost of the 3-4kW to run the pumps and concluded that until/unless he can use RE for the pumps he may as well burn the gas (and produce the CO2) in-house.
So he's going round the insulation loop again. (And writing about it for the local paper: http://www.chroniclelive.co.uk/north-east-news/the-environment/go-green-news/2008/01/22/dodging-the-draughts-72703-20379926/ )
Given the history and construction of this house, I'm pretty sure that there's a good bit of reduction I could/should do first: For example, the job today is to put a door back on the sitting room. For the last ~20 years the sitting room has been open to the hallway...
(Hmm, yes, our sitting room doesn't have a door into the hall either...)
If you have a mains natural gas supply then burning that for heating is about the same in terms in CO2 terms overall as using GSHP with mains electricity AFAIK. Mains gas is also much cheaper than GSHP in capital costs.
Some things that may change the balance:
But burning gas is otherwise a least-bad thing to do, especially once you've insulated to the max.
All AFAIK/IMHO of course.
Chris said:
I didn't get the sitting room door hung today, but did a few other useful things including measuring the amount of front wall usable for solar panels: there's 6.4m^2 of space available between+beside first floor windows which gets midday sun on the shortest days (perhaps 10-14). Today it was in sun ~8am to ~15:30. So that's something else to add to the mix.
I do like the idea of a thermal store from several directions:
- it can be cold here in winter, so storing energy as heat rather than converting it from other forms makes sense.
- collecting should will be more efficient than collecting electrons.
- storing heat in water/rock/... should be cheaper than storing electrons.
I'm going to do some calculations along the lines of your milk tanks to see what's possible with ummm (assuming there's 1m available in the foundations over the full area of the house) 60m^3 of water and between 6 and 12m^2 of solar panel.
There's a blog I read (http://www.eurotrib.com/) which has several energy-savvy people (the founder, 'Jerome a Paris', is a banker for the RE sector, specialising in wind farms). There's been some interesting postings and discussions about gas supply policy in Europe, the most recent http://www.eurotrib.com/?op=displaystory;sid=2008/2/8/23619/65712 which again highlights the problem of depending on The Market to provide energy cheaply when there's only one major supplier (Russia) who doesn't benefit from competition so won't play. Anyway, that's politics ... My point being that I'm expecting gas prices to increase massively in the coming years _and_ would not be surprised at supply shortages on top: Most of the rest of Europe have long-term contracts with Russia at fixed prices. What's left when the long-term contacts have been filled goes on the market where the UK gets what it can.
So heat is good! Without large capital outlay is even better.
If gas supplies run short it wouldn't even be safe to curtail domestic demand (low pressure in the pipes, air getting in, etc, etc) so industrial demand, including 40% of our electricity supply, would have to take the hit first. The UK is building storage, etc, like crazy to blunt Putin (+successor) attempts to turn off the taps, but yes I'd like to reduce dependency on gas all round.
Chris reported that he'd done some work to chase down his 350W 'base' load:
OK: I've done some testing. (Mmmmm this has turned into a saga!)
Background
- 5 circuits on the DB
- real-time monitor on the DB<->meter cable never showing less than 350W even after turning off computers and subtracting known devices
Oven
Oven showing clock, but not otherwise in use shows 0 [good!]
Downstairs Lights
All accounted for: the main hall light (on most of the day) is rated 32W and measured 35W.
Downstairs Power
The new-ish combi-boiler uses 50-123W, usually 70-87W.
I need to check the duty cycle for this.The fridge and freezer each use about 80W when running and the fridge has a 25W bulb.
Record the duty cycle of these (and write it down this time).The joker here is the music-centre in the kitchen (on/standby 24x7): 50W on standby, not noticably more when in use as a radio. I've put it on a switched power block until it can be replaced.
Replace, possibly with switched active speakers driven by SqueezeBox.Upstairs Power
As expected, the main use is 'The Back Room' (the office). The main server is 130W up to 160W in heavy use. But firewall, switch, WAP, printer, 2 LCD screens, C's workstation bring the total to 280-300W on 24x7. I've rearranged the power distribution so that C and I each have a switched power block for screen and misc. power supplies (radio, phones, laptop, ...) that can be switched off when not in use. The remaining devices are classed as:
- always connected+filtered (server, switch, WAP, firewall)
- always connected (not filtered) (lamp, DECT phone PSU, Ethernet-over-Mains)
- not reconnected: DAT drive, old SCSI CD-R drive, old Mac server, ...)
There was also 13W previously measured in the front-attic from DECT phone, music centre, SqueezeBox + Ethernet-over-Mains.
Put all but the phone onto a switched power block.Upstairs Lights
I found that the main front/back/half-landing lights (on when anyone is active in the house) were taking 158W: but these are just 32+20+15=67W CFLs. I narrowed it down to the biggest (and oldest) lamp and replaced it ... to find that the 20W replacement used 70W! Basically whichever combinations of lamps I used, the draw would be more than double!
Look for lose connections at lampholders, switches and junctions, lamps wired under the floorboards, check the cable run length.
So between the extra computer equipment, the boiler, the kitchen stereo and the upstairs landing lights, I've accounted for the mystery consumption. I've made some changes and got an action list for future work. I really don't know what to make of the upstairs lights...
Indeed a saga, and a bit startling to discover the rogue CFL energy monster! Another reason to get one of the at-the-meter power monitors like Chris has.
After a bit more testing Chris discovered that his rogue '70W' CFL wasn't quite as bad as his power meter was claiming.
... Well [the power meter] measures VA rather than W.
Testing various light bulbs with a power meter:
Rating Light W VA PF[*1] 100W incandescent 102 102 1.00 15W OSRAM Dulux 14 25 0.58 18W Phylex ecolamp 22 36 0.59 32W OSRAM Circolux 37 62 0.60
*1 Power Factor: supposedly W/VA but not very accurate when non-zero!So that's where my 158 not-W-but-VA from 67W of CFL comes from.
To test this I tested actual live usage. Having turned off all other circuits, I compared the power usage reported by the real-time power monitor (Efergy Wireless Energy Saving Monitor) and the spinning disk of the meter with different lights:
Lamp Efergy KW[*3] Meter rotations/minute[*2][*4] 1x75W R80 incandescent 0.035 0.34 6x15W R80 CFL 0.210 0.32
*2 300 rotations/KWh == 3+1/3Wh per rotation
*3 this doesn't make sense to me -- it should be ~0.075 for the incandescent and (assuming a 0.6 power factor for CFLs) ~0.150 for the 90W of CFLs. How does it manage to under-report incandescents by over 50% and over-report CFLs by 50% even allowing for it reporting VA instead of W!
*4 this doesn't make much sense either, it's supposed to be 75W -vs- 90W so the incandescent should use a little less than the CFLs. This might be measurement error, I was trying to record start and stop measurements on a spinning disk.
So what I take from this is that once you're down to a few hundred W, which is mostly CFLs, computer fans/disks and fridge+freezer motors, this particular power monitor isn't very useful :-(
At least I don't have to pull up floorboards again looking for more bad wiring!
That's interesting to note. So that rogue CFL that you described before (I assume the last on the list) was not so bad after all.
(Many days later I woke up in the night with the following thought: if that meter is clip-on and works as I imagine, all it can measure is current. It can guess that the voltage is the standard 240V RMS or whatever, but has no idea of phase, so cannot correct for non-resistive power factor.)
Correct! (As long as I'm being charged for what it says on the meter, rather than the real-time monitor.) I'm looking for the Efergy monitor documentation because I don't remember any mention that it measured VA rather than W (the screen only says KW/WKh) which is rather naughty.
Another possibility is to put the CFL in a standard/table lamp and see what one of the plug-in meters claims. A W-accuracy-bakeoff!
The W/VA/PF numbers came from a Maplin (cheap-n-cheerful) meter: I had to build a light-socket:plug (because the only lamps we've got left have things like SES connections!)
I was rather surprised at the W/VA disparity because I'd not noticed any difference with other things I'd been measuring (and I had wondered what the point of the VA button was): admittedly kettles, toasters & etc. would've been resistive loads.
As long as I'm being charged for (approximately) the rated value it doesn't really matter, but it would've been nice if the power monitor tallied with the meter -- then I wouldn't've spent a week worried about arcing joints or warm cables.
Anyway, he's greener by the minute, dodgy VA vs W or not...
The 50W-standby music centre has gone, replaced by a pair of active speakers (still 6W when idle, but I've put then into a switched socket so they're easy to turn off when not in use). So music/radio is now 23W when in use, 4W with just the SqueezeBox on for time & RSS feeds. (But that excludes the server running slimserver, the WAP, switch and firewall needed for radio!).
So the audit has saved about 1KWh/day from the music centre and about another half KWh/day through more devices on switched sockets. I'll be able to tell by next Sunday if there's a noticable difference.
...
Still a lot to do on the electrical front, but I'm going to look at heat for a while now: Get some clear polycarbonate sheeting to play with thermo-syphons on the windows and check termperatures around the house. So far my new thermometer has confirmed known draught locations, but also shows the other side of a wall supporting a radiator as 3C above the rest of the wall. So: I need to do something about all CH radiators on external walls. And the ground-floor floorboards are consistently 2-3C cooler than the internal walls, so that's another area to investigate.
We've put cardboard behind our external-wall radiators to reduce heat loss through the wall. Theoretically they should be mirrored with foil too, but I'm happy to reduce conduction first and postpone [dealing with] radiation losses to another day.
We've also not yet really considered putting shelves above radiators to push heat back into the room, though in a couple of cases I think it could really help.
Phil Turner wrote to me soon after I put up this page:
I've just read through "Going Green in Newcastle" and I thought it might be worth sharing my experience in trying to reduce the power usage/consumption of computers that I have in my home. I have a couple of servers running, one of which does the same job as Chris' "house server". I assume like my machine, Chris' is just an old unused PC. As I have my server up in the loft, I control it remotely via "ssh" and so it has no need of a monitor (an obvious power save there). But if you delve a little deeper you can reduce power usage by quite a bit by simply pulling out all the unused devices in the PC. Unless you have a need for things like cdroms, sound cards, modems, extra USB ports etc then get rid of them!, you will be surprised how much power some of these devices consume (on some machines you can even remove the graphics card).
One other thing that I have done for my "house server" is to shut it down overnight automatically by setting up a cron task (I'm using Linux as my server O/S) to power it of in the small hours, and then using a plugin mains power timer to kill the power about 10 minutes after. I do it like this because the server is so old that it cannot be powered of by the O/S, and also this means that when the timer powers back on, the server automatically restarts.
With a few other tweaks as well I have got the power down to around 35W whilst idle, and a maximum of around 60W whilst serving files. Averaged out over a week, the power consumed by this machine came to just over 32W per hour!
That's pretty good and not much above the ~27W from the mains for my low-power laptop. There's a lot of scope for trimming consumption of these boxes!
Chris has brought to mind a number of suggestions that I should implement in 2008:
Need help? Don't know your solar from your loft insulation? Try ExNet's energy advice: free to get you started, then there for the tricky questions later too!
The machine that serves this site is partially powered by off-grid solar and wind renewable energy, and that as of 2008/03 the entire remaining consumption is more than offset by a grid-tie solar PV system.
Please email corrections, comments and suggestions.
Please read our privacy policy.
Copyright © Damon Hart-Davis 2007-2008. [home]
