Wind power in Malaysia – the missing map

March 28th, 2011
TNB's wind power in Malaysia map

TNB's wind power in Malaysia map

Just a quick note to myself really – many, many times online I’d replied to people saying “why doesn’t Malaysia use wind power?” with my belief that it doesn’t have any. I couldn’t find a good source for that  before today, relying on maps of global wind patterns and the argument that “this isn’t called the Doldrums for nothing”. I found the link on a nuclear -versus – solar page at TNB which bizarrely concludes that renewable energy is ‘unpalatable’. As if somebody has to eat it.

The whole “Think nuclear, think green” section at TNB appears to me to have been written by a 12-year-old set a ‘justify nuclear at all costs’ exercise. Why include graphs with illegible legends? In what way is ‘refuel every 18-24 months’ a guarantee of energy security forever? The material is so devoid of facts that it’s impossible to conclude that it’s anything other than marketing gloss.

http://www.tnb.com.my/think-nuclear-think-green/nuclear-vs.-renewable-energy.html

Java console “Content-Encoding: null” message on loading applet

March 26th, 2011

Sun JavaI thought at first glance this might be an error in an website I’m working on, so looked into it a little bit further – turns out it’s nothing at all.

I turned on request debugging in Spinneret and could see that the Sun Java plugin in FireFox was sending the request header:

accept-encoding: pack200-gzip, gzip

… so perhaps it felt obliged to point out that there was no Content-Encoding specified in the response. I don’t set a Content-Encoding header for the response, as it’s not encoded. As far as I can see there’s no need to encode the response I’m sending – it’s a jar file, so it’s already in Zip format – compressing it again would probably make it bigger. I can’t help thinking though that the message shouldn’t be there at all if unencoded responses (identity encoding) really are OK.

It’s nice to see that the Java developers have taken the Header Field spec at face value and are sending all lower-case field names in requests from the plugin. I wonder if that ever catches out anybody who expects Initial-case or Camel-Case header field names?

Bye bye spider.my

March 24th, 2011
Spider.my has shuffled off its mortal coil

No more spider.my

I’m taking spider.my down in anticipation of a move from Malaysia in the next few months. The split-brain nature of the site, with a front-end application on a small local VPS and the search (and a few other bits and bobs) back-end on a stack of servers on my landing makes it sufficiently awkward to move that by far the easiest thing to do with it is to unplug it. I’m unplugging it now because it’s simply one less item on the ‘do before move’ list.

There are no dependent users of spider.my who will be harmed by the switch-off (according to the log). There are plenty of downloads of the Damerau Levenshtein sources, so I’ll move those onto this blog (probably) in the near future (probably). If there’s anything else that anybody misses and it’s easy to put back up somewhere online, let me know.

It has been interesting and I suspect when life settles down again the code may be reincarnated, but not as spider.my – that’s gone.

Spider Icon

Power from the sun (and gun): Bandar Dataran Solar

March 18th, 2011

A couple of days ago when I wrote about how I thought Malaysia should generate electricity from the sun, I realised that the picture that I carry in my head might not be the same as a reader might form in their own. Worse, I dashed off an image of how I think a ‘residential area solar roof’ should work using the GIMP and a few images I found scattered around the Web, and produced something awful!

Bandar Dataran Solar

Bandar Dataran Solar Sauerbraten-style

Ideas like this pretty much live or die depending on whether people can see for themselves what it will look like, so I tried asking at a popular local forum if there was anybody who liked the idea and might be willing to do an “artist’s impression”. I got a couple of suggestions to use Google Sketchup, which I didn’t know about, and will try. Someone was sure the whole solar roof thing is a non-starter – that’s okay, I’ll add their concerns into the mix soon. A few other replies were ‘No money no talk!”, which seems to be a popular local affliction – the same one that is going to get everyone nuclear power here, because that’s where the biggest profits / commissions / kickbacks / facilitation fees are.

All I had installed last night and ready to go, that I knew could draw 3D with, was Sauerbraten. I’d never tried editing a map in Sauerbraten before, but it’s actually a masterpiece of simplicity. OK, I know nothing about 3D design or art and it’s obvious to me that the finished article still looks like a cheap video game, but it got me started, and quickly too.

The solar roof

The solar roof ... something like that

What impresses me most about using Sauerbraten to produce this model is just how fast and pain-free creating such a model with Sauerbraten is. Sauerbraten is heavily optimised to do a few things only but do them very well, and that carries over into map-editing. On the down side – and perhaps an experienced Sauerbraten modeller would know how to work around this – is that everything must be arranged around cuboids and simple fractions of cuboids. The roof ended up being much more ‘clunky’ than the image of it that I have in my mind because my 6 hours of Sauerbraten modelling experience just didn’t reveal to me the ‘way of the very large, slim, gently-sloping plane’. It’s good enough to show most of what I mean.

So what does the image show? Well, despite not having any (that I could find) mapping of model dimensions to real world dimensions, it does show the scale of construction I’m talking about: pairs (facing in opposite directions, like in the development I live in) of link-housing units under a single roof. The roof is separate from the houses below.

Sloping solar roof

Sloping solar roof - but not this much!

One thing I did in this model which is different from my original suggestion is to extend the roof past the ‘front yard wall’ of the house. I did this in the model just to make it easier to put the supports in for the roof. My original idea was that the solar roof would extend only a few metres past the front wall of the house. That’s important so that there’s enough sun falling on at least some of the plot for laundry and growing plants. Without some fairly hard-core light modelling, it would be difficult to say how far out the roof can come. The further out the roof comes, the darker it will be beneath it, but the more electrical power can be produced from the same residential area.

The solar roof slopes, the house roofs do not. Malaysia is slightly north of the equator, so a fixed installation of solar panels will point slightly south. Because the link-housing installation depicted here is very large, I think the houses would have to run West-East so that the tilt can be from the fronts of one row to the fronts of the houses behind – or otherwise the difference in heights from one edge of the giant roof to the other may be exceptionally* great.

Flat roofs and a dark back alley

Flat roofs and a dark back alley

I imagine the solar roof would be a complete roof. When it’s built, it should be rain-proof, perhaps constructed with corrugated panels over a beam lattice. The solar panels would sit directly on top of this roof. If the roof were not waterproof, there would be a problem with mini-waterfalls dropping metres onto the houses / ground below, possibly causing damage. Another hazard is dropped tools during maintenance – a solid roof (without the solar panels) avoids all that. Having a complete roof might slightly reduce the cost of building the houses beneath (can build in slightly wet weather), and also allows the houses below to be built with flat roofs constructed from cheap materials. If these materials allow heat to escape easily from the house, so much the better.

The back alley of such a development would be a dark place, but if my experience of link-housing back-alleys is anything to go by, it’s a dead space anyway. I think it would make sense to include some night-time lighting on the underside of the solar roof (or possibly installed along the roofline of the houses beneath), so that the back alley could be well-lit at night. If it’s lit from down-lights on the bottom of the solar roof, the light should intrude less into bedrooms at night than lights mounted on poles in the road.

In case you want to look at the map yourself, here’s a zip file with the Sauerbraten map inside it. Just unzip it in your [sauerbraten_directory]/packages/base/ and you should either be able to select it in a menu or load it with “/map bds”. If you load it in DMSP mode (Death Match Single Player), you can even shoot a few monsters – there’s some ammo in the longkang (not in real life). Any resemblance between the monsters and my neighbours is entirely unintentional. Also, the name is a play on the name of my residential area ‘Bandar Dataran Segar’. I’m absolutely not suggesting that everyone in BDS cut their roofs off and buy shares in solar – this idea would almost certainly only work well if a residential area was designed this way from the beginning.

bds.zip (Bandar Dataran Solar)

* Using a large example from my earlier post, the big solar roof is 180mx50m. Kuala Lumpur is about the same latitude as Bandar Dataran Solar – 3°N. The sun averages 0°N, so rough calc puts the roof slope at 3°S. Height difference (Opposite) of a roof at 3° incline, for a span (Hypotenuse) of 50m is 50m * sin(3°) = 2.6m. For the 180m span (the row of buildings runs North-South), the height difference would be 9.4m – that’s a long extra (it’s 2m+ at the small end, remember) gap between the top of the building and the bottom of the solar roof.

Solar power in Malaysia – the residential area solar roof

March 15th, 2011
Malaysian house, blue sky

Malaysian house, blue sky (on a rainy day)

‘Part two’ has some better images that should make things a bit clearer!

I keep suggesting this as a ‘real’ solar power alternative whenever someone says (as they invariably do) that solar power cannot produce enough energy to compete with traditional sources. Alternative power sources are in the news at the moment, as Japan’s earthquake and tsunami woes are compounded by exploding, non-productive nuclear power stations. Malaysia is also planning to have the first nuclear power station in South East Asia, led by Minister Peter Chin, a man who responded to requests for reassurances that wind-borne fallout from Japan would not fall on Malaysia with “don’t speculate”. Perhaps I’m confused. I think it is his job to give the Malaysian people the information they need to reassure them about their safety and that of the people they care about. He seems to believe his job is to reassure speculators that they can go ahead and build a nuclear reactor in Malaysia come what may.

It’s reasonably obvious to most people that  plonking a few solar panels on the roof of a house that was designed to shed rain is not going to solve anybody’s power supply problem. The best that can be hoped for from an expensive one-off installation like this is to shave a few pennies off one’s electricity bill in good weather. Malaysia, although it’s near the equator, doesn’t have quite as much sun as you’d expect – it mostly has punishingly sweaty heat. It’s not bad though. Figures in this article are taken from an insolation map at Wikipedia from which I reckon that Malaysia has an average insolation of 200W/m². The good thing about Malaysian sun is that it’s about the same all year round. The worst thing about sun in the UK – for example – is that it’s best in the summer, when everybody wants to go outside and not use any electricity, and worst in the winter, when everybody wants to stay indoors and turn the heating on.

Pulau Kapas Solar Diesel Power Station

Pulau Kapas Solar Diesel Power Station - note separate 'sun roof'

I’ve been toying with this idea for quite some time. Solar obviously needs large expanses of plan area because the power density of solar radiation is low, compared to that of a traditional power station furnace. Dedicating land to such a project makes its price prohibitive. The great thing about solar photovoltaics is that they don’t need dedicated plan area, as evidenced by their current most likely deployment – plonked on your own house roof. One house’s roof is too small a scale to create a solar power station though, so on a recent visit to a coal-fired power station near Pontian in Malaysia, I noticed that the dry end of the coal yard was covered by a huge roof, far, far above the coal heap below. Why not do this for solar?

On a family holiday in Kuala Terengganu a few weeks ago, we spent a delightful afternoon on Pulau Kapas, where tucked away behind a fence by the jetty is a combination diesel / solar power station. There are several solar arrays on short legs fixed to the ground, but the diesel generator shed itself is shaded by a single, separate ‘sun roof’ that is entirely covered on its top surface by solar panels. This is how I think Malaysia could implement solar power stations – by combining them with residential / commercial developments. A standard (modern) Malaysian house is like a large concrete solar oven. Why not shield it from solar radiation, and at the same time convert that radiation to usable power?

Solar panels built over link housing

Solar panels built over link housing

Here’s an image taken from Google Maps of the housing development where I live which I’ve crudely edited to show how I think solar panels could be added to such a development. (The picture is a few years old, and the built-up area has since more than doubled). The 7 solar roofs in the image, 180×50=9,000m², 60×40=2,400m², 80×40=3,200m², 100×40=4,000m^2, 60×40=2,400m², 75×40=3,000m², 90×40=3,600m² have a total area of 27,600m². Using baseline figures of 1KW/m² for peak insolation and 200W/m² for average insolation, power input from the sun is 27.6MW peak, 5.5MW average. Using a photovoltaic efficiency of 10%, peak electrical output is 2.76MW, average is 550KW.

The price of the work is likely to be dominated by the expense of covering such a large area in solar panels. Searching online for solar panels in early 2011, I see a minimum price of USD1.65/W for a minimum order of 10 panels. I think this could be cheaper for a purchase of thousands of panels, but let’s use that. Working backwards from the peak power output, the cost of solar panels for an installation such as the one shown here would be 2.76M * 1.65USD = 4.6million US Dollars, or about 14million Ringgit.

Houses under separate solar roof

Houses under separate solar roof

If the output of the solar roof was sold to the national grid as though from a power station, and we thank the government in advance for mandating that power from solar should be bought by the grid at commercial residential rates (about 40sen per KW/h), then this installation’s nominal average income from the grid should be RM220 per hour, or about RM37,000 per week, Rm1.9m per year. At that rate the investment would take about 8 years to pay off the cost of the solar panels alone. I see notes online to the effect that the solar panels are around half the cost of the project, so let’s say the project’s build cost is equal to around 15 years’ operation. If the price per Watt for panels is lower than that quoted, this timespan is reduced. The same goes for an increase in efficiency from 10%.

The way I envisage this working on a personal level is that I would buy a house in such a development and live in it. I would buy electricity from the national grid at residential rates. The ‘solar roof’ would be owned and operated by a power generation company, not related to me in any way. There would be a legal right on the deeds of my house for the power company to use so many plan square metres in the void immediately above my house (say 2metres to 5 metres above my house, weight per square metre and thickness restricted to avoid cynical development) in return for that right, the power generation company would pay each homeowner a small monthly ‘rent’ of RM10 or so.

There are additional benefits. The solar roof would function as a shade for the houses below, reducing the need for air conditioning. If designed carefully, the void between the bottom of the solar roof and the roof of the house below may encourage air-flow, further reducing need for forced cooling. If the solar roof were also function as a rain roof with its own guttering and downpipes, the houses beneath would be exposed to less weathering and hence require less maintenance. It would even be possible to supply local hot water either from solar-heated tanks or from the power conditioning / storage equipment’s cooling system, reducing the need for expensive-to-run bathroom water heaters.

The timescales for projects like this to make a sizeable contribution to power supply in already built-up areas is long. It is almost certainly not possibly to retro-fit such a project to an existing development. It would have to be designed in before building commenced. In areas such as East Malaysia where development is still at an early stage, the relative contribution from new development would be much greater. A modest-sized development such as that shown here could be generating a few MW of excess power during office-hours.

Bandar Dataran Solar

An update with (slightly) better pictures

I’d be interested to know what you think. Particularly, I’d be interested to know whether you could see yourself living in such a development. If you have any questions, don’t hesitate to ask, and I’ll do my best to answer or – hopefully – someone who knows will see your question and help out. The worst part of a development like this is that it doesn’t offer the same opportunities for trimming off fat in the form of commissions, leakages and ‘contingencies’ that a mega-project like a dam, nuclear or fossil-fuel power station offers. It’s totally transparent (!), so it won’t attract get-rich-quick speculators. It needs massive public support to make it a reality. Will that come?