Infomania.co.uk » White papers

What is the difference between raisins, currants and sultanas?

admin — Thu, 21 May 2009 08:11:34 +0000

Ever wondered what raisins, currants and sultanas were before they were dried out? The answer is simple:

Raisins = dried grapes
Currants = dried grapes
Sultanas = dried grapes

Of course it isn’t quite as simple as that or raisins, currants and sultanas would all look and taste the same. The differences are primarily due to the different varieties of grapes used to produce them.

Raisins = any dried white grapes
Currants = dried Black Corinth (also known as Zante) grapes
Sultanas = dried white seedless grapes, originally from Turkey. Usually Thompson Seedless variety.

Sultanas may have been bleached to make them lighter in colour than raisins.

Barbecue gas – butane or propane?

admin — Tue, 05 May 2009 23:05:56 +0000

Buying a gas barbecue can be more than a little complicated. You think you have worked out what you want, then you discover there are two types of LPG (Liqueified Petroleum Gas) – Butane and Propane. Not only that, but the gas bottles come with different sized valves. And what on earth is Patio Gas?

Butane was more commonly used for barbecues than propane at one time but the situation appears to have reversed in recent years. In fact many, if not most, barbecues can use either butane or propane depending on the regulator valve fitted to the barbecue. Most barbecues now come fitted with a red propane regulator as standard. But is one gas better than the other?

A gas bottle with blue butane regulator attached

Here is the rundown on the properties of different cooking gases:

For comparison, natural gas - as delivered to millions of homes via the gas mains - has the chemical formula CH4. When burnt, a cubic metre of natural gas will provide 38 Megajoules of energy.

Propane, chemical formula C3H8, produces 96 Megajoules of energy per cubic metre.

Butane, formula C4H10, produces 126 Megajoules of energy per cubic metre – far more than either natural gas or propane.

So butane must be better then? If only life were so simple. Inconveniently butane doesn’t work well at low temperatures. When too cool it stays as a liquid and the gas pressure drops. That means a butane gas cylinder will produce lower pressure once the bottle temperature drops below 10 degrees C. The bottles actually get colder in use, as heat is removed from the bottle when the liqud butane boils into the gas that comes out of the regulator. The higher the rate you use gas, the colder the bottle gets. Use it too fast in cool temperatures and the gas pressure can drop significantly. On one level it doesn’t matter - who wants to barbecue in chilly weather after all? How about someone who has suffered a power cut due to snowstorms?

A given quantity of butane will burn hotter than propane but in fact propane regulators release the gas at a higher rate to compensate. In fact many people will tell you that propane burns hotter. Both gases are heavier than air, which is one reason why barbecues have holes in their bottoms. If they didn’t, any leaking gas, (for example from a burner that has blown out) would build up in the bottom, ready to explode in the face of someone lighting the barbecue.

Gas bottles come in a variety of different sizes and, confusingly with different regulator fittings. The clip-on regulators used for barbecues are blue for butane, with a standard internal valve size of 21mm. Propane regulators are red with 27mm in internal size. That means that it is not possible to connect to a propane bottle using a butane regulator or vice versa.

A clip-on propane regulator

Typical bottles sizes vary from 3.9 kg to 13 kg but note that those are the weights of the amount of gas in the bottles, not the total weight. The total weight of most bottles is around three times the gas weight, so a 13kg bottle may weigh as much as 40 kg! BP have started selling Gas Light bottles which are made of glass-fibre reinforced plastic. They are translucent so it is possible to see how much liquified gas is left in the bottle. Needless to say they weigh a lot less too – around half as much as a steel bottle.

Note that in use the gas bottle has to be upright. The liquid has to be at the bottom of the bottle – attempting to use the bottle on its side could result in liquid gas being forced out. Which is one reason why barbecue gas pipes always seem a bit on the short side…

Oh by the way Patio Gas – it just seems to be a marketing term for propane. It will work in both barbecues and patio heaters at all air temperatures, so is a good general-purpose bottled gas that is useful for all the gas-powered devices that are in common use on patios.

For information about the differences between gas and charcoal barbecues see Garden barbecues – charcoal or gas?.

Memory card speed ratings – what do they mean?

admin — Thu, 05 Mar 2009 22:28:08 +0000

SD and CompactFlash memory cards often carry speed ratings but it isn’t clear what the numbers mean exactly. To make matters worse the two card formats use different methods of designating the speeds. Although many devices only take one card type what do you use if you have a camera or media player that will take either?

SD card speed ratings are quite simple – the cards are rated into three different speed classes – class 2, class 4 and class 6. The numbers indicate the minimum transfer speed of each class in Megabytes per second:

Class 2 – minimum transfer speed 2 MBytes/sec
Class 4 – minimum transfer speed 4 MBytes/sec
Class 6 – minimum transfer speed 6 MBytes/sec

CompactFlash speed ratings are completely different. A few manufacturers have applied their own simplified labels, such as SanDisk’s Ultra 2, Extreme 3 and Extreme 4, but the majority of manufacturers use an ‘x’ rating. You will find CompactFlash cards with speed ratings from 12x to 300x. It seems clear that 300x is faster than 12x (especially when you look at the prices) but what is the actual transfer speed?

It may seem odd but CompactFlash speed ratings are the number of times the card is faster than an old 1x speed CD-ROM. This is because CF cards use the same IDE interface as CD-ROMs and can be thought of as an alternative type of removable storage.

The speed of those original CD drives was 150 kBytes/sec or 0.15 Mbytes/sec, so a 12x card will support a transfer speed of 12 x0.15 = 1.8 Mbytes/sec. As you can see this appears to be comparable with a class 2 SD cards. Unfortunately some card manufacturers use maximum write speeds not minimum speeds, which may be significantly lower. Major brands are more honest with Lexar for example making it clear that their card ratings are minimum sustained write speeds.

Here is a table showing some common card ratings. You can clearly see why high-end digital cameras have continued using CompactFlash cards, as all but the slowest are faster than the fastest SD cards:

12x = 1.8 Mbytes/sec
40x = 6 Mbytes/sec
80x = 12 Mbytes/sec
100x = 15 Mbytes/sec
200x = 30 Mbytes/sec
266x = 40 Mbytes/sec
300x = 45 Mbytes/sec

So there you have it – speed ratings for SD and CF cards explained.

Digital cameras – why sensor size matters

bill — Sun, 15 Feb 2009 20:05:46 +0000

A wide range of sensor sizes are used in cameras today. The size of the sensors is not related to their resolution (measured in megapixels). The smallest, 1/3.6″, has a tiny light-sensitive surface of only 4mm x 3mm. That is just 1/72 of the area of a 35mm frame. Needless to say there are some compromises in performance. Other common sizes are 1/3.2″, 1/3″, 1/2.7″, 1/2″, 1/1.8″, 2/3″, 1″, 4/3″ and 35mm (the same size as a 35mm film frame). The inch sizes are the diagonal size of the sensor, 1/3.6″ being equivalent to about 7mm. In comparison a 35mm sized sensor is about 43mm diagonally.

Most sensors have an aspect ratio of 4:3 – the same shape as a standard TV or the original “academy” feature film ratio. The 35m stills film format is wider, at 3:2. You might think that in this multimedia age somebody might have produced a widescreen 16:9 aspect ratio camera but that is clearly too much to hope for!

The Four Thirds (4/3″) camera format was developed jointly by Kodak and Olympus and is also used by Fuji. The intention was to create smaller, lighter cameras and lenses that were functionally equivalent to 35mm SLRs. Standardising the lens mount meant lenses could be made that would fit all 4/3 cameras. The sensors are about half the height and width of a 35mm film frame.

One consequence of the differing sensor sizes is that lenses offer a reduced angle of view compared to a 35mm film camera. Even with the relatively large sensors fitted to most digital SLRs the effective lens focal length is around 1.6 times that for 35mm film. A 50mm lens designed for use on a 35mm film SLR would act as an 80mm lens on a digital SLR. This is an advantage at the telephoto end of the lens range as the lenses are much smaller. Unfortunately it is pretty unhelpful in making a true wide-angle lens. The smaller the sensor the worse these effects get.

The depth of field available also depends on the sensor size. Small sensors give large depths of field. Great for snapshot cameras with fixed-focus lenses or cheap auto-focus with few steps. Not so good for taking pictures through chicken wire or dirty glass – all those obstructions that would have been invisible or vague blurs with a 35mm SLR are now in sharp focus. Another result is that artistic use of differential focus by controlling the lens aperture (to adjust depth of field) is virtually impossible even if the camera allows the relevant adjustment. The serious photographer is left attempting to emulate the effect in Photoshop.

The smaller the sensor the better the lens performance has to be to focus fine detail onto the sensor with good contrast. In fact there is a finite limit to the resolution of a lens at a given aperture and the nearer the limit it gets the lower the contrast. The measure of lens performance in reproducing fine detail is called the Modulation Transfer Function and it reduces as we get closer to the limit of the lens’s capability. Even the highest quality lenses are incapable of delivering adequate detail to the smaller sensor sizes at f16.

Unfortunately it also works out that the smaller the sensor the smaller the amount of light collected too. That’s the reason that pictures from cameras with small sensors produce noisy (aka grainy) pictures. Comparisons of the image noise in pictures from large and small sensor cameras can sometimes be seen in reviews on the dpreview.com website.

In summary small sensors have advantages in terms of price and convenience but these are paid for in image quality and camera sensitivity.

GSM phone technology

admin — Sat, 15 Jul 2006 20:08:28 +0000

GSM cellular phones use a combination of Time and Frequency Division Multiple Access. TDMA simply means that several stations can share a frequency by taking turns. The difficult bit is making them think they’ve all got the channel to themselves. FDMA simply means that the spectrum is split into channels spaced 200 kHz apart. The TDMA works by transmitting taking place during 0.577mS burst periods. There are eight of these bursts in a TDMA frame which lasts 4.615 mS. One channel consists of a single burst per frame. Two channels are needed for two-way communications of course. The result is that four duplex conversations can take place on the same frequency at the same time.

Channels are either common channels, which are used by mobiles in “idle” mode, or dedicated channels which are used for calls in progress. Remember though that these are “virtual channels” – they are simply slots for data to be exchanged, not RF channels!

Dedicated channels contain sequences of 26 TDMA frames. Each sequence lasts 120mS and contains one TDMA frame used for control purposes, 24 for calls, the last being unused. The channels for uplinking and downlinking a call are separated by 3 bursts, so there is maximum separation between transmit and receive periods.

Common channels are used to set up calls and arrange handover between base stations. There are six different types of common or control channel. The Broadcast Control Channel sends base station ID, frequency plan and frequency hopping schedule. The Frequency Correction Channel together with the Synchronisation Channel define when the burst periods happen and the time slot numbering. Another channel is the Random Access Channel which is used by mobiles to log onto the network. The Paging Channel is used to inform mobiles about incoming calls. Finally the Access Grant Channel is used to give a mobile permission to use the signalling channel to set up a call.

The modulation technique is called Gaussian-Filtered Minimum Shift Keying (GMSK). It’s designed to minimise spuraii while still having good spectral efficiency and not needing too much complexity in the modulator.

The digital speech coding is called Regular Pulse Excited – Linear Predictive Code (RPE-LPC). This works in a similar way to MPEG video. The coder predicts the next sample based on the last few. The decoder can also predict the sample the same way. Most predictions are quite close to the actual sample in which case the system will send just the difference between prediction and sample. This needs much less data than sending the complete sample. Speech is divided into 20 mS samples each of 260 bits, giving a data rate of 13kbit/s.

Error correction is added to the speech data with the bits likely to cause the worst errors having better correction. If the sample is too corrupt to correct it is replaced with an attenuated version of the previous sample. Radio channels suffer from bad burst errors, typically caused by impulse interference, so the samples are interleaved (ie shuffled around in time). Each burst in fact carries data from two different samples. The result of interference on an interleaved signal is a reduction in quality rather than complete drop-out for the duration of the interference.

The system has auto equalisation to minimise the effect of multipath fading. A known 26 bit “training sequence” is transmitted in the middle of every burst. After checking the sequence for errors the effect of multipath on the off-air signal can be calculated and an inverse filter applied.

The final twist to this amazingly complicated system is frequency hopping. The mobiles have to be frequency agile because transmit, receive and adjacent base station monitoring all take place on different frequencies. The GSM system makes use of the ability by applying slow frequency hoppping, where each TDMA frame is transmitted on a different carrier frequency. The algorithm for this is sent on the control channel. There are two reasons for frequency hopping – reducing co-channel interference and multipath effects. Both effects will still impair signals but will only cause momentary problems rather than complete drop-outs.

“Digital transmission systems are like hovercraft. They fly – but only just.” – John Wilkinson