BT Smart Comes Of Age


For those wanting to develop peripherals with BT Smart (previously known as Bluetooth Low Energy or BLE), the options so far have been limited. Either, work directly with chip manufacturers like TI or use a module such as the excellent BLE112 (based on the TI chipset) from Bluegiga. Chipsets give the benefits of low component cost, but require high volumes to be used to make the development and approval costs worthwhile. Modules however, are normally pre-approved, but come with a higher price tag. Typical component costs for a solution in volume would be in the order of $4 or $15 respectively.

BT Smart of course, is the new ‘version’ of Bluetooth that actually has nothing in common with traditional Bluetooth, but allows for devices to run at very low power, able to run off of coin cells for years. They also require less overhead in an operating system, typically within iOS for example, being able to connect directly from an app, rather than having to go ‘techie’ to connect a remote device. More than one device can be connected at a time, too.

Bluegiga's BLE development Board

Bluegiga’s BLE development Board

Bluegiga’s great feature is the fact that an application can reside on the BT Smart device, allowing it to run autonomously, too. Basically access is given to the processor on the device, and Bluegiga have developed a neat scripting language to enable these apps to be implemented on their devices.

Just a couple of weeks ago, Bluegiga sold their business to SiLabs, we have yet to see what will happen as SiLabs absorb the Bluegiga business and IP, how much of Bluegiga’s innovation will continue, too.

But, almost at the same time, a new pretender to the throne has arrived. Anaren, with a Broadcom based module. Pre-certified too, this comes in at a much lower price tag. Developers working with the BLE112 will find it difficult to justify staying with that solution with Anaren’s similarly featured A20737 device coming in at around a THIRD of the price of the Bluegiga module.

Anaren's BT Smart development board

Anaren’s BT Smart development board

Demonstration apps are simple to develop, with their online ‘Atmosphere’ development environment meaning that it is platform independent, no development tools have to be installed. Bluegiga on the other hand only support the PC for their script compiler, a slight disadvantage now given that most silicon vendors support Mac and often Linux too, with their tools. It’s fast too, I got my first app running and communicating from iOS to BT Smart device in about thirty minutes flat. That’s impressive.

Anaren provide a visual development environment, allowing a developer to quickly prototype an application and demonstrate it, with connectivity to iOS and Android supported. What is unclear, and remains to be seen, is how simple this is to move to production. The programs that go onto the BT SMART module and the phone are not clearly delineated. GATT profiles are not clearly published from Atmosphere, with some rather vague explanation in their support Wiki pages (I’m sure this will be improved with time). Certainly, all of the communication between host and target goes through in one large GATT characteristic (called “Atmosphere Transport Protocol”, not really the best implementation or optimised for low-power applications.

So, Atmosphere does seem like a bit of a blind alley. Great for concept proving, but not really much use to move onto production. But, help is at hand. The module, being based on Broadcom’s BCM20737 device is supported by their ‘Wiced’ (pronounced ‘wicked’) development tools. An Eclipse IDE and the compilers are all free to download for PC or Mac, and allow development in ‘C’. Example frameworks are supplied, I haven’t developed anything just yet, but they look fairly straightforward to use, and you can see the commonality carried over into Atmosphere, so if you use that first, it should be easier to move onto the Broadcom tools. Time will tell how simple it is, but it’s certainly worth investing a little time, given the difference in price with the modules. It’s really not worth moving to a BT Smart chipset unless you are using really high volumes, probably in excess of 200K units or more, I would think.

It does look like Anaren have hit the mark with this device, and likely that Bluegiga have sold their business at just the right time!

I’ll post more as I learn more, I have an active BT Smart project right now that is running on Bluegiga already and is being ported to Anaren before production.

Prices of the development kits? Bluegiga’s is around $200, Anaren weigh in at about $50, software tools for both are free.


The Pebble Watch

Pebble watches in different  colours

Pebble watches in different colours

I’d been looking at the Pebble Watch for a while after buying one as a gift for a friend. It had been available in the retail channel in the UK for a couple of months. After receiving mine for Christmas, I set about writing an App for it to function as a Rugby timer and scorekeeper.

I’m pretty impressed with the watch overall. If you have a program like this that is useful to you, it’s going to be worth having for that alone. Other watches might be prettier and offer more functions, but (I’m pretty sure) that most are limited to working with one watch OS alone. Android watches need an Android phone, and so far, I think the Pebble is the only watch to work with iOS. It’s not sold in the Apple store of course, it’s a competitor to the forthcoming Apple Watch, but they will continue to hit different markets as the Apple Watch is likely to be at least twice the price of the Pebble.

The watch is unpretentious. It doesn’t claim to be the most powerful or best smart watch out there. It has a mono screen, not colour like most of the others, so it’s not flashy at all. What that gives the watch though, is an advantage in terms of battery life. They claim 5-7 days in normal use. You charge it using a magnetic cable that takes power from a USB port, so you can charge from your computer or a wall adapter, and it only takes a couple of hours for a full charge.

It’s an “e-ink” display, this is the sort that doesn’t really take any power if it doesn’t change. So, if your watch face changes every second as opposed to every minute, then it will drain the battery a bit faster.

Watch face with weather display

Watch face with weather display

My favourite watch face is a simple watch with the day and date as well as an outside temperature reading. This of course, comes from a data service that comes via the internet. Like all smart watches, access to the internet is via your phone – connected via Bluetooth – so, if your phone isn’t available these sources of data (there’s also some selectable watch faces with news and so on,) are not available either.

This watch face also shows your current location, there are others that even show a little map on the display.

Hardware wise, the watch has an accelerometer and compass, so can function as a basic fitness tracker, there are apps for that too. An impressive compass display is another. It has standard Bluetooth and Bluetooth SMART, so this probably helps the battery life somewhat. I’m not sure what data is transmitted via Bluetooth SMART, but it would make sense for things like the temperature data to go via Bluetooth SMART. It’s fully water-resistant too – that’ll be handy for those rainy Saturday afternoons on the touchline!

Built in is a music controller (AVRCP) so you can play/pause forward/reverse music on your phone.

Rugby timer & scorekeeper app

Rugby timer & scorekeeper app

Watch faces and apps are downloadable from the ‘Pebble Store’. At the moment, all of the apps are free, so you can download what you like, but a few developers use what are called ‘companion apps’ which the watch app will require to operate. They then charge for these apps via Google Play or Apple’s App Store, so they get paid that way. That said, most are free. Up to eight apps and faces can be stored in the watch, a ‘locker’ in your phone is used as an overflow for others, so they are quickly available to you. Watch faces range from digital displays with big numbers, words and analogue displays.

I got the basic watch, there are two models, the Pebble Watch (£99) and the Steel (£179). The Pebble Watch is a clunky looking, plastic affair available in different colours – most will opt for black like me, I would imagine. The Steel is a more robust looking and probably more conventional looking watch. The standard watch comes with a rubber strap, which I quickly replaced with a leather watch strap from my local jeweller. I got a decent quality Hirsch strap, but any 22mm strap or bracelet will work. The steel has a non-standard fitting, so you’ll need to buy a strap or bracelet designed for that watch.

For developers, the development tools are very straightforward. You don’t even need to install anything on your computer, you can develop and compile using ‘C’ via CloudPebble – an online development portal. Publishing a finished app is simple too, your app goes straight into the store. By comparison, iOS development is much more complex and the approval process to get into the store is much more long-winded. But then, Pebble is a simpler concept from the start. You can install a compiler onto your computer if you want to be able to develop off-line, OS X, Linux and Windows are all supported.

Pros and cons? Well, price, battery life, simplicity and iOS compatibility are things this watch have going for it. Lack of a colour display and possibly the clunky looks might be things that will put you off.

When the Apple Watch arrives, this might change things, but we will see. Apple have done some really clever things with the watch bracelets and straps, making them easily interchangeable without tools. In all the years the jewellery industry have been making watches, they haven’t done anything as clever as Apple have with their straps! Apple Watch might be a game changer, but I’ll still be using my Pebble as a rugby timer on Saturdays at least, I’m sure.

Pebble Development – a first App


So, for Christmas I received a Pebble Watch. I looked at the SDK for this and there’s a quite comprehensive development system with event driven functions as well as all the normal ‘C’ type functions.


Pebble display when the App starts.

Shooting rugby most weekends, and writing reports, I often struggle to keep track of the score, elapsed time and cards, so an obvious app for me is one to do that for me. So, I set to developing just such an App for the Pebble watch.

When the App starts, it’s ready to go for kick-off. The clock is set at 40 minutes. The three buttons Up, Select and Down have different functions during the program use, indicated by the selected part of the display. Press the Select button to select the three different areas of the display (the very top always shows the current time of day).

The first section is the match time, with the display showing the elapsed time and the remaining time in the match half. Press the Up button for time off and on, indicated by X and o in the right-hand part of the display. The watch vibrates when the clock runs out. A long press on the Up button restarts the timer from 40 minutes – use this when the second half starts.


Match underway, with one card in progress.

Press Select and you move to the Cards section. Press the Up button to add a card timer. Up to six cards can be added, with the watch vibrating each time a card expires. Hopefully six cards will be enough!

Press Select again, and you get to the scoring section. Press Up to increment the Home score, and Down to increment the Away score. Press and hold the relevant Up or Down button to decrement the scores in case a score was added by mistake.

Finally, press Select once more to return to the timer control mode.

My first attempt at a Pebble watch App. I guess it could look a but prettier, but it’s functional! You can see it here.

Update :

I’ve updated the program a little with some ‘hints’ on screen, and various other improvements.

December 2014, Simon Taylor

Photography Essentials


photo_s4pro2_500I’m surprised how many so called “photographers” don’t calibrate their screens. Calibration ensures that your screen displays colours to a conformed set of ‘standards’. A hardware device (such as the Datacolor Spyder) measures colours from your screen and creates a profile which the display and operating system can use to correct colours and brightness. The screen is calibrated at regular intervals to keep the colours true.

One of the big benefits is that if you use a calibrated print service (like ProAm or have your own printer calibrated), the prints you get have accurate colours. Hold the print next to the screen and the colours will be true. Get another print month later, and they’ll be the same. Try it if you’re currently uncalibrated.

Apart from all that, the calibrator seems to give monitors a new lease of life. Calibrate your screens and projectors and all of a sudden, they seem to come to life, obviously because colours are truer.

While we are on a small rant, lots of people don’t understand the meaning of “backup”. They think that putting files onto an external drive is a backup, when it’s still the only copy of those files. Simply ask yourself, “how would I feel if I lost my hard drive/computer etc.” If you would worry, you need a backup, which is a second copy. Preferably a third copy. One of my cats knocked one of my drives onto the floor recently, the drive was rendered useless. I had a backup, so thankfully, nothing was lost apart from a few pounds and a bit of annoyance at the cat…

Farnborough Airshow 2014


Red Arrows MINIs from Barons

What an amazing show! Over $201BN of business for civil airliners and engines was made during the show. Key manufacturers displayed their products, both on the ground and in the air.

Organisations like the air cadets, disabled flying organisations, historic and preservation societies, even the Bloodhound SSC team were there amongst many others.

For Farnborough, the airport is the lifeblood of the town, and the airshow brings a huge amount of additional business, not just in the week of the show, but local hotels accommodating contractors and business people for six months before, too. Without the airport, Farnborough would be just another vanilla town, like many others in the UK.

Local restaurants and other businesses have a boom time during the show.

Over 100,000 trade visitors attended during the first five days, with the same number expected over the weekend, being the public days. The public visit, with children allowed in free, local people given a “buy one, get one free” ticket offer, and many exhibitors encouraging the engineers of tomorrow with all kinds of attractions. Meet the Red Arrows, sit in an F-35, build rocket powered cars…

A nice touch from Breitling, apart from their wing walker displays and the beautiful Super Constellation was their lounge, just flash your watch and get entry to a little bit of peace and quiet in the show with complimentary drinks and air conditioning, too with a nice viewing platform.

The Red Arrows were stunning as usual, the A380 never fails to impress and the Boeing Dreamliner is probably the most stunning airliner in the skies today.

How the F-18 can do the acrobatics it does is incredible and the Eurofighter Typhoon is great, but not really a match visually for the American plane. No doubt the F-35 when it does finally arrive will trump them both!

The media are well catered for, with the usual WiFi, press information and so on, and a well elevated viewing platform – but as usual a bit manic up there!

Local car dealer Barons also hosted an event on the Thursday, half-way through the show and showed off their nine Red Arrows liveried cars.



Many people watch from outside the event, seeing much of the displays, but missing out on the action inside the airfield perimeter.

Something for everyone, and the weather was good too.

A full gallery from the event is available to view at my main web site.

Electric Cars Are NOT A Green Alternative…


First published in the September 2012 issue of Desig, Mensa’s Design Special Interest group.

Following on from my previous post on the matter, the Nissan Leaf was been robustly advertised over Christmas 2011 on UK television, proudly announcing its status as “2011 Car Of The Year”. Now, the Vauxhall Ampera is “2012 Car Of The Year”.

While alternative fuel cars (hydrogen?) are probably the future, I have to question these.

First, the Nissan Leaf costs £23,990. That’s about 30% more than a reasonably specified MINI Cooper D, with similar on road performance. The Vauxhall costs £32,250. These figures are including a government £5,000 subsidy for purchasing an electric car. I’ll concentrate on the cheaper Nissan Leaf in the examples below.
It does benefit from a zero road tax rating, but then, so does the MINI. But, there are some big downsides. The quoted range is about 100 miles and reviewers have had ranges of about 47 miles in city driving. So, it’s not one for the motorways, and not really one for the cities either. I live about thirty miles from the centre of London, so it would be a sixty mile round trip – and I don’t think my nerves could stand the adventure of the possibility of getting stranded on the M3.

Then, there’s the issue of a British winter. When it gets cold, ‘normal’ cars use excess engine heat to warm the cabin. This one will have to use battery power, reducing the range even further.

They say you could save about £1,500 a year on fuel. Let’s do some maths.

To save this sort of money, you’d need to do almost 14,000 miles a year. That’s 38 miles every day, quite a lot for a car with a short range. That’s an average, many people will have driving patterns that include days with no driving and others with hundreds of miles. I’ve calculated this against a car doing 40mpg, petrol at £1.40 a litre and a charge costing £2.40 (the battery is 24KW/h).

Now, the car is warranted for 3 years or 60,000 miles, and the drive train and battery for 5 years / 60,000 miles. That means if you DO use the car for 14,000 miles, the battery will be out of warranty just after four years.
What will the battery cost? I’ve seen rumours of £7,000. Being generous, if it’s £4,000, then your average mileage of 10,000 miles will save £1,080 a year, so £5,400 after five years. Murphy’s law dictates that the battery will need replacing once the warranty has expired, so that should leave you £1,400 better off than a 40mpg car.

The MINI does average 50mpg, that will save you that extra money in the first place, and the car is cheaper to start.

So unless you are a raving ecologist and willing to pay significantly more to feel good about doing your bit, what’s the point?

You might say that electric power is greener, but again, I’m not so sure.

One benefit of a vehicle that uses an engine is that the fuel is stored in the vehicle and goes through a single process to produce mechanical energy to the wheels. We could probably assume that this is 70% efficient.

But, electricity in the UK is mostly produced by fossil fuels too. About 74% of our electric power is from oil, gas and coal, 20% from nuclear, and the remaining amount from ‘green’ sources such as wind.

So, for the purposes of calculations, let’s consider that nuclear is ‘green’.

Our electric car will need power from the grid. We could be generous and say that the battery to motor power conversion will be 80% efficient. But we need to charge that battery from the mains, that is generally considered about 70% efficient. Getting that power to the house requires transmission, which involves conversion to and from high voltage and transmission itself. 70% efficient would be good. Then, the power needs to be generated in the first place. Lets say 80% efficient.

So, for each unit of fossil fuel in our petrol car, 0.7 would get to the wheels using our 70% efficiency guess above.

But, for the electric car, we need to apply 80%, 70%, 70% and 80% as above. Allowing for the 26% of fuel which we consider ‘green’, we can apply a 1.35 multiplier which improves things a bit, but we get .42 of each unit getting to the wheels of our electric car. We’re burning more fossil fuels as a result, so how is this greener?

Unless we charge our electric cars from home using entirely wind or solar power I can’t see how electric cars make sense. They just move pollution out of the cities and cause us to burn more fossil fuels than we already do with our current petrol and diesel cars. The idea of calling electric cars ‘zero emission’ is just propaganda in my opinion, until we have renewable energy for the majority of our electricity needs.

The Nissan Leaf is only greener because if you buy one, you’ll probably end up using it a lot less than you would a normal car…

A New Wireless Camera Controller


Update 18th July – the second prototype

For a couple of weeks now, I’ve been working on a new camera controller.

Like many photographers, I often use a cable release to activate the camera. This has a number of advantages in allowing the camera to remain steady on a tripod, or simply to allow me to look over the camera while shooting. More recently, I’ve been using a radio remote controller which has the same simple shutter release control, but is wireless.

I’ve been involved in Bluetooth technology for some years now, and a new development is Bluetooth SMART – also known as Bluetooth Low Energy and included in the Bluetooth 4.0 standard that is appearing in mobile phones, tablets and computers now.

Bluetooth SMART is a very low power protocol, designed for low-data throughput devices such as temperature sensors, switches etc., and it lends itself quite well to the camera controller function I mentioned. Devices will run from a
coin cell for years – unusual for a wireless device.

So, I set about developing such a device, but rather than just have simple shutter control, I have added timer functionality.

You can trigger the device and take a single shot, just like a normal radio remote, but you can also set parameters that allow you to start a timer running which will wait for a period of time, then take a number of shots at a programmed interval. Once the parameters are set and the timer started it will run autonomously. Range is about 100 metres, so it’s a very useful remote trigger too.

All of this can be triggered and controlled from any Bluetooth 4.0 capable device, such as the iPhone 4S, iPad 3, most new Mac computers and newer PCs. Most new devices include Bluetooth 4.0, so expect to see many more in the coming months.

The prototype is all working, with a program in the device to give this functionality. The next steps are to produce a PCB (printed circuit board) to allow these to be made in higher volume – and neater – rather than like the hand-built prototype above. I also need to develop iOS and Android apps that will allow the device to be controlled from smartphones. This can be done at the moment, but via a rather clumsy generic Bluetooth application on these phones.

Bluetooth SMART Controller Prototype

I reckon the device could be made available in the retail market at £50, with the iOS app being a free download. Different versions to connect to Canon, Nikon, Sony cameras would be available (a different connector is needed for each manufacturer) and the controller is your phone – which you have with you, rather than a second box for the trigger.

So how to do the next steps? Design the PCB myself, design the iOS app? both of these are new to me, but I know people who could do this. Maybe use Kickstarter to finance the project? Not sure.

Watch a demonstration of the controller here:

But, I am looking to move this forward – if you have any bright ideas, let me know!

23rd July – added datasheet – Phooto iSMART Datasheet

27th July – new video with the second prototype controller –

2nd August – If you would like to make one of these, let me know, I can provide the circuit diagram and object code to load into the Bluegiga BLE112 used here. You will need to source the module and components. I would be interested to get feedback on the device – you might even be interested in developing iOS software for it?

Blexplr iOS program used to control the timer

6th August – Dr Michael Kroll’s ‘Blexplr‘ program has been invaluable in developing this application – and has now been updated with the UUIDs to support the device. (see picture).

19th August – First Nikon D90 version made.