Basis

A light bridge in its simplest form is a light source and a light receptor. When the light source can no longer be “seen” by the receptor, the bridge is “broken”.

Light bridges are most common in shops where, when you enter, some kind of “ding dong” chime is activated.

Light bridge

What you can use it for

1. Sensing movement in your house – Is there movement? How much? Can you correlate the data between multiple Light Bridges?
2. Simple alarm system – Is there movement while you are away? Should there be? Who coult it be?
3. Checking if the cat is still alive – This is what triggered this project in my case. Is the cat still alive? Is she still moving through the house on day 3, 4, 5 and 6 after we left?

Types of Light Bridges and which will be covered in this article

1. Visible light- Covered in this article. The simplest form is using a bright visible light source, in our case bright LEDs, and an LDR. The LDR is a resistor sensitive for visible light. The more light falls on the surface of the light sensitive resistor, the less resistance it has. The problem with normal LEDS is that the light disperses. Meaning two things:
   1. Visible – At night it will light the envorinment
   2. Loss – A lot of light is “lost” as it does not reach the LDR but is shone upon the surroundings
2. Laser light – Not covered in this article (yet). Laser light is a focused bundle of light, with a minimal dispersion. This means that most of the light from the LED will reach the LDR and that it will be less visible at night.
3. Infrared light – Covered in this article. To cut all visible light, you can move to using infrared LEDs and an infrared receiver

Reliability

Related articles

Basic workings of the Light Bridge

Light Bridge block diagram

The light bridge consists of two parts:

1. The light source (a) – Which is either a bright LED, a Laser LED or an Infrared LED
2. The light sensor (b) – Which receives the light and will change state based on:
1. Recieving light – Meaning that everything is OK and nothing is happening
2. Not receiving light – Which means that something is blocking the light source and something is happening

The output of the Light Sensor reflects the “state” of our Light Bridge, which is “low” when all is well and “high” when something blocks the light source.

Passing data to your processor: using a Wired and wireless State Bridge

1. Wired State Bridge - Means that we will run wires from the Light bridge to the device that will take that state and act upon it (either a electronic chime or something else)
2. Wireless State Bridge - Means that we will use a radio transmitter to broadcast the state – in this case – “broken” – over the air
3. State Processor – Your PC, laptop, netbook or Android device, linked to an Arduino, Android IOIO or similar IO board

Wireless State Bridge

As our base-station responding on anything passing the Light Bridge will be on another location, we will use a wireless State bridge, which can be an existing solution based on – for instance – a wirless doorbell, or something you build yourself. Both solutions for a Wireless State Bridge will be explained in a separate article here [to be written].

States – and what it means

In any process-technology, including electronics and informatics, we talk about a “state” when something can be two things or more.

For instance, the base-states of a fridge can be: “empty”, “full”, “clean”, “dirty”, “open”, “closed”, “cooling” and “not cooling” and any shade in between.

As discussed the Light bridge has two states: “no activity” and “activity sensed” which is derived from “light is hitting our sensor, so we assume all is well” and “no light is hitting our sesor, so we assume something is blocking the path and activity is sensed”.

We respond and want to respond on activity, which is the “up” part of the little icon we use for state.

In the schematic below, you will find a solution that connects module MOD1 to GND, using T2 as a “power switch”, allowing current to flow through that module when the Light Bridge is broken.

1: The LDR Light Bridge

The LDR Light bridge is the most simple Light Bridge to start with.

In this example we use a 12 Volt schematics, but with some minor adjustmens (different values for the resistors) you can make it into a 5 Volt Light Bridge. The main reasons why you might want to do that are:

1. Lower costs for power supply – 5 Volt phone chargers are in principle sufficient to power this Light Bridge.
2. Arduino – Arduino has a 12Volt Power in, but actually runs on 5 Volt. When connecte to a USB, 5 Volt is all you get

I focused on simplicity before anything and building this schematic for the first time should be possible in one hour.

When you are done building and testing the Light Bridge, please take care

1.1: Schematics of a simple 12 Volt LDR Light Bridge

Schematics of a simple Light Bridge (click to enlarge)

This light bridge uses only two transistors, 4 resistors and a potentiometer / variable resistor to do all the work.

It will switch any connected module “on” by pulling M3 down / closing the circuit to the GND when the Light Source is no longer registered by the Light Sensor.

Choice of parts

I choose parts which are most common, and will not be in your way when you buy them due to obscurity.

In my case I am using the parts in standard “200 Resistor/Capacitor/Transistor/Diode” packages provided by Velleman, but similar packages are sold by DIY electronics parties in the likes of Velleman and Radio Shack.

Labelling

1. Ω, k and kΩ – Represents the resistance measured in Ohm and kilo-Ohms. For example: 1k or 1kΩ is 1000 Ohm

Parts list and role of each part

1. R1: 1kΩ – This resistor reduces the current to below 10 milli Ampere. As the maximum current for each diode is 20 milli Ampere, it will assure a long livity for your diodes
2. LDR1 – A run of the mill LDR. This is a Light Sensitive Resistor and the centerpiece of this design
3. R2: 50 kΩ potentiometer – You use this potentiometer to set the sensitivity of the Sensor. Together with the LDR this potentiometer basically forms a voltage divider. As the resistance increases on the LDR when light reduces, the potentiometer will “pull down” the voltage level on the Basis of the trnasistor to the point where it will stop conducting current between Emitter and Collector, thus switching “open”
4. T1: BC457B – An NPN transistor, used to amplify or multiply the voltage differences created between the LDR and the potentiometer. A slight change in Voltage on the Basis is multiplied/amplified between Collector and Emitter of T1
5. R3: 400kΩ – This resistor “pulls up” the voltage on the Collector of T1 and reduces the current-flow through the Basis of T2
6. T2: BC337B – An NPN transistor like the BC547 but with a larger allowed current throughput. Where the BC547 only allows to a maximum of 0.2 Ampere, the BC337 allows for 0.8 Ampere. We connected the Basis to the Collector of T1, using the “pull up” to switch T2 “on” and “off”. This way we reverse the state of the first part of our Sensor, which is “on” when it registers light and “off” when it registers no light
7. LED3: red or green indicator LED – This is your indicator of the registered State. It will help you to fine-tune the Light Sensor. It should be on when no light is registered and off when light is registered by the Sensor
8. R4: 2k2Ω – This resistor limits the current running through the LED to below 10 milli Ampere
9. D1: 1N4007 – This Diode pulls up the voltage level of the Emitter of T2 to 0,6 Volts. As the Collector of T1 never really is pulled to 0 Volts we use this Diode to raise the Basis of T2 above that level so that for T2 it is. The 1N4007 is a general purpose and  allows for 1A of current flowing through, so it will be

1.4: Understanding the LDR Light Bridge

Understanding the Light Bridge (click to enlarge)

1. State 1: Light on sensor – When the LDR in our Light Sensor receives light from any source, its resistance value will drop, “pulling” the Voltage “up” at the junction point where the Basis of T1 is connected
1. T1 “closes” the “connection” between Collector/Emitter – Due to the raise of voltage on the Basis of T1, the Collector/Emitter will start to conduct, “closing” the circuit it is part of
2. T2 “opens” – As T1 has pulled down the Voltage on its Collector (see Measure point M2), the Basis of T2 is pulled down was well, preventing T2 to “close” between Collector and Emitter
3. MOD1 is not connected to the GND – As T2 is “open”, MOD1 is not connected to the GND and for MOD1 it is as if the power switch is set to “off”
2. State 2: No light on sensor (“dark”) – When the light source is blocked, the LDR no longer receives light and its resistance value will go up to “infinity”, leading to a drop of voltage on the Basis of T1 as it is “pulled down” by Potentiometer R1
1. T1 “opens” the connection between Collector/Emitter – Due to the drop of voltage on the Basis of T1, the Collector/Emitter will stop to conduct, “opening” the circuit it is part of
2. T2 “closes” – As T1 opens, the resistor R3 will “pull up” the Basis of T2, allowing it to “close” the connection between Collector/Emitter and pulling Collector down to GND
3. MOD1 is connected to the GND – Via the now “closed” connection between GND and the “GND” provided to the Module, MOD1 now receives 12 Volt power and will start doing whatever it is designed for

Recalculating the resistors for different voltages

This circuit is designed for 12Volts, but 12Volts might not be the smartest choice.

5 Volt

5 Volt might be your most conveniant choice as 5 Volt power supplies (the ones to charge your mobile phones) are almost half the price of 12 Volts power supplies (6 euro for a 5V phone charger compared to 12 Euro for a 12 Volt transformer).

For 5 Volt, we roughly reduce the resistance by 2 or 3. This might not be the most scientific approach, but for this kind of circuits precision is not really required, so as long as you do not blow up your components, you are fine.

1. R1: 220 Ω - Each LDR takes about 1.4 Volts. Added up (as they are in serial) this is 2.8 Volts. The remaining Voltage is 5 Volt – 2.8 Volt = 2.2 Volt. As we allow roughly 10 mA to pass, we divide the remaining Voltage by 0.01 (10 milli) which gives us: 2.2V / 0.01A = 220 Ω
2. LDR1 - Unchanged
3. R2: 50 kΩ potentiometer - Unchanged
4. T1: BC457B - Unchanged
5. R3: 100kΩ – We roughly divide it by two, keeping the maximum current through Basis Emitter around 50 micro-Ampere and well below the threshold where you will kill T2
6. T2: BC337B - Unchanged
7. LED3: red or green indicator LED - Unchanged
8. R4: 1kΩ - This resistor limits the current running through the LED well below 10 milli Ampere
9. D1: 1N4007 - Unchanged

2: Schematics for the Infrared Light Bridge

[To be added]

Schrödinger's cat-dilemma: is it dead or alive?

We travel. We have a cat. The cat is old. The cat can not come with us in most of all cases. We currently have no-one to take care of the cat when we are gone. As the cat is old, she can die while we are gone.

We want to know if she is well.

Observing the cat by its movements

As long as the cat is alive, she will move through the house. If we can observe these movements, we can establish with certain certainty that she is alive. Movements are:

1. To the kitchen to feed herself
2. To the living room to sleep in her hammock or observe the pigeons on the roofs of the houses at the other side of the street
3. To the bathroom to relieve herself

Requirements

1. Movement-detector - We need something to moniter the cats movements. The simplest solutions are to use a light bridge or infrared-sensor. We go more into detail about this later
2. Wireless transmission of states – To avoid running cables through the house, I want to use wireless technology.
3. IO board to connect to computer/server – When we receive the signals, we need some kind of technology to feed it into a computer
4. Computer/server – To broadcast the state of the cat: dead/alive, we need a computer/server

Jumping towards the conclusion of this blog: Learning basic electronics

As I was researching several options including cheap electronics like 6 euro wireless doorbells and Infrared movement sensors, I found none really satisfied me.

Using wireless parts like XBee and Pololu Wixels to send the state of each sensor feels like overkill: like calling a cab to cross the street. Most parts in the chain can be solved by using very basic electronic components.

Additional blog posts: basic schematics to build sensors and do wireless IO

As my own knowledge of the basic electronics needed for this project is very limited, I will need to do some research and re-vitalize my knowledge: which will be posted in later posts.

Bottom line of these posts:

1. Building your own sensors and electronic components – For solutions like Arduino, Pololu Wixels and the IOIO board for Android
2. How does it work? – Basic descriptions of each part and how they work and work together

Using:

1. General purpose components – Like transistors, capacitors and resistors: which are multi-purpose and which you can buy in any electronics shop
   
2. Simple schematics – Which are easy to grasp and easy to build
3. Simple explanations – Which will be easy to understand for people who never did any electronics

Basic setup of the Schrödinger Box

Basic setup of the wireless observers in the Schrödinger box

To know with some certainty if the cat is still alive, we need to observe either the cat or the environment.

We choose to observe the environment for certain state-changes, in this case movement, to use as the basis of our assumption that the cat is still alive.

Movement in house == “cat is alive”

In other words: if we can somehow sense that something is moving through the house while we are gone, we can assume with high probability that the cat is still alive.

Possible solutions

1. Infrared movement-sensor – After testing with a low cost Infrared sensor, I discarded that solution. The 6 euro sensor simply is not reliable in its response. It is very sensitive to surrounding light, the additional light sensor interferes too much and the so-called PIR (Infrared sensor) seems to be only responding when I trigger it from very close distance
2. Light bridge – The alternative is to use a light bridge. A light bridge consists of two parts: a light source, shining on a light-sensor. If the light hits the sensor, all is well. If the light-source is not detected, something is passing the bridge and “breaking” it, marked as “activity” in the illustration above. The schematics for a light bridge is very simple and easy to build
3. Wireless doorbell – In the first prototype I chose two 6 euro wireless doorbells to broadcast the state of our activity-detector. Used separately, the doorbells function fine. When you activate the transmitter of one, the other remains silent. When you activate the transmitter of the other, the one remains silent. However, when connected to my experimental setup, both receivers start to influence each other, leading to both receivers being triggered when one doorbell is activated and not responding at all when the other is activated.
4. 2 transistor FM transmitter – About 25 years ago I built a very simple FM transmitter, consisting of 2 transistors and a hand full of resistors and capacitors. The schematic is shown below

Light bridge, "activity" is sensed when the light source is no longer registered by the sensor

Advancing the possible solutions

1. One receiver, many transmitters – To avoid the issues I had with the electronic doorbells, I decided to go for a “one receiver, many transmitters” solution. This solution has some disadvantages and limitations, but most are irrelevant for this project as we only have one cat. One receiver, many transmitters means that:
1. All transmitters broadcast on the same frequency – As you only have one receiver, you can only tune that receiver to one radio channel an d receive one state at the time.
2. You can not receive more than one signal - If more than one transmitter is activated at the same time, their signals will “compete” for reception. In most cases, the strongest signal will win. So only one of the two signals will be received
3. You can not use this when you have simultaneous updates – Since only one signal will be registered and processed
2. Using a different signal per sensor – To distinguish each sensor, we will be using a different tone / wave length per sensor. For instance:
1. Sensor 1: 100 Hz – a low humming tone, representing Sensor 1
2. Sensor 2: 1000 Hz – a higher pitched tone, representing Sensor 2
3. Sensor 2: 2000 Hz – even higher
3. Interference – The 2-transistor FM transmitter is quite basic and will “echo” its signal over multiple radio channels, apart from the channel we want and need to broadcast on.
1. Only short bursts – As our signal will be sent for only a short amount of time (less than one second when the cat passes) and is quite weak, we can assume it will not spoil the radio-listening pleasure of our neighbors
4. Wave / sound generator – To create the specific tone/signal per sensor, we need a wave/sound-generator.
5. High and low-pass filters – To filter out specific the signals from each specific sensors, we need to filter out each specific tone/signal/wave length and feed that to a “switch” on the receiving side
6. FM receiver – To receive the signals from our Sensors we need an FM receiver. As FM receivers are relatively complex (definitely more complex than FM transmitters) we use a cheap receiver from wherever we can find one
7. Tuning the transmitters and the receiver – Each transmitter will be tuned to broadcast on the same channel. Using the individual tone/wave generators as input for each transmitter and using the speaker of the FM radio/receiver, we can easily test and fine-tune the transmitters by simply tuning the transmitter until we hear the beeping sound on our receiver

Block schema for simple wireless activity monitor

Block schema of our solution

Brief:

1. Activity registered by sensor – Cat breaks light bridge
2. Power switch – “Activity” is sensed and used to turn on the “Power Switch” for the Wave Generator and Transmitter
3. 100 Hz signal transmitted – Wave Generator starts generating 100HZ signal which is broadcast on FM by the Transmitter
4. 100Hz signal received – The Receiver receives the signal and passes it to the Filters
5. 100Hz signal filtered – The 100Hz filter filters out the 100Hz signal and creates a “pulse” which is sent to “pin 1″ of the IO board
6. Activity registered by IO board – The IO board registers the activity related to the sensor that registered activity

Simple 2-transistor FM transmittor, used to transmit state: “activity”

I am a fan of your tablets.

I bought the Archos 7 Home Tablet and the Archos 32 and Archos 101 tablet as I needed them for my experiments. I will skip the G9 Archos 80 – 2×1000 as I want the G9 Archos 80 – 2×1500.

Increasing presence

I see your tablets pop up in positive reviews and in deals with phone companies next to the RIM tablet and the Samsung and Asus tablets. And I am very happy you finally seemed to have moved to higher resolution screens with (according to the reviews) a proper viewing angles.

(Awesome) Value for money

Your tablets deliver good (and almost awesome) value for money. I pay for what I get and – except for the crappy screens in the older series – there are no hidden issues. I recommend other people to buy Archos because I believe what you do- and deliver is not some slapped-together me-too product.

End of 2009

End of 2009 you were one of the few tablet producers with prices that made sense. For 200 or 300 euro I could buy a tablet that did what it needed to do and using the XDAdevelopers site I could find hacks for the Android Market that made it into a full product.

Sweex/Yarvik

Since then a lot has changed. This year Sweex suddenly appeared (in the Dutch market) with the below 110 euro ARM 11 Yarvik series. And the top-range of tablets have dropped in price to levels of 399 euro and less. You yourself launched the Arnova brand for the lower end of the market to – as I call it – sell last years technology for a good and very competative price. Basically in the same line and range as Yarvik.

For your information: I bought the Yarvik for research purposes. It is a decent product for the price it is sold for (88 euro on discount at “Computerland”).

Clouded product communication

But something else happened with your product communication: it became more clouded.

When I look on your site and especially when I walk into a shop it is unclear where your tablets stand now. How do you differentiate from other tablets? How do you differentiate between your own products? Why is the online pricing of your Arnova 10 with 199 euro so close to your Archos 101 (249 euro in your shop and 199 euro at Computerland) ? What are the real differences between the two 10 inch tablets? And why – with missing specs for the CPU on the Arnova (in the physical shop stated to have 600 MHz and sold for 119 euro instead of the advertised 199 on the Arnova site) – is it so difficult to find out what makes them different to begin with?

[Edit: October 22, 2011] I added some screenshots to the bottom of this page to illustrate this confusion.

Is Arnova itself a smart choice when it seems to lie so close to your Archos brand? Or is it just that the prices online need to be updated?

More choice, but unclear communication

I can buy an 7 inch Archos/Arnova Tablet with a 600, 800 (the old G8 HT) and 1000 MHz CPU (the latest G8 HT). Internal memory might range from 128 to 256 MB and maybe goes up to 512 MB but – specifically on your site – nothing clarifies that to me.

An overview on the current product/price strategy

Right now the Archos/Arnova 7 inch series is mostly competing with itself. I list:

1. The 99 euro 600 MHz (?) Resistive Touch Arnova 7 inch tablet
2. The 129 euro 800 MHz (?) Resistive Touch Archos 7 HT
3. The 135 euro 800 MHz Capacitive Touch Archos 7C
4. The 239 euro 1000 MHz Capacative Touch Archos 7 Internet Tablet

Less is more. Simpler is better. Especially with all other tablets people can choose from as well.

Sales people in shops seem not to “get” you

Sweex, with the Yarvik outsells you, if I can believe the “Computerland” shop (The Netherlands) sales rep. I spoke with. The tablet clearly has less to offer on all accounts except the screen, but the price beats you. And honestly, I did not get the vibe that the sales-people in the shops “get” what you offer more in the mid-range.

People go for simple choices

You seem to assume people recognize your brand and are well informed beforehand. You seem to assume people will take effort to research your products, read the reviews and have used it before they buy or sell it. In 99% of all cases they are and do not.

We feel. We compare on shallow levels, assumptions that do not hold in reality and incomplete data and then we buy. Our decisions are based on what we see and what we hear, yesterday and right there right then. In general we do not research. We simply go to a shop with a specific budget in mind, to buy “a tablet” for “a specific price”.

We go for – what at that point seems to be – a clear and simple choice, which is either/or based on:

1. Price – What we pay for what they think they will get
2. Trust based on familiarity – A familiar brand is automatically trusted more than some unknown brand
3. Advice – At the end, the advice of a shop-sales person is an important factor. If the sales person says: I would either go forYarvok or Asus while Archos is on display, something is wrong in your communication.

And so instead of an Archos, people might buy a Yarvik or an Asus tablet as the brandname is more known, the marketing and the buzz much louder (in the case of Asus, Motorola, Acer, Samsung) and/or the price much more competitive (in the case of Yarvik).

I love you

I trust your brand. I want to buy your shares on the stock market. I want your brand to kick ass in sales and presence on the mid-price range. As I believe that – as said before – you create good products and you deserve to rule in that region.

What I think is important to communicate (more clearly)

When I step back and look at the factors, I think the following matters to me as a customer:

1. Speed – CPU and GPU. Does it carry a 600 MHz, 800MHz, 1000 MHz or 1500 MHz processor? Is it an Cortex A8 or Cortex A9/ Is it single-core or dual core?
2. Generation – What is more? What is newer? Luckally you now mention “G9″ on your G9 products.
3. Memory – Why is it so hard to find out how much RAM your tablets have? It is one of the more important factors for speed and user experience. And for Flash. More is better.
4. Design – Does it stand out, does it appeal to me? To be honest: your G8 tablets are not really awesome. The G9 series seems to be an improvement. But you can do better. Release a new series next year that really looks 2012 instead of what might have been cool around 1998.

Suggestions to make your product stand out more

Right now your tablets are almost invisible. They do not stand out. Their (this counts for the G8 and Arnova series) design is not overwhelmingly awesome or different.

When looking at this, why not simply do something like this:

1. Arnova 70 G2 – 1×600 HT – Stating that it is the 6th generation 600 MHz single core Archos 7 inch tablet
2. Archos 70 G9 – 2×1000 HT – Stating it is the 9th generation, 7 inch, dual core 1 GHz processor
3. 1200 x 800 pixels / 256MB RAM / 4 GB internal storage / 1080p (1920 x 1080 pixes) true HDMI out – Add a sticker in one of the corners of the screen (handy for display models) stating what the screen resolution, RAM and internal storage is. Are you outputting true HDMI? Or just a blown up version of the 600 x 480 pixels you run on the screen?

Make changes that promote the strong points of your tablets and sales people in the store will notice.

The people buying thousands of your tablets for sales in shops AND the people on the floor will be able to compare your tablets and your brand to the Yarvik tablets and other systems and provide better feedback to the client. People who have access to all versions in a specific range (7 inch, 8 inch, 10 inch) can make a simple price/performance/possibilities decision.

Change the casing-design

Like stated before, I think your design of the outside is outdated. Write out a competition. Offer eternal fame, the newest generation Archos tablet (in that design) and a financial reward of 1000 euro for the one that wins. Approach some awesome product designers and pay them for conceptual designs which you will evaluate. Make your product look like an awesome mid-price tablet that is smart, looks cool and holds properly when you work with it. Halfway 2012 you would be able to launch a new tablet with a new awesome look.

Make clear where you stand compared to other Tablets

I do not read your PR to the buying departments of big parties like the big chian that MediaWorld is part of. Probably you communicate something similar as on your site.

Communicate how your tablets compare to the leading brands (a 300 euro tablet which is almost as fast as/faster than the Samsung Galaxy Tab II, can be connected to a generic keyboard, LCD screen, mouse and your computer via two generic USB ports and a mini (true?) HDMI out port) in the coutries where local or European law allows you.

Add prints to reviews of independent sites everyone knows so people can read how your tablets compare. Emphasize the parts where you are almost as good – or better – than your competitors.

Better communication means more sales

In your case, better communication means more sales. Your prices are very OK (although the general price drops caused by the dump of the HP WebOS tablet has made that driver less distinct) but what you lack is the support on the shop floor.

Make your tablets the darlings of the mid-range tablets. You deserve to be just that.

Thank you for reading.

Peter

Screenshots and additional info of confusing brand communication

The Yarvik 211, a 7 inch Resistive Touch ARM11 1GHz tablet

The Yarvik 211. Lower in price (due to discount) than the Arnova 7.

ARM11, 1GHz, 256 MB RAM, 3 hour battery life. Is it faster/better/worse than the Arnova 7?

Arnova 7. Apparently with 600 MHz Cortex A8 processor and 256 MB RAM

The Arnova 7. 600 MHz ARM Cortex A8 as far as I can find. RAM? I assume 256 MB. The tech specs on the Arnova site tell me nothing.

Hardly any info on the hardware. RAM? CPU clock speed?

How do I know this is better than the Yarvik? Is it? Or not? And if not: why not release something that is?

The Archos 7 Home Tablet v2. RAM? Processor?

The Archos 7 HT V2 is probably a Cortex A8, 800 MHz machine. Again, no info on the Archos site.

The Touch screen is Resistive.

The Archos 7C

For only 6 euros more we suddenly have the Archos 7C. This one sports a Cortex A8, 800 MHz. The Touch screen is Capacitive.

What is the difference apart from the Capacitive screen? Even the stock image is the same.

Archos 7 Internet Tablet

The question is: why not drop these budget tablets from the Archos brand and move them underneath the Arnova umbrella?

The next Touch project on my list involves a Surface Acoustic Wave Dual Touch panel.

I am choosing this technology for the following reasons/advantages:

1. Affordable multi touch – Capacative screens become increasingly expensive as size increases. SAW stays relatively low cost.
2. Less vulnerable – I noticed with a capacitive Touch screen from [insert here] that it is prone to scratching
3. Scalable – At NijTouch you can order SAW panels up to 42 inch. Resistive Touch Panels go up to 22 inch and apparenly it stops there for now, for the Business to Consumer market.
4. No ridges/Flat surface – The alternative to SAW is Infrared Touch, but these “panels” involve a ridge on top of your touch surface to house the IR LEDs/Sensors

Possible disadvantages and limitations:

1. Water and dirt – The SAW panels might be more subjective to water and dirt: acting as “touch shadows” casted by the dirt and water instead of by your fingers. I will have to test that. As SAW is used for more “prone to abuse” type of use like: “ATMs, Amusement Parks, Banking and Financial Applications, public information kiosks” I think it might be fine
2. Latency & precision – Also I do not know what the latency and precision is.

Application

The application I have in mind is for table-based Touch Surfaces, like in a cafe or in your home. The main goal is to have a multi-touch surface that can deal with the stress and abuse of day to day use of objects placed on the surface.

The alternative

The alternative I might fall back to for the final product is Resistive Single Touch.

Surface Acoustic Wave Touch (SAW)

From TVI Electronics :

The Surface Acoustic Wave (SAW) technology is one of the most advanced touch screen types. The technology is based on two transducers (transmitting and receiving) placed for the both of X and Y axis on the touch panel. The other important element of SAW is placed on the glass, called reflector. The controller sends electrical signal to the transmitting transducer, and transducer converts the signal into ultrasonic waves and emits to reflectors that are lined up along the edge of the panel. After reflectors refract waves to the receiving transducers, the receiving transducer converts the waves into an electrical signal and sends back to the controller. When a finger touches the screen, the waves are absorbed, causing a touch event to be detected at that point.

Compared to Resistive and Capacitive technologies, SAW technology provides superior image clarity, resolution, and higher light transmission. Because the panel is all glass, there are no layers that can be worn, giving this technology the highest durability factor and also the highest clarity. Disadvantages of Surface Acoustic Wave (SAW) technology include the facts that the touch screen must be touched by finger, gloved hand, or soft-tip stylus (something hard like a pen won’t work) and  that the touchscreen is not completely sealable, can be affected by large amounts of dirt, dust, and / or water in the environment.

The Surface Acoustic Wave technology is recommended for ATMs, Amusement Parks, Banking and Financial Applications, public information kiosks, computer based training, or other high traffic indoor environments.

Surface Acoustic Wave itself apparently is a process/theoretical principle already discoverd in1885. Read the Wikipedia article here. Touch screens are briefly mentioned in that article:

SAW sensors have seen relatively modest commercial success to date, but are commonly commercially available for some applications such as touchscreen displays.

[Edit: Oct 16, 2011] One of the readers of Hack a Day pointed out that a 23 inch Touch panel could not be found. I guess I made an err here.

My version of a 23 22 inch Android Touch “mega tablet”

It uses a single-touch resistive 23 22 inch Touch screen with an Difrnce DIT7000 Android tablet I bought for 135 euro on a heavy: “everything must go” sales discount.

I had hoped for higher resolutions of at least around 1200 x 800, but the HDMI output resolution of this tablet is probably also 800 x 480 pixels.

The video that triggered this experiment

The Megapad by Martin Drashkov.

As I already had done some experiments with resistive touch screens and seen the insides of another Resistive Touch tablet (see this post on taking the Archos 32 apart), I guessed it would not be that difficult to get stuff working

Also see

1. The Scrablet – The concept, the execution. A 15 inch DIY mod, turning a Laptop screen into an external Touch panel
2. Taking apart the Archos 32 – See this post About fixing the broken power switch, including a photo series of the insides ofd this tiny tablet
3. Touch tablet concepts – Here. Once made for Wacom, for a possible gig that did not get through.
4. “Minority Report interface” With FLAR Toolkit and Flash – Here. Originally to test AR-toolkit markers for a table-setup, but done with “finger gloves” for fast prototyping
5. HotForestGreen – an “Internet of things” framework here that will be used in future projects with this (and other) screens

Basic setup

Schema of basic setup - Android 23 inch Touch Screen

The Android tablet is connected to the screen via a standard issue HDMI cable (with Mini HDMI connector for the Android tablet).

We connect the 4 wire Resistive  Touch Panel (or Touch Digitizer) with a home brew cable to the Android tablet, replacing the built in Touch screen of the Tablet itself.

To make sure we can run this setup until kingdom come (or one of the components fails) we connect the tablet to a power supply.

Wiring – and flipping the pair

Jumping ahead in the story: this is the (assumed) setup of the Touch Digitizer wiring and the expected and implemented layout of the home brew cable to connect the external (23 inch) Resistive Touch Digitizer to the tablet.

4 wire resistive Touch Panel layout and wiring

A 4 wire Resistive Touch Panel has a relatively simple layout. See “A: Example layout of touch digitizer” in the image. I did some research online (see for instance this PDF document) and deducted the 4 wire connection schema by simply measuring the resistance between each wire pair, before flipping the wires.

In most cases, the wiring would be straight forward (see “B: Expected layout”): A to A, B to B, C to C and D to D. For some reason, the Android tablet I used considered the Touch film to be  implemented “upside down” where: “up” become: “down” and vice versa (see first part of my video).

This was corrected by twisting the pair connecting to the Y-axis digitizer (see “C: Implemented layout”). The layout became A to C, B to B, C to A and D to D. (In the image I show the position of the Touch Panel wires: C, B, A, D.) The Touch screen input then worked as expected.

As I only did this experiment with one Tablet, I do not know if this is the exception or the rule.

Photo: Getting the wiring right

Getting the wiring right: connecting A to C and C to A

In the picture above you see how I solved the issue on the wrong wiring: by using 4 clamps and switching “A” to “C” and “C” to “A”. If you look closely you see that the yellow (going from “A” to “C”) and green (going from “C” to “A”) clamps are the ones I switched.

Ingredients (and where to get or buy)

1. A resistive touch Android tablet – To run Android and allow you to attach a 4 wire touch screen. I used the 135 (199 without discount) euro Diffrnce DIT7000. It runs Android 2.2 and has mini HDMI out. I bought it at a local store that had a big sale
2. A Do It Yourself resistive 4 wire Touch digitizer – I bought mine on eBay: a flexible touch film for a 23 inch screen. Cost around 90 euro. There are two sellers I trust: Visual touch (for flexible film Touch panels) and NJYTouch (Touch panels on glass surface). Both delivered the ordered product as promised. Additional info: NjyTouch also came up to their promise to replace a broken Touch screen when the first arrived in that state.
3. A LCD screen with HDMI input – I had a 150 euro, 23 inch Samsung from a previous experiment, bought in a local store
4. Thin flexible wires – I harvested the insides of an USB cable from some device I killed
5. Pins to connect your cable to the Touch digitizer – Bought at local electronics store
6. A HDMI/Mini HDMI connector – To connect your tablet to the big screen, bought at a local store

Tools and other stuff

1. Screwdriver, knife – To pry your tablet open
   
2. Scissors – To cut the 4 wire flatcable from your tablet
3. Sandpaper – To uncover the wires in the tablet touch screen flatcable
4. Solder tin – To solder
5. Soldering iron – To solder
6. Super glue and paper – To make your connector hulls (see photos). I used Loctite flex gel (superglue) and the paper from an envelope
7. Clamps – As uses in the photo above. If you do not have them, buy them. They proved out to be very handy

Estimated required build time

1. 4 hours to modify Android tablet/Touch cable and make it work – It took me about 4 hours to do all the handy work and get the first proof of concept running. As I already had the setup with the LCD screen and the Touch Panel, that time is not included.
2. 6 to 8 hours for all – I think a first time implementation of the full setup (including the LCD screen and Touch Panel) can be done in 6 to 8 hours time.
3. 2 to 3 hours when repeated – Naturally, the first time implementation includes learning time. I made several small mistakes I had to correct, and there were some things (like the wiring of the Resistive touch Panel) I had to figure out. When repeated, most of that learning time can be skipped

Trying it first – Frankensteining two tablets

As I did not have the touch screen at hand at first, I used another resistive tablet for reference. The main question to be answered was: are 4 wire touch screens interchangeable?

So I opened both low cost tablets (110 euro 8 inch Yarvik Tab310 and the 135 euro Difrnce DIT7000) I bought for tinkering purposes. I then did a Frankenstein with the touch screens. Photo and video below.

Frankensteining two tablets: linking one's touch screen to the other tablet

The tablet on the bottom is the Difrnce borrowing the Yarviks Touch screen.

Going pro: Connecting it to the 23 22 inch Touch screen

Once I had access to the Touch screen and monitor I started to do the required modding. Below are the steps:

1. Open the resistive screeen tablet – Depending on the manufacturer this might be quite the ordeal. The Difrnce DIT7000 was somewhere in the middle. Not easy, but simpler than – for instance – the Yarvik
2. Disconnect the 4 wire Touch screen cable – This is a 6 mm wide flat cable with 4 wires
3. Cut a 1 cm long piece from the flat cable – Just long enough to use it as your connector and short enough to reconnect the remaining length of the original cable to restore the tablet in its old function
4. Clear the end of your 1cm flatcable – Using sandpaper until you see the bright shiny surface of the copper ends
5. Cut 4 lengths of at least 30 cm of thin wire – To allow some space to work between the tablet and your touch screen
6. Solder the cleared ends of the flat cable – To create your connection points for the wires. Photo of the end result below
7. Solder the wires to your flatcable/connector – Be careful and use a soldering iron with a thin tip
8. Solder the wires to your connector – Using tin. As i used cheap quality — the hull melted a bit. Prepare the pins with tin beforehand to make soldering easier and the reuslt more reliable
9. Test – I found:
1. Not all the same wiring: down is up, up is down – That the Difrnce tablet crosses the wires for the Y-axis readout of the 4 wire Touch digitizer. So “up” is “down” and “down” is up. See my movie and the layout of wires above.
2. Direction matters – There is a clear “pin 1, 2, 3, 4″ situation. If you connect “pin 4″ to “pin 1″ the tablet will not respond to touch
3. Try the standard setup first – Connect the pins straight forward first and make it work before you start solving possible issues like #1: down=up, up=down
10. Make the hulls for both connectors using paper and super glue – To avoid your cable/construction to be too vulnerable. The wires you use have a 3 or 5 wire core of very thin copper wires. They break easily. Photos of the end result below.
11. Use – Your connector cable from your 4 wire Touch digitizer to your tablet is done.

Photos

The insides of the Difrnce DIT7000 - the disconnected 4 wire Touch cable

The make-shift Touch connector with cleaned wires

Make-shift connector with USB cable wires soldered to it - the white is paper superglued to the connector

The connector to the 22 inch Touch digitizer

Superglue and thick paper to make a solid wrapping/encasing for our connector

Almost done: finalizing the connector with extra support/sturdyness for the wrapping

The setup: screen and tablet, with Touch screen connector

When you build stuff that needs some kind of computing power beyond the 20 euro PLC/Arduino scope, Android devices might be your first weapon of choice.

When I started to focus my attention on Android in 2010, the cheapest hardware you could find where the Apad tablets, offered for 130 USD, excluding the import costs.

Archos had already launched their first series of 7 inch tablets: running on a 600 MHz ARM Cortex A8 processor and with 128 MB of RAM, obtainable for around 190 euro in that period. And in the end I bought just that.

But for home made projects, where your “computer” is hidden in some closet somewhere, even 150 euro is still a lot of money. So I am patiently waiting for the hardware to drop below the 100 euro point. And this months (from Juli on) it has.

Two reasons why I prefer Android above other platforms

1. Java and code portability - My (Java) code is not locked to one specific platform as is the case with – for instance – iOS and Symbian. If I do it properly: by abstracting and encapsulating platform dependent parts, I can compile and run that same code on any platform of choice, including Linux versions, the Mac and the PC.
2. Price tag and competition – Android is not locked to one vendor or one producer. Anyone can build and sell Android devices. This means that there is more choice and more competition in the market. As one consequence, more and more “last year devices” (like the Yarvik and Arnova products) will continue to hit the market to target cheap skates like me, dropping in price as production costs drops on the long tail as well.

Two devices to toy with

88 euro Yarvik Tablet: 256 MB RAM, 1 GHz ARM 11 processor

Yesterday I bought the Yarvik Tab 211. Yarvik is a sub-brand of Sweex, which specializes in OK hardware on the low end of the market. It is not the best you can get, but it is good for its price. And with the specifications and build of the Tab211 I think they hit bulls eye.

The tablet cost 88 euro in a special offer at “Computerland” (see image below), a retailer part of the same group that contains “Media World” / “Media Markt”.

Yarvik Tab211, 88 euro at "Computerland"

It is the “high end” tablet in their range of two 7 inchers and below is a part of the list of its specifications:

The resistive screen is quite OK and it comes with a USB host cable included.

Yarvik USB Host

The USB Host was something I actually did not expect to get with the tablet.

It allows you to connect a keyboard, a mouse and a drive to the device. And if you use a USB hub (with an independent power source to feed the harddrive) you can use all at the same time, as I did with the Archos more than a year ago. (see this post of my “Hacking away with Android” series).

Brief: it works.

100 euro LG GT540: 600 MHz, 256 MB RAM

The LG GT540 has hit and went under the 100 euro price tag in Italy, Germany and the Netherlands. Unfortunately this offer has apparently expired and the cheapest offer I could find now (August 2011) is 103 euro.

LG GT540 for 99 euro

LG GT540 for 103 euro

GT540 and Android 2.2/2.3

The GT540 apparently still comes standard with Android 1.6 and can be upgraded to 2.1 (Eclair).  As this phone is selling for over a year and gaining more and more user, there are mods as well to upgrade it to 2.2 (FoYo) and 2.3 (Gingerbread).

FroYo makes it possible to use this phone as a WiFi hot spot, using a data plan from your telephony provider..

I have done an upgrade for another “old” phone: the Samsung Galaxy I, and assume it will be equally painless for the GT540.

But what if you want to do more?

Protecting your idea before entering a partnership

Say you have the golden opportunity to start a partnership with other people.

And the outlook is really promising. Something like: “several million euros in contracts in the next twelve months”.

Where you deliver the technology, they deliver the sales, the network, the corporate knowledge and commercial know how.

If all goes well, you will all be happy and have a lot more money than when you started.

However: if things go wrong, you might lose your awesome solution, for instance because your partners until then will sue you dearly if you ever use it again, somewhere else, in any future project. Simply because it is their policy and because they can.

There are several ways to protect your idea and your software, but the main question is: “what do you bring into the deal to begin with?”

Legal headaches

How do you protect what you perceive to be yours? How do you make sure you can te-use what you brought into a business deal when things go pear-shaped?

Who can help you avoid putting your head into a legal noose?

Your expertise, not your solution

One way I feel is best – if you feel an unfortunate premature exit is one of the possible outcomes – is to bring in your expertise alone.

Whatever you produce during the partnership might be of all involved at that time, but your original product might better stay yours.

Again the Public Domain might be a good solution, if you dare to take it. Where you might loose the cutting edge – as others, including your competitors, will have access to it too – you might keep the thing you put your heart in to begin with.

If the market is small or: when this might be a bad idea

If there are only a hand full of players, wolves prowling on the other side of the road and only a hand full of potential clients, releasing  your source code (as this is what we talk about) might be a big mistake.

Also, you might have dreams of world domination.

In that case keeping your stuff secret might be (or seem to be) the only way not to lose your position.

When your business model is service driven

On the other hand, maybe your idea is not that unique to begin with and making it open to the public domain allows you to move much easier.

What to keep in mind with going Open

If you go Open, there is hardly any room for mixed feelings. If you go, go fully and make sure you have no second thoughts.

Open Source has a completely different business model than Closed Source. One reason is what I mentioned before: if it is in the Public Domain, your competitor will have access to it too.

What defines your cutting edge?

The question is, however: what do you compete on? And what changes if your business model changes?

I like to use the example of the Boiler and Geiser for this. Both are used to provide hot water. Both are produced in mass and both are based on relatively simple technological principles. Both are omni-present. Any and all households in the Western world has either one or the other.

As they are mechanical constructs, you need maintenance.

As they are everywhere, you need relatively many people to do that maintenance. Enough to build a business around.

What defines youir cutting edge?

Are you special due to some secret sauce, or because of what you can do for others with the tools you produced?

Do you want to go for secrecy and (as a possible next step) to complex legal shit, or simply release and build your business on services or spin off products? Both are valid as a starting point.

Your secret sauce makes it possible to keep the cutting edge (if you are good) and be the only one licensing that software. As Microsoft shown, that can be very lucrative. It also involves a lot of extra overhead as you will need more and more effort to protect that secret and your market share.

When you open stuff up, your main competitive edge is the quality of you and your vision. Put into an example: Even though anyone can make fried potatoes, but yours might be so super awesomely tasteful that people travel from the other side of town specifically to get yours. Here too is increased effort required to remain in the fore front as time progresses and others learn from your success, but as you have fewer things to worry about (no secrets to protect and a simpler organizational model) you need relatively less effort to make it all work. You can basically simply focus on doing your job.

Where I stand

I want several things:

1. To build a steady business – Which allows me to keep and increase my financial independence
2. To create for creations sake – As I like to make beautiful things that gives me satisfaction. It is basically my legal drug.
3. To create something awesome with others – As I am limited in what I can do alone and working with the right people (as in: the right fit/chemistry) can be very exciting, very stimulating and a lot of fun.

Starting point

One of the reasons was a somewhate strange situation with a client who – after a project was closed – suddenly started to press on having a contract signed by me with which I would transfer all the copyright of all the work I did for their clients over to them.

Copyright, Freelancers and the Netherlands

In the Netherlands, as a freelancer, you are an independent agent. All the work you create for a client is your Copyright.

The only way to transfer that Copyright is when you sign a contract with a Waiver of Copyright and by transfer on a case by case basis.

Framework

As some work was done with a framework I was working on at that time, I said “no”, as that framework was not in the public domain and did not contain any Licence Info. By signing the contract I would (possibly) automatically transfer the Copyright of that work into their hands as well.

Fight

For a while the relationship was not funny and the pressure applied did not add up to for what it was worth. There are many ways to assure that:

1. I would not abuse the code I wrote for that project
2. They (and their client) would not have to fear any claims from me on hind sight

After some heated calls and mails we agreed and we would only transfer the copyright of the work I did for that specific project.

Libraries of awesomeness

When you are like me and create work for a Client as an independent, there are three possible parts:

1. Generic stuff – The work you create specifically for the Client and could not care less about afterwards
2. Re-usable (awesome) stuff – Some smart and generic solutions you can re-use for other clients
3. Your library of (awesome) stuff from past projects – As you move on, more and more tricks have been added to your pouch of wonders

Number 2 and 3 are what makes you more valuable/relevant for your current and future clients. Your re-usable stuff already went through several cycles of improvement and allows you to make several shortcuts that saves you time and allows you to produce better results than when you had to do it from scratch.

To me, this is more important than anything. Including future assignments from one specific client.

Your libraries, agencies, Open Source and Copyright

Bottom line: If you have built libraries of your past awesomeness, written in your own time or re-useable with the consent of your past Client, based on your experience and to be used in future Client projects, consider to make it Open Source.

Any Agency is in principle right to ask you to waiver your copyright for the assignments you do via them.

Not because they have a right to claim your work as their own, but to avoid trouble for their clients in the future.

It will not be the first time that a Freelancer causes a lot of issues later down the line as he or she executes the copyrights on his or her work and claims several thousands of euros or blocks the right to use that software – leading to direct or possible damage for the End Client as whatever commercial endeavor could have started is halted until the issue is solved via the lawyers or a judge – as the End Client (might have) infringed something somewhere.

If and when you make your general purpose Libraries available as Open Source to the Public Domain via one or another license, the work is much more likely to stay yours when you (re) use it in a future project.

Clients who do and do not care about Copyright

Depending on who you work for and what conditions they are used to, you might have two type of assignments:

1. Free to use and re-use your own work – One type of client just wants the job done. They assume you are not there to fuck them over afterwards. Whatever you produce is yours and they are completely OK if you (re)use your code. As long as you do it in a way that does not harm them or feels like a backstabbing job.
2. Anything you create is theirs – The other type wants to be sure you will not fuck them (or their clients) over and has the standard policy to anyone and everyone to have them (and you) Waiver their Copyright.

What to keep in mind

When you sign a contract as a Freelancer, in which you waiver all your copyright to the Agency (and usually via the Agency to the End Client), the work you produce under that contract and during that project is not yours.

If you want to incorporate what you learned under a contract in which you Wavered your Copyright, refactor the code. Any line you copy and past from the result of such an assignment is you breaking the law and becoming liable for a lawsuit, as the code and the Copyright is no longer yours.

Re-use of a concept is not the same as a copy

Whatever concepts you introduced in a project where you wavered your Copyright, you are completely free to re-use those. (Unless someone ran to the Patent Office and patented your solution).

If you use the existing (Copyrighted by someone else due to contract) code as a starting point in – for instance – an aforementioned Refactoring, you are fine.

Respect non-competition

If your solution is also to direct strategic benefit for the competitor of your past client and you use the (refactored) code for their benefit, you might not directly breaking a law, but you are threading on thin ice.

Selling a solution Client “A” paid a lot of money for to Client “B” for a lot less, removing the potential advance “A” might have had will piss people off when they discover the link to you.

Apart from that: it is simply not done.

Wireless RFID card readers

Looking for a simple way to create a series RFID readers which are able to operate wirelessly, I came upon the following post by Josh Tananbaum / the GeekMovement .

It describes the creation of a RFID reader “TUNE glove” and uses 3 main components:

1. The Innovations ID2 RFID reader
2. An Arduino board
3. An XBee module

I bluntly copied the RFID part of “parts needed” from the GeekMovement post here.

Parts Needed:

• 1 x Innovations ID12 RFID Reader
• 1 x RFID Reader Breakout
• 1 x LM7805 5v Voltage Regulator

The ID12 is a short range RFID reader that sends the ID on theRFID  card in a serial stream through pin 9 to any device including the Arduino board in the TUNE Glove.

In that solution the Arduino board then sends this data to the XBee module, using the NewSoftSerial library.

XBee might be sufficient

What the code for the TUNE Glove seems to do is contstruct the numbers and then pass them from Arduino to XBee.

It is possible that you do not need the Arduino. This post by Lynnlinse seems to suggest (using a different type of RFID reader) that is doable. As the ID12 sends serial data in a specific Baud-rate, the XBee can be used to simply broadcast that wirelessly to the XBee base station. The software on the main computer, which can be a solid state machine like the FitPC2 or a low cost machine like a Netbook, can then translate the raw serial data from the reader.

Something to test out. It would dramatically simplify remotely readable RFID readers as you will only need the RFID card reader and the XBee module.

The ID12 RFID reader

ID 12 and Arduino board. Image curtosy to hc Gilje

Creating an RFID reader appears to be relatively simple, thanks to the use of serial data.

The ID12 has two Digital Out ports: D0 and D1. In the schema presented for the TUNE Glove, D0 is used, located at pin 9. The other pins connected are:

• Pin 2 (RST) and pin 11 (SV) to +5V
• Pin 1 (GND) and pin 7 (FS) to GND

3.5 and 5 Volt voltage regulators

As a layman, what I did not know existed are the single-chip Voltage regulators. The base models are step-down regulators with an output of 3.5V and 5 Volts. Perfect when you need to add XBee or WiFly to your experiments.

Arduino NewSoftSerial library

Another neat thing is the NewSoftSerial library for Adruino, allowing you to turn any set of pins into additional RX/TX pins with free to set baud rates.

It is based on Ladyadas  AFSoftSerial, uses “interrupt driven receives” and is written by Mikhal Hart.

What is handy to know is how to use the library, which is quite simple.

I copied and pasted the SoftSerial elements from the GeekMovement TUNE Glove project.

// Define the SoftSerial object called "xbee" and the pins it will use
NewSoftSerial xbee(rxPin,txPin);
// Start it with a speed of 9600 baud
xbee.begin(9600);
// Write data
xbee.print(code[i], HEX);

Other RFID projects

I found another RFID project on the Arduino Playground Wiki. It does not state who wrote the article. The cote is attributed to: “vgrhcp”.

That project also uses the NewSoftSerial library to allow communication to both the RFID reader and a computer or device.

Electric strike lock

Electric Strike lock - Curtosy to Alibaba

The Electric Strike lock is what you will find in almost any door with an electric opener via an intercom system.

Its internals contain an electro magnet that – when activated – releases the strike and allows the door to be pushed open.

When the electromagnet is disabled, the strike will block again.

You can buy Electric Strike locks operating at 12 Volts. Conveniently the same amount as for an Arduino board.

This post explores the basis for a scalable, low maintenance system to automate the commercial exploitation of office space: including simple ways to grant and dany people access to Flexible Office Space.

The purpose of this series is to create the foundation for a concept in which existing and vacant Office Space can be broken up into commercial units which can be leased separately without creating extra overhead for the Owner/Exploiter.

Related links

1. Basic principals – See this post for my basic exploration on office space, the classic models of exploitation and the added and negative value of property for its surroundings..

Three issues on exploiting a building

We begin this post by summarizing three issues on the exploitation of a building. More details can be found in the “Basic principals” post as mentioned in “Related links”.

1. Administrative costs – Including contracts, invoices, assuring and handling payments by clients.
2. Handling costs – Including handing over keys, creating keys and delivering the office space
3. Efficiency – The more office space is leased to less clients the better

Three issues on current office space

1. Supply and no demand – There is more office space available than needed in the business. The result is loss of revenue and loss of value.
2. Decay – Empty office space is prone to vandalism and decay.
3. Negative added value – For the city, as there is no business, no flow of people and the building itself is a dead space.

Three potential groups of clients

Offering attractive conditions to rent office- and working space

1. European technology start ups
2. Freelancers and small businesses in the city
3. Social businesses – like restaurants, small educational companies

Main focus of the sytem

1. No overhead – The system should reduce any form of overhead towards the building owner
2. Automated – Any transaction between a Client and a building should be automated: without the need of any additional human interaction
3. Clear and simple – Any Client should be able to make transactions and understand exactly what the limitations and options are as a result of that transaction
4. Hard to fraud – Anything in between that can instigate fraud should be kept out. Even if there might be a: “business opportunity” left out.

Seven elements

1. Online – Finding and booking vacant space. Using a database with real time information on which units in which buildings are available, reserved and taken
2. Local – Closing the deal via PIN-payment and a Touch screen.
3. RFID – Using general purpose RFID cards to gain access to a building and a unit.
4. Payment up front - All units and work spaces are paid for up front.
5. Limited time access – All and each unit are made available for limited time. A space can be reserved for a long period of time, spanning multiple months, but if payment is not done for the next period, access will be denied to that unit to that client.
6. Full Client responsibility – The Client is fully responsible for maintaining access rights to a building and office
7. Automated administration - Invoices and reminders (regarding new payments) are sent via e-mail to the Client and the Building.

The basic flow to access a building and office

Basic flow to get access to an Office and Office space

Three ways to get access to an Office Unit

1. Reserving online, payment on spot – The first way is by reserving a space online and pay at the office. Included is an QR-Code produced after online reservation and to be shown to the local system when starting the payment process.
2. Selection and payment on spot – The second option is by selecting the unit and period of time on the spot. And do the payment immediately.
3. Via a guest-account – The third is via a guest account, where the Client grants you temporary access to the building and the unit the Client paid for.

In all cases general purpose RFID cards can be used

OV Chip cards as issued in the Netherlands

General Purpose RFID cards are any card with a readable RFID chip. This includes:

1. OV Chip card
2. Oyster card
3. RFID tag on sticker or 3rd party (bonus/air mile) card

In most, to all cases, payment is on the spot

1. No fraud – To cut out any possibility for fraud between the service provider and the Commercial Exploiter
2. To allow scalability and independent development - To allow for maximum decentralized scalability of the concept
3. Simplification – Adding a man in the middle adds an extra layer of complexity, more administration and requires a bond of trust. To keep the financial process simple, payments are made electronically, directly to the party exploiting the building.

Payment is done via electronic payment system, coupled to the bank account of the Commercial Exploiter of that building.

Gaining the right to Access to an Office Unit

Reserve online, pay on the spot

Reserving online, paying at location

The biggest benefit – when you are a first time Client – of arranging the reservation online is speed.

Since you have to go online to find out which location has what space available you just as well might fill in all the basic data required to reserve a space properly:

1. Your e-mail address
2. Your name
3. Your home-address and registration number for taxes, if required

The check-in process on a location is – in this case – limited to the following steps:

1. Present your printout with the QR code
2. Insert your RFID card
3. Make the payment via the electronic payment system.

Reservation and payment on the spot

Reservation and payment on the spot

The check-in process on a location is – in this case – limited to the following steps:

1. Insert your e-mail address (for announcements and your invoice)
2. Insert your name (if it is not known yet)
3. State the required time frame
4. Find and select an available unit
5. Insert your RFID card
6. Make the payment via the electronic payment system

When space is not available on that spot

It is possible to link all available locations and suggest an alternative location / Office / Building offering a matching Office Unit in the matching time frame.

Making

What RFID cards provide

RFID Cards provide:

1. A unique machine readable ID number per card – Making it possible to distinguish separate clients and keep access to a specific unit exclusive to those RFID cards registered to a unit
2. A quick and reliable machine read out – Using an RFID reader, the scan of the ID on a card is done under one second
3. A cheap, safe and easy to gain object- Due to the introduction of the OV Chip card and the overall increasing use of RFID cards anywhere, each person has or will have at least one RFID card in the pocket. As the number per RFID card is unique and personal, the RFID card as a key is safe as well.
4. A simple way to grant a person access – To a building and unit, by adding that card ID to the “Granted Access List” of a building and Office Unit.
5. A simple way to deny a person access - To a building and unit, by removing that card ID from the “Granted Access List” of a building and Office Unit.

Four safety aspects of using RFID

1. No copies - Unlike a normal key, an RFID card is not a copy issued to one or more people.
2. Unique, physical and bound to a person - Each RFID card is unique. Granting access via an RFID card is granting access to the person who holds that RFID card. Denying access is also bound to a specific person.
3. Access is personal and explicitly based on the unique number or ID of the card - What we use is that unique aspect per RFID card, bound to only one person: represented in a number that can be read from that card.
4. Machine readable and hard to reproduce - Unlike a pin-code you type in, the RFID code is hard to copy or reproduce. As it is machine-readable, it is also a very fast and easy way to gain access once access is granted to the uniqque number on that specific card.

Limited time access

By default the access to a building and Office Unit via an RFID card is time limited. This means that:

1. No access is granted outside the granted time period – The door(s) will not be opened if the card is used outside the granted time period.
2. Access to a building will automatically expire – Unless the Client actively renews and prolongs the period to which he or she needs access.
3. No action will automatically lead to expiration of Access Rights – Meaning that the Client can not enter afterwards due to human error.
4. Maintenance of Access Lists is simplified – Due to the automatic expiration of access rights, Clients and users will be automatically removed from the Access Lists unless action is taken to prevent that.

Design rules of Guest access

A brief overview:

1. Definition of Client – Clients stand for “paying clients”.
2. Definition of Guest – Guests are guests of the paying Client. They can be people hired for temporary jobs (temp workers, interns, freelancers) or guests visiting for a specific period of time, with the need of individual access to the building
3. Client Access – Client Access is for the duration of the pre-paid period.
4. Temporary Guest Access – Guests can be given temporary access to the building and the specific unit of the Client, by pairing the Guest RFID card to the RFID card of the Client.
5. Set to name – The registered Guest RFID card should be registered by name, for later reference. See: “Repeated Guest Access”.
6. Maximum time limit for Guest Access – The maximum time a Guest is granted access should be 8 to 10 hours per day, from the moment of Guest/Client RFID card pairing: to allow temp workers and freelancers free access to the office during that working time.
7. Guest Access time limit set by Client – The maximum time a Guest is granted access is to be set by the Client.
8. Repeated Guest Access – In case of Repeated Guest Access a once paired RFID card can be scheduled to be “activated” online by the Client.
9. Scheduled activation of Guest Access is limited – The Client can only schedule repeated Guest Access for a limited period. This should be no longer than one work week, ranging in most cases from Monday to Friday.