This module concerns sensors, and you might be thinking in terms of electronic components.
But your wonderful new commercial product could be a snazzy new gadget, or it could just be a clever new software application. So we must widen the scope to include the exploitation of sensors and their software that are already built into desktops, laptops, smartphones and the host of other consumer products.
You could be exploiting a new sensor that has just come onto the OEM market, or you could be tapping into the sensors that are already serving some device.
In either case, you must address something that will make it attractive to buyers.
When I was very young, there was a need to prevent your kettle running dry. A marvellous new invention was a whistle, driven by the steam.
Today, the sensor of an electric kettle is still driven by steam, but by the steam’s heat on a bimetal switch. If you leave the lid open, the steam will not be directed to the bimetal strip, and the kettle will boil until it is dry.
These switches are used in almost every heating-based appliance, irons, room heaters water heaters and such. Why should an electronic device compete with them? Foils of two dissimilar metals are bonded together having different coefficients of expansion. Heat will cause the strip to curl and open a contact. A magnet, or a bent tag that acts as a spring, gives snap action, so that slowly opening contacts do not burn out.
My Portsmouth UK consulting company looked at the problem of sensing something boiling inside a Sharp microwave oven. We could not use the heat of the steam, but had to rely on detecting water vapour. But thirty-odd years ago there none of the ‘vapour detector’ add-ons that you can now attach to your Arduino.
The device that we were asked to try relied on the fact that water vapour has lighter molecules than air, so is more efficient at removing heat from a heated filament. I have to confess that it did not work very well, and did not go into production. But maybe there is no shame in that:
Do a web search for “Microwave oven detects boiling.”
As far as I can see, nobody has succeeded yet. Perhaps there is still an opportunity for you.
An earlier consultancy had involved detecting leaky tins. They were filled with compressed sulphur hexafluoride gas, and an electron capture detector was used to sniff out any leaks.
A thin foil of a radioactive isotope of tin emits beta radiation, electrons. A voltage difference draws them across a gap to be captured by an anode, causing a minute current to flow. But if there are electron-hungry molecules in the gap, such as sulphur hexafluoride, the electrons are snaffled before they can reach the anode.
My involvement was to devise a way to amplify that minute current.
I used it to charge a small capacitor. When the voltage reached a threshold, the circuit discharged the capacitor. So the output came in the form of the interval between discharges. It worked well.
A company director told me that the detector was very sensitive to explosives, such as nitro-glycerine. I mentioned this to a friend at the BBC, and it seemed that almost overnight every airport was equipped with ‘bomb-sniffers’. I am sad to say that Cambridge Analytical Instruments was wound up in 2016.
Quite independently it was realised that the electron capture detector is excellent at detecting combustion products. All over the world, it is at the heart of fire safety detectors.
But not in Australia!
Out of fear of all things nuclear, Australian smoke-detectors have to use an LED and a phototransistor, to detect nothing until the smoke is thick enough to see.
Fear is driving a lot of needs. One fear is of intruders.
For decades, arriving customers have been detected by an infra-red beam across the shop doorway. But there is a simple sensor that can detect the warmth of a human body. This now gives security lights their cue to switch on.
Search the web for ‘passive infrared sensor’
As well as technical details, you will see the strange nature of commerce. Many firms can sell security sensors for hundreds of dollars, even though they are based around the same sensor that is part of a ten dollar ‘special buy’ torch.
The sensor is high-pass filtered, so that the circuitry only reacts to change. A plastic lens is placed in front of the sensor that focuses ‘stripes’ of the scene into it, with gaps in between. As the intruder moves across these stripes, the signal toggles and the intruder is sensed.
The intruder might instead be a resident, enjoying a light that turns on automatically when a room is entered. The particular ‘special buy’ torch that I am thinking of is mounted in a charging socket, and its light turns on when you walk past it. But it also turns on if the mains supply fails, addressing another fear that will grow when power cuts become more common.
“I wish that I did not have to remember that password.”
That wish has been addressed by face and fingerprint recognition, though I have to doubt the level if their security. The sensors that you need for these are already built into the laptop or mobile phone, and to exploit them for your own application you need only find out how to access the data. That is explained in other modules.
“I wish that I could dictate my assignment.”
Yet another wish fulfilled. The machine digitises the signal from its built-in microphone, then uses extensive software to analyse it. Now you only need to find the link into the data.
To look a bit more deeply at the nuts and bolts of audio, load and try out Audacity, https://www.audacityteam.org/download/windows/ .
Getting the machine to read your text back to you is a simpler matter entirely. In 1975 my Cambridge team demonstrated primitive voice output, in an open-day demonstration on a machine that only had sixteen kilobytes of memory.
Today you can perform voice output with something as simple as a Micro:bit. Maybe you can devise a simple app to answer the wish:
“I wish I could remember what I came upstairs for.”
If you can master Bluetooth, you will have access to a host of sensors already in the purchaser’s gadget collection.
“I wish that I could get fit.”
There are plenty of applications for smart watches already. With information about blood oxygen, heart rate, number of steps taken, time spent sitting and sleeping, plus data entered by the user, you might be able to think of much more.
If you have a pair of Micro:bits, perform the experiment described in Microbit Bluetooth and read about later developments.
There is more on smart watch sensors in the next section.
A recent headline made me wonder if this will be a dream come true, or a nightmare. It will take some time to find out.
Boeing backed Wisk to bring self flying taxis to Brisbane for 2032 Olympics | The Australian
◄ ►