Small And Inexpensive MEMS Gravimeter

April 10, 2016 by Al Williams 18 Comments

A gravimeter, as the name suggests, measures gravity. These specialized accelerometers can find underground resources and measure volcanic activity. Unfortunately, traditional instruments are relatively large and expensive (nearly 20 pounds and $100,000). Of course, MEMS accelerometers are old hat, but none of them have been stable enough to be called gravimeters. Until now.

In a recent edition of Nature (pdf), researchers at the University of Glasgow have built a MEMS device that has the stability to work as a gravimeter. To demonstrate this, they used it to measure the tides over six days.

The device functions as a relative gravimeter. Essentially a tiny weight hangs from a tiny spring, and the device measures the pull of gravity on the spring. The design of the Glasgow device has a low resonate frequency (2.3 Hz).

Small and inexpensive devices could monitor volcanoes or fly on drones to find tunnels or buried oil and gas (a job currently done by low altitude aircraft). We’ve covered MEMS accelerometers before, although not at this stability level. We’ve even seen an explanation from the Engineer Guy.

Open Sesame, From A Galaxy Far, Far Away.

February 14, 2016 by Anool Mahidharia 15 Comments

[TVMiller]’s description of his project is epic enough to deserve a literal copy-paste (something our readers often praise us about). In his own words, “Having discovered several spare Midichlorians in my liquor cabinet, I trained and applied them to opening a large cumbersome gate. The FORCE motion travels through my inner what-nots and is translated by the Pebble Classic accelerometer toggling a command sent to the (Particle) Cloud (City) which returns to the Particle Photon triggering a TIP120 to fire a button on an existing RF transceiver. May the ridiculous hand gestures be with you, always.” Thus was born the Gate Jedi , and you’ll need exactly 47 Midichlorians, and some other trivial parts, to build one.

The Pebble watch hooks up to his android smart phone. A Pebble ~~(android)~~ app sends the accelerometer data to the Particle (previously called Spark) cloud service. From there, the data is pushed to the Photon IoT board which runs a few lines of code. Output from the Photon turns on a TIP120 power transistor, which in turn triggers the existing RF trans receiver that opens the Gate.

This looks way cooler than the Light Sabre hacks. Check out the video of him summoning the Force. And if you’d like to do more, try integrating gesture controls with this Pebble Watch hack that turns it into a home automation controller.

Continue reading “Open Sesame, From A Galaxy Far, Far Away.” →

RaspiDrums Uses Expensive Sensors

September 4, 2015 by Bryan Cockfield 3 Comments

Piezoelectric sensors are great for monitoring mechanical impacts with a microcontroller. Whether you’re monitoring knocks on a door or watching a heartbeat, they are a cheap way to get the job done. They do have their downsides, though, so when [Jeremy] wanted to build an electronic drum set, he decided to use more expensive accelerometers to measure the percussive impacts instead.

Even though piezo sensors are cheap, they require a lot of work to get them working properly. The ADXL377 3-axis accelerometer that [Jeremy] found requires much less work, plus provides more reliable data due to a 1kHz low-pass filter at the output. In his setup, a Raspberry Pi handles all of the heavy lifting. An ADC on each drum sends data about each impact of the drum, and the Raspberry Pi outputs sound via the native Alsa driver and a USB sound card.

This project goes a long way to show how much simpler a project like this is once you find the right hardware for the job. [Jeremy]’s new electronic drums are very well documented as well if you are curious about using accelerometers on your newest project rather than piezo sensors. And, if you’re into drums be sure to see how you can have drums anywhere, or how you can build your own logic drums.

Continue reading “RaspiDrums Uses Expensive Sensors” →

Hackaday Prize Entry: Project Dekoboko 凸凹 Maps Bumpy Roads On A Bike

August 17, 2015 by Richard Baguley 14 Comments

If you live in New England (like me) you know that the roads take a pounding in the winter. Combine this with haphazard maintenance and you get a recipe for biking disaster: bumpy, potholed roads that can send you flying over the handlebars. Project Dekoboko 凸凹 aims to help a little with this, by helping you map and avoid the bumpiest roads and could be a godsend in this area.

The 2015 Hackaday Prize entry from [Benjamin Shih], [Daniel Rojas], and [Maxim Lapis] is a device that clips onto your bike and maps how bumpy the ride is as you pedal around. It does this by measuring the vibration of the bike frame with an accelerometer. Combine this with a GPS log and you get a map of the quality of the roads that helps you plan a smooth ride, or which could help the city figure out which roads need fixing the most.

The project is currently on its third version, built around an Arduino, Adafruit Ultimate GPS Logger shield, and a protoboard that holds the accelerometer (an Analog ADXL345). The team has also set up a first version of their web site, which contains live data from a few trips around Berlin. This does show one of the issues they will need to figure out, though: the GPS data has them widely veering off the road, which means that the data was slightly off, or they were cycling through buildings on the Prinzenstrasse, including a house music club. I’ll assume that it was the GPS being inaccurate and not them stopping for a rave, but they will need to figure out ways to tie this data down to a specific street before they can start really analyzing it. Google Maps does offer a way to do this, but it is not always accurate, especially on city streets. Still, the project has made good progress and could be useful for those who are looking for a smooth ride around town.

Continue reading “Hackaday Prize Entry: Project Dekoboko 凸凹 Maps Bumpy Roads On A Bike” →

Hand Controlled Robot Uses Accelerometer

August 16, 2015 by Al Williams 3 Comments

What do orchestra conductors, wizards, and Leap controller users have in common? They all control things by just waving their hands. [Saddam] must have wanted the same effect, so he created a robot that he controls over wireless using hand gestures.

An accelerometer reads hand motions and sends them via an RF module to an Arduino. This is a bit of a trick, because the device produces an analog value and [Saddam] uses some comparators to digitize the signal for the RF transmitter. There is no Arduino or other CPU on the transmit side (other than whatever is in the RF module).

Continue reading “Hand Controlled Robot Uses Accelerometer” →

It’s Time To Roll Your Own Smartwatch

August 12, 2015 by Kristina Panos 29 Comments

Giant wristwatches are so hot right now. This is a good thing, because it means they’re available at many price points. Aim just low enough on the scale and you can have a pre-constructed chassis for building your own smartwatch. That’s exactly what [benhur] did, combining a GY-87 10-DOF module, an I²C OLED display, and an Arduino Pro Mini.

The watch uses one button to cycle through its different modes. Date and time are up first, naturally. The next screen shows the current temperature, altitude, and barometric pressure. Compass mode is after that, and then a readout showing your step count and kilocalories burned.

In previous iterations, the watch communicated over Bluetooth to Windows Phone, but it drew too much power. With each new hardware rev, [benhur] made significant strides in battery life, going from one hour to fourteen to a full twenty-fours.

Take the full tour of [benhur]’s smartwatch after the break. He’s open to ideas for the next generation, so share your insight with him in the comments. We’d like to see some kind of feedback system that tells us when we’ve been pounding away at the Model M for too long. Continue reading “It’s Time To Roll Your Own Smartwatch” →

IoT Enabled Thomas The Tank Engine

June 2, 2015 by Anool Mahidharia 6 Comments

This month the popular “Thomas the Tank Engine” toy celebrated its 70 anniversary. As a fun project, [tinkermax] wanted to bring this traditional toy into the age of IoT, while preserving its physical appearance and simple charm.

He used a model called the “Diesel” which seemed big enough to house the electronics, but proved otherwise once he inspected the innards. He needed to fit in an ESP8266 module, an accelerometer breakout, some discrete parts, a nifty analog multiplexer, and a 14500 3.7V LiPo. Once done, he was able to control its speed remotely over WiFi, with an auto “throttle-boost” that kicks in when the accelerometer senses that the train is going uphill, and has remote monitoring of battery state, engine load, inclination and track vibration – all in real-time using MQTT over WiFi. It’s quite a demonstration of the power of these super-cheap WiFi modules that are powering the current wave of IoT innovation.

The train motor works off a single 1.5V battery, so [tinkermax] tried a couple of boost converters to get the ESP-12 to work. But the modules were a tad bigger, and couldn’t provide the high peak current needed by the ESP-12. So he used a 14500 3.7V LiPo battery instead. A series diode drops the LiPo voltage to a circuit friendly 2.9V ~ 3.6V range. The ADXL345 accelerometer is used to measure “pitch” to detect going up and down a hill, “roll” to check for tilt or tip over and vibration to identify track defects. It communicates with the ESP-12 using a special Lite-SPI library that he wrote.

Two analog measurements are performed. One uses a resistor in series with the PWM driven motor to measure its current, with a low pass filter to smooth out PWM noise. The other is a resistor divider network used to monitor battery voltage. But the ESP-12 has just one ADC channel. Instead of adding another ADC module, [tinkermax] used a neat device – the FSA3157 – which allows two analog inputs to be channeled to a single output much like a SPDT switch. One PWM output is used to control motor speed and a second one to pulse a LED.

The sensor data is streamed 5 times a second over the MQTT protocol to a Raspberry Pi based MQTT broker. Finally, a JavaScript webpage receives the MQTT messages and plots the data graphically. One upgrade he would like to implement is speed measurement, to allow constant speed drive. If you have any ideas on how to extract that information from an accelerometer, chip in with your comments below. Check out his build log in the short video below. And if you’d like to see how all of this can be used in the real world, check this other video where [tinkermax]’s colleague gives a run down about a commercial enterprise IoT cloud platform hooked up to Thomas the Tank Engine.

Continue reading “IoT Enabled Thomas The Tank Engine” →