Pages

Tampilkan postingan dengan label meeting. Tampilkan semua postingan
Tampilkan postingan dengan label meeting. Tampilkan semua postingan

Senin, 20 Juni 2016

Humboldt Microcontrollers Group July 24 2014 Meeting Notes

Redwood Curtain Brewing Company
This is a summary of the July 24 meeting of the Humboldt Microcontrollers Group at The Link in Arcata, California, behind the Redwood Curtain. Speaking of the curtain, maybe the Humboldt Microcontrollers Group should schedule an upcoming meeting to be held at Redwood Curtain Brewing Company. Microcontrollers can be used in brewing...

One of the reasons for this blog and for the formation of the Humboldt Microcontrollers Group is the redwood curtain, which is both an actual and perceived barrier between Humboldt County and the rest of the world. In many ways, that curtain is a much appreciated and highly beneficial barrier, preserving natural beauty, creating a slower pace of life, and leading to a better balance between work and non-work activities. But in other ways, that barrier brings economic challenges and limits both the job opportunities for people and growth opportunities for
The curtain between Humboldt and the rest of the world
companies. One way to reduce the redwood curtain barrier for people and companies in this region who work with microcontrollers (MCUs) is to connect, strengthen, and promote the Humboldt microcontroller community. This blog and the Humboldt Microcontrollers Group are both working to connect, strengthen and promote that community.

So, the main topic of the July 24 meeting of the MCU group was the #8 Jeremy Blum Arduino video tutorial (see also my previous post about that tutorial). The focus of the #8 tutorial is SPI (Serial Peripheral Interface), and Jeremy uses an AD5204 digital potentiometer to demonstrate the use of SPI. Ed Smith used a different digital potentiometer than the hard-to-find AD5204 when he did the #8 tutorial exercise, so the sketch Jeremy wrote in the tutorial doesnt work correctly because of the differences in the two potentiometers. Well have to figure out an appropriate alternative potentiometer and accompanying code for other Humboldt people who want to do the #8 tutorial exercise. One more item to add to the list of things-to-do regarding developing a set of effective learning tools for people in this region interested in MCUs.

No one at the meeting had actually worked through the exercise in tonights video. Since none of the people at the meeting tonight had done the exercise, we didnt spend too much time talking about either the technology of a digipot or about interesting or unique aspects of SPI. Maybe at a future meeting? My plan is to talk with Ed, Nick and anyone else interested to agree on a good alternative digital potentiometer and to modify or rewrite the Arduino sketch to work with the digipot we choose.

In addition to the #8 tutorial, the following topics were discussed at the meeting:
  • Somewhat related to the #8 tutorial, Bob asked about using a digital potentiometer (or other components / circuitry) to match color and intensity in a set of three LEDs that are simultaneously given the same input signal but seem to generate different colors. The LEDs are Cree PLCC4 (plastic leaded chip carrier) 3-in-1 SMD LEDs. Based on the little bit we could figure out during tonights meeting about how the LEDs are powered and controlled, Nick suggested the first step should be to put an oscilloscope on the lines going to each 3-in-1 LED to confirm theyre all seeing the same signal. Nick also said its likely the LEDs are adjusted by PWM (pulse width modulation) rather than by varying the power to them, so the digipot would probably not be a possible way to adjust the colors. If all three are seeing the same signal, it would appear unit variability in the LED color output may be the issue. The spec for the LEDs says "Tolerance of measurement of dominant wavelength is ±1 nm" so that spec appears pretty tight. I need to research the issue of bin codes and figure out how to determine if the 3-in-1 LEDs we purchased for this order have green and blue LEDs all from the same bin code. According to the datasheet, the red LEDs only have one bin code (which identifies the wavelength range for that LED), but the blue and green LEDs for this particular component each have seven bin codes. So if some blues were at the bottom of the wavelength range and others were at the top, that might result in a noticeable color difference even with the same applied PWM signal. Of course, keep in mind that my knowledge about PWM signals applied to 3-in-1 LEDs, or other types of RGB LEDs, or even single color LEDs is extremely limited. Researching this will definitely be a learning exercise.
  • HLH frame
  • An update about the Humboldt Laser Harp (HLH) project was given to the group by Nick and Ed. Nick has the frame of the HLH just about completed. In the upper picture on the right you can see the laser harp frame. If you look near the top of the lower picture on the right, a bit of a laser beam is visible -- first sighting of the infamous HLH laser beams. One outcome of tonights meeting is that Gordon will be working with Nick to design and build a base for the harp. Gordon also had an interesting suggestion for laser beam presence enhancement in brightly lighted areas. He suggested shooting the laser through a diffusing or side-emitting light pipe for part of the distance from the laser to the light sensor.
  • From a HLH programming standpoint, Ed has been crafting the code that will convince the Humboldt heavenly harp to generate many marvelous melodies. In version 1.0 of the HLH, the goal is to have a demonstration device that produces simple tunes and has a reasonable degree of casual interactivity with curious members of the general public. Future versions are envisioned that incorporate music-correlated lighting effects and more musical instrument capabilities, e.g. lots of simultaneous notes, note strength and fade effects, simulation of different musical instruments, allowing notes from several octaves, different type of controls for the musicians playing the HLH, etc. As with many other maker
    HLH laser beam
    topics, there are lots of people who have specialized in the field of DIY electronic music, and much of what well be doing in the early days of the Humboldt Electronic Light Orchestra will be research on whats already been done, connecting with people in the DIY electronic music field who are willing to share their knowledge or who are interested in what were doing, and starting to develop new (as far as we can tell) electronic music ideas of our own.
  • The plan is for Nick and Ed to get together this weekend, connect the hardware with the software, flip the switch, and see what happens on the HLH. Stay tuned for updates and videos...
  • We discussed the upcoming August 2 maker fair in Eureka, which has been organized by Origin Design Lab. Nick and Gordon will figure out if Gordon has any maker projects that would be good to have at the Humboldt Makers table. Shawn Dean from InPrinting will be at the table doing 3D printing demonstrations. It would be great if other Humboldt Makers can spend an hour or a couple hours at the maker fair, which sounds like it will primarily be a craft fair aimed at being a selling opportunity for local craftspeople. Even if you cant be at the Humboldt Makers table, if you have an item that will be interesting to the general public, please contact Nick Appelmans or me (my email is arcatabob (at) gmail {dott} com).
  • Justin suggested the Humboldt Makers might consider doing a minor repair on the 3D printer that InPrinting nearly finished building some months ago, then figure out what else needs to be done to get the unit printing. We can discuss that at the next meeting as either an activity to spend a couple Thursday meetings on, or a project that one or several people would work on as time becomes available.
Hope to see lots of people at the next Humboldt Microcontrollers Group meeting, which will be on August 7.

**********
Read More..

Minggu, 10 April 2016

Wireless Communication For Microcontrollers August 7 MCU Group Meeting

Traditionally, microcontrollers (MCUs) didnt need wireless communications because they did all their interactions in a limited space. If they interacted with something a few inches or a few feet away, wires connected the interacting devices. However, in 2014 some MCU applications require wireless communication, and this post takes an initial look at the wonderful world of wireless for MCUs.
Breadboard setup for #9 video wireless communication

This posts overview of MCU wireless looks at Jeremy Blums #9 Arduino video tutorial, the wireless exercise Ed Smith developed as his alternative to using the BOM components (bill of materials) and code in Jeremys #9 video exercise, and a few recent developments and web resources for MCU wireless communication. In addition to presenting an updated look at MCU wireless, researching and writing this post is a step toward locating or developing a training exercise for Arduino wireless communication that is more up to date than Jeremys #9 video, which was published in Febrary 2011.

Two of the primary driving forces behind MCU wireless communications are the internet and the increasing power of mobile or miniature computing devices. As the increasingly ubiquitous smartphone and the rise of the Internet of Things (IoT) create a sharp increase in the production volume and tech innovation in MCU wireless components, the number of wireless technology solutions continues to rise and the cost to do wireless communications continues to drop. If you search online for MCU wireless projects, tutorials or components, check the dates on your search results. Information from five or even two years ago may be totally outdated.

First up -- points of discussion about Jeremys #9 video at tonights Humboldt Microcontrollers Group meeting.

Most of the participants in tonights meeting had watched all or part of the #9 video. But no one had done the exact wireless exercise that Jeremy demonstrated and explained. The primary reason no one duplicated his exercise was that the components in Jeremys exercise cost too much to buy for learning about wireless. If you already had a couple spare Arduino Unos and the XBee components sitting around idle, then Jeremys exercise would be reasonable, but it didnt make sense to anyone at the meeting to buy all the components just to do the #9 video exercise. The main value of the #9 video seemed to be that it focused everyones attention on MCU wireless communication, and even if we didnt follow Jeremys exercise exactly, a couple people in the group did develop somewhat analogous wireless systems. And there was a lot of good discussion about the various wireless options and components, probably more than if everyone had used Jeremys components.

Nordic RF module
Second topic -- Ed Smiths MCU wireless communications components, circuitry and code.

Ed put together a low-cost wireless exercise using a couple Nordic RF transceivers he had available. The parts used in Eds setup include Texas Instruments Tiva-C launchpad 296-35760-ND, Texas Instruments MSP430G2553 Launchpad 296-27570-ND, Nordic 2.4 GHz RF transceiver NRF24L01+, Pin header 952-2247-ND, Pin socket header S7031-ND, perfboard, jumper wires, solderless breadboard, TFT LCD 2.2" 240x320 and an accelerometer. The breadboarded circuit for the MSP430 and accelerometer is shown in the picture below on the right. Eds hardware sent 3-axis information from the accelerometer from the MSP430 to the Tiva-C board and displayed it on the small LCD screen. The code was written using Energia, which is designed
MSP430 with Nordic RF 
"bring the Wiring and Arduino framework to the Texas Instruments MSP430 based LaunchPad". Ed will be writing up a separate blog post about his setup and the Nordic transceivers, including code and pictures, so look for that to show up on this blog soon.

An alternative to the wireless components Jeremy used in the #9 video and the components Ed used is the JeeNode. Nick got two JeeNodes for about $22 each with 915 MHz radios and worked with MCU wireless using those boards. According to Digital Smarties, the UK fulfillment transaction processors for the boards, the JeeNode is,
"a compact wireless board with an Atmel 8-bit RISC microprocessor. JeeNodes are compatible with the Arduino platform and can be programmed under Windows, Mac OS X, or Linux using sketches created with the Arduino IDE. Each JeeNode has 4
identical “ports” for attaching analog and digital I/O logic, I2C devices, and more. The JeeLib library makes it easy to interface to these ports. The JeeNode Kit contains all the parts needed to build a JeeNode v6 – including an ATmega328p processor and an RFM12B radio module."
Modern Device does much of the manufacturing for JeeLabs and is the US distributor for JeeNodes. The JeeLabs website describes the board this way:
"The JeeNode is a wireless micro-controller board designed for a variety of Physical Computing tasks. From measuring and reporting temperature, humidity, and other environmental data to tracking and controlling energy consumption around the house. It was inspired by the Arduino Duemilanove and Uno boards, and by the “Real Bare Bones Board” (RBBB) from Modern Device."
The JeeNode component bill of materials (BOM) for learning MCU wireless would only be slightly more expensive than the Nordic RF components. Ed said for various prototypes or applications, the JeeNode would be more capable than the Nordic systems, and would be well worth the few extra dollars.

Third subject -- Worthwhile web resources for newbies to understand MCU wireless.

I havent done an exhaustive search for web resources on MCU wireless, so please send me links to sites you feel are good on this topic, especially if you think theyre better for a particular aspect of wireless compared to the sites Ive listed. If youre new to wireless communications, you might want to read through the Wikipedia entry for Wireless Networks. If you are already have a good background on wireless networks, consider looking through the SparkFun Wireless Buying Guide. The SparkFun guide is where I read about Nordic RF modules yesterday for the first time ever. Then I found out what Ed used for his wireless exercise, and it turned out to be Nordic transceivers (or Nordic clones). The wireless section of your favorite MCU vendors website is another good place to learn about wireless, especially as it applies to the MCUs you use and understand. Atmel, Texas Instruments and Microchip main pages for wireless are listed below.
  • http://www.atmel.com/products/wireless/
  • http://www.ti.com/lsds/ti/wireless_connectivity/overview.page
  • http://www.microchip.com/pagehandler/en-us/products/wireless/home.html
If you go through the above resources and want to know more, search for your topics of interest and youll find lots of relevant websites to sift through for more specific info thats relevant to you.

Fourth issue -- Developing an inexpensive MCU wireless educational exercise for the Humboldt Microcontrollers Group.

We decided tonight that Eds setup is about as inexpensive as youre going to be able to do a two-MCU, two-transceiver wireless exercise. Another wireless learning exercise we will probably research is one with a single MCU and a Wi-Fi system to directly connect the MCU to the Internet. Figuring that one out will wait for another day.

Along with our discussions about MCU wireless, tonights Humboldt Microcontrollers Group meeting covered a multitude of topics, including but not limited to freezing hard drives to recover data from them, buying vacuum tubes, flying quadcopters that have a mind of their own, the unpredictable adventures of buying electronic components from Chinese eBay sellers, good power supplies and power supplies to stay far away from, how to load Arch Linux on a Chromebook, repairing welders, upgrades to automated doors on chicken coops, the Humboldt Laser Harp, and what the topic of the next MCU group meeting should be. If you have suggestions for MCU meeting topics, please contact me at arcatabob (at) gmail [dott] com or contact Nick A.

If you are interested or experienced in MCU wireless, it would be great to connect with you -- email me at arcatabob (at) gmail {dott} com. Hope to see you at the next Humboldt Microcontrollers Group meeting on Thursday, August 21.

**********
Read More..

Kamis, 07 April 2016

Solar Energy And Arduino Solar Charge Controller

Theres an intriguing microcontroller (MCU) project in an August 8 article on Design News titled "Gadget Freak Case #260: Arduino Solar Charge Controller." I havet read a lot about solar energy and Arduinos, or any other MCUs, so I took a closer look at the article.
Residential active solar energy system (from Wikipedia CC)

Many Humboldt County people seem interested in solar energy, although the often-foggy or cloudy climate of Arcata, Eureka, and other coastal communities isnt as well suited to solar energy collection as the climate of New Mexico or Arizona cities. However, it is still good to be experienced with technologies used outside the North Coast, and there are plenty parts of Humboldt County not immersed in marine fog banks or redwood forest mists. For this reason it would be useful for participants in the Humboldt Microcontrollers Group to know the basics of active solar energy systems and to have experience with the electronic components and operation of those systems. The topic of the Design News article above isnt an inexpensive basic active solar energy system, but I think Ill do some research regarding different types of inexpensive systems for a future group project.
Solar charge controller schematic from CirKit.com

The solar charge controller that is the topic of this post is located between the solar energy collection device and the energy storage system, usually a battery. The controller regulates both the voltage and the current going to the energy storage system from the energy collection device. The schematic at the right from CirKit.com shows the general concept (although that schematic is not from Gadget Freak project). The Design News article describes the function of the Arduino this way:
"This gadget uses Arduino to control the whole process and takes a voltage reading from the solar panel and the battery to be charged. Then, according to voltage levels on either side, it charges the battery using PWM control signal. Energy flow is driven with MOSFET transistors that ensure low energy loss. The charging controller is equipped with basic filters on both the battery side and the solar panel side. It is also equipped with things like overcurrent, overvoltage, PV panel reverse current, auto load disconnection, and overcharge protection."
Gadget Freak #260: solar charge controller (Design News PDF)
The solar charge controller build instruction PDF is linked to in the article, and seems like a good quality document. It has pictures, circuit drawings, Arduino code, and the type of helpful tips that many project documents do not include. An example of the helpful tips is where the project creator describes how to choose MOSFETs. For people who arent already familiar with MOSFET properties and how theyre used, he says this about drain source voltages for MOSFETs:
"When the MOSFET is turned off, the whole supply voltage will be measurable across it, so this rating should be larger than your supply voltage to provide sufficient protection so that the MOSFET does not fail. The maximum voltage a MOSFET can handle varies with temperature."
He also talks about whether to use an n or p channel MOSFET, continuous drain currents, thermal loss, gate threshold voltage and more. Tips like this are especially useful to people like me who are new to microcontrollers.

In months and years to come, it is my hope that the Humboldt Microcontrollers Group will build many MCU devices and systems. Nicks automated chicken coop door is the first one, as far as I know, and the Humboldt Laser Harp is the second, I think. It would be great if we can develop and put online as much information as possible about these MCU projects, assuming theyre not going to be commercialized so they can earn us millions. Or even thousands. Anyway, part of gathering, organizing and publishing this MCU project information should be a good documentation format. There are many possible formats, but the Gadget Freak build instruction PDF used in the solar charge controller project seems like a good place to start if someone in the MCU group doesnt have a different project information format they prefer.
Gadget Freak (from Design News and Allied Electronics)

This Gadget Freak topic appears to be a regular feature of the Design News website. This solar charge controller is labelled #260. I dont know if that means there are 259 previous Gadget Freaks, but here are links to The Best of Gadget Freak Volume 1 and The Best of Gadget Freak Volume 2. Design News collaborates with Allied Electric, and they invite you to submit your electronics projects for publication. This is just one more way that people in the Humboldt MCU community can participate in the expanding world of microcontrollers and possible earn a few bucks. The Gadget Freak page on Allied Electrics site says:
"Are you a Gadget Freak? Allied Electronics and Design News would like to send you a check for $500 to spend at alliedelec.com or anywhere you please! Submit your design for a gizmo or gadget that any Gadget Freak would appreciate, and you just might win! If your project is selected, you’ll receive a $500 check from Design News and will be featured in an upcoming issue of the magazine with your invention. In addition to the $500 awarded for being selected as a Gadget Freak, all selected gadgets...will be included in the Gadget Freak of the Year contest. Starting in November, the readers of Design News will then vote on the best gadget...The winning gadget will receive an additional $6,000 with two runners-up who will receive $2,000 each."
If youve built solar energy devices or systems with MCUs, it would be great if you came to one of the every other Thursday meetings the MCU group has. The next meeting will be on Thursday, August 24. If you cant make it to the meeting but would be interested in discussing your solar energy device or system, please email me at arcatabob (at) gmail {dott} com
. It would be great to meet you for coffee and a tech discussion, or possibly youd be willing to show the MCU group your MCU in operation!

**********
Read More..

Little Bit of Information About LittleBits

So Im pretty sure everyone reading this blog knows what the Internet of Things (IoT) is. Anyone who likes to use or read about microcontrollers (MCUs) has probably read a couple news stories about the IoT or at least about how some microcontroller is being used in the IoT. The topic of tonights blog is littleBits, which are now "an easy and open way to contribute to the Internet of Things without wasting time prototyping devices from scratch."

littleBits circuit with module snapped and wired together
In addition to that preceding description of littleBits, the July 23 Engadget article about these modular electronic marvels says:
"The magnetic, interconnecting circuit boards that make up LittleBits library of electronic modules make it easy to build all sorts of neat (and noisy) devices with almost no technical knowledge at all...today LittleBits is announcing the Cloud Bit, a new module that, as company CEO and founder Ayah Bdeir puts it, allows builders to "just add internet" to almost anything...More complex modules -- like the Arduino module and todays WiFi-enabled Cloud Bit -- raise the ceiling of complexity for potential LittleBits projects. Its her hope that the product will be used for rapid prototyping of new ideas in addition to being a fun hobby toy. "If you wanted to recreate a Nest or recreate a Sonos or a DropCam, you could. If you wanted to create the next billion dollar idea, you could do that..."
cloudBit
The Wired article announcing the cloudBit has this to say about the new component and about littleBits in general:
"The cloudBit, announced today, is a new component that’ll connect any littleBits creation to the internet, allowing it to be programmed through a simple web interface. There have been other kits like this before, powered by Raspberry Pi and Arduino—but while those remain powerful, flexible, and inter-operable, none of them possess littlBits’s plug-and-play simplicity. So with the addition of the cloudBit, littleBits could become a reasonable, DIY gateway to the Internet of Things...Rigging together motors, sensors, and noisemakers can be painstaking work, requiring lots of soldering and testing equipment to make sure all your circuits are done right. littleBits, instead, is a modular library comprising dozens of neat electronic components that simply snap together in a line, with magnets...To get started, users first sync their cloudBit to the web app. From there, the web app can link to any any web API, including IFTTT (a service that allows you to chain together websites with simple “If this, then that” logic). Users can then use those data streams to turn the cloudBit on or off, or activate any partial state in between. By linking the cloudBit to any number of other bits, users can create whatever chain of mechanical reactions they like...the littleBits ecosystem now has 59 different components, ranging from basic items like motors, a microphone, and a keyboard to a growing body of sensors that detect everything from light to bends to motion to pressure. (There’s also an Arduino component, for more serious hacking.)"
littleBits tagline is "the easiest and most extensive way to learn and prototype with electronics." Maybe its because Im an engineer and not a marketing person, but from my perspective, it seems unlikely for any product to be both the "easiest" and the "most extensive" way to learn and prototype with electronics. My guess is that littleBits is probably a lot closer to the easiest claim of their tagline than the most extensive claim.
littleBits Big Drone Synth 

In spite of their overreaching marketing department, which may be an apt description of the marketing department at pretty much every company, I do feel littleBits does have a strong appeal for non-technical people who want to incorporate technology into unique or interesting products that theyve created in their minds. If an artist, sociologist, psychologist, disaster worker or many other types of people with no desire to figure out how to breadboard an Arduino circuit can fairly quickly learn how to snap together a couple modules that accomplishes something meaningful to them, they will tend to use that product again in the future and start learning how to do more with that product and complementary products.

What this potential appeal for non-technical people says to me is that if littleBits can figure out kits that are useful to a large number of target users and successfully market those kits, it could very likely lead to an active, growing and loyal user community. By removing the Arduino barriers (perceived or real) of learning at least the basics of electronic design and programming basics, littleBits has given itself a much larger potential market. The challenge for littleBits now is figuring out what modules will appeal to and get buzz among non-tech early adopters.

littleBits DIY smartphone-enabled thermostat
I think it would be interesting for the Humboldt Microcontrollers Group to build a product with littleBit modules, then build an MCU-based product that has the same capabilities and features of the littleBit product. We could benchmark the operation of both products, compare the component and build costs for both products and keep track of the actual build time to go from loose components to a properly and reliably operating unit. The main thing stopping us from doing that is the lack of components. If I can identify clear benefits for sponsors, Ill put together a project proposal for doing that comparison and pitch the project to potential sponsors.

One side note of particular interest to me is the IFTTT aspect of littleBits. Ever since I read about if this then that, a couple years ago, Ive wanted to experiment with it. Doing a littleBits project would be a good excuse to set up some IFTTT recipes and figure out how much value the service has for me. Ill have to bring up IFTTT at the next MCU meeting to find out how much, if any, people at the meeting have done with it.

If youre reading this blog post and have used littleBits, please come to the next Humboldt Microcontrollers Group meeting, which will be from 6 to 8 PM on Thursday, August 7, to talk about your experience with them and to demonstrate how to use them.

**********
Read More..