How to build a $5 Arduino clone
hwstartup.wordpress.com
hwstartup.wordpress.com
Once you've prototyped using an Arduino you can then go select the minimal set of components you need using the smallest appropriate ATMEGA for your application and get it under $5.
author: please proofread the post... i'm no Grammar Nazi, but the plethora of errors in that post made me cringe...
You'll need to do some soldering.
http://dx.com/p/arduino-pro-mini-microcontroller-circuit-boa...
The great thing about dx is that they will ship a single $9 board to your house for free :)
Ordering from Deal Extreme is somewhat fun and interesting because of this though because by the time I receive an item I forgot I ordered it, so it is like an awesome surprise gift I sent in the mail to my future self.
I buy from sparkfun because: - They provide eagle files and data sheets for most things they carry - They provide real time customer support in IRC - They don't screw up orders or take 6 weeks to ship - They test, use, continuously improve, and deprecate their own stuff
I know it doesn't apply in this case, but go try to find board files, schematics, or datasheets for most of this stuff on dx. There's a reason sparkfun is so popular.
For people who don't know much about hardware, that kind of customer service is invaluable. Without it, I could have spent days trying to figure out what was wrong, only to give up in frustration. Instead, I got quick confirmation that there was something wrong with the board and was immediately sent a replacement (at no extra cost).
So many chip manufacturers will send a few parts to whoever wants to order them.
Atmel sample store: https://secure.atmel.com/forms/secure/verifylogin.aspx?targe...
For students on a budget, or anyone interested in electronics, you should be ordering free samples.
May latest score is two ADXL362 chips, normally sell for ~$9 each in qty 1. I got them shipped express from asia, totally free.
I've gotten tons of microcontroller chips, small graphic LCDs (as well as the associated touch-screens), even a laptop touchpad, all types of LEDs, and many other components. My workbench would make most EE labs blush, and a lot of it was totally FREE, including shipping.
Back a few years ago, you could buy very cheap microcontrollers, then with a few $ of components make a simple programmer from the PCs parallel port or serial. http://picpgm.picprojects.net/hardware.html for a nice collection of them.
These where great, they cost about £1.50 each iirc, and I really liked them, by the time I went to uni I knew my Microchip PIC MASM by heart.
However people who are just learning about programming electronics don't really want to have to figure out the joys of GPIOA3 being shared with the ADC when trying to attach an LED.
Arduino providing a nice simple formfactor for allowing people to simply plonk them together, no worries about which pins can do what, how much can be driven. All the packages made it quite clear what could play nicely together.
This doesn't change that. Nor does it mention that if your the kind of geek who is thinking of buying say 3 or more of them you'd be much better off with oldschool prototyping board, a vertible smorgasboard of uCs and a simple programmer. There are some great PIC ones that even have the ICD (In Circuit Debugging!) on ebay for next to nothing.
The other thing I like a lot (even more I think) is that there is such a great community with many, many examples, so you don't have to dive into the bit level to get a sensor or module working.
But. I do think that when someone is finding they want more boards, they are probably envolved enough in it to go naked, and look at the uC world.
A friend wanted some help on their project, they had never considered using the 8pin uCs available for 50p each, they where trying to tie together Ardiunos and a lot of wire. Luckily they asked for help rather than just throwing money at it and using what they know. But I fear if people just stay in a LegoLand they might not realise that there are other things often more appropriate. I do still love LegoLand mind, despite not being there for a long time.
The Arduino and similar products (microcontrollers right down to PICs) seem interesting and appealing, but I'd much rather approach the issue from an EE perspective, and the thought of learning to do this like in the article - with the chip itself on a breadboard, seems much more appealing.
It's just a very intimidating space, and I have no idea how to approach it iteratively without jumping straight in to something like the Arduino, which presumably hides a lot of the complexity. I've looked at Sparkfun and would love to understand enough to start using some of the cool looking components on it.
Do you have any insight or suggestions for a path to learning this?
You can see all the pieces needed to make an 'Arduino'. A voltage regulator, a reset switch with a pull up resistor, some filtering capacitors, and a crystal oscillator. That's it. If you make a board yourself that runs off of a Lithium battery, you don't need a voltage regulator, might not need a reset switch, you can use the internal oscillators, and you have almost nothing outside of the microcontroller. It's really amazing.
I don't think you meant 'EE perspective' this way, but if you meant getting closer to the machine code, one thing you can do and still be able to do everything nice is rip open the Arduino source code, or re-implement the functions they provide. Try rewriting digitalWrite(PIN, STATE), or analogRead(PIN), and see how many instructions those functions need.
However, I - and many of my peers in CS-land - don't have the EE knowledge to correctly put together a analog circuit. Particularly as so many of the components for interesting stuff have different voltages... understanding how to link this stuff is really hard!
However, an Arduino abstracts that away. The 'hard' stuff has been done, for the cost of being inefficient and expensive.
I spent about 6 hours with a circuit simulator in the last few weeks trying to design a simple LED based circuit relating to a some 555 timers. Totally failed. I simply got exposed to my own ignorance even more! I got the corresponding Arduino thingie working in about an hour.
[1] http://arduino.cc/en/Main/Standalone
[2] https://www.google.com/search?q=minimal+breadboard+arduino
EDIT: judgmental statement removed, as it was unnecessary
http://aeturnalus.com/robotics/poor-mans-breadboard-arduino/
... which, in turn, reminded me of a similar guide which was even more explicit about wiring in the 5V power supply:
http://log.liminastudio.com/itp/physical-computing/breadboar...
To me, the parent post just smacks of blogspam (mostly links to high rank domains and content developed by others).
The the OP: As a matter of constructive criticism, you speak about how much you've gained from the open-source hardware community, but the hardware you sell, which by your own definition is based on this, seems to be closed-source. Your blog, thus-far, appears to offer little new content and seems to re-create common content from others. It would be most helpful for the community if you would explore what contributions you could make that would, perhaps, be new information for many. Here's a suggestion or two: consider publishing the hardware schematics, and talking about your "wireless communication protocol" between the Grow nodes and the host station. Talk about what the software looks like, what challenges you faced, and how you overcame them.
... just a suggestion.
That hasn't been the point.
I can go buy a nice, stackable hardware platform with everything Open and just not worry about it. The bootcode has already been written, and the IDE is short and sweet. Everything "Just Works". And then, we can discuss the crazy good amount of modules that have extended the Arduino IDE (and associated hardware).
It's well worth $30 for a rapid prototyping and design. Then, once I get a set design, I can go the chip+regulator+resistors route. And then I can tear the breadboarded setup on my Arduino and use it for another project testing!
I found this article easier to read than the one on Arduino's wiki. It's basically the same information, but with better explanations and pictures.
Also take a look at the parallax propeller: http://www.parallax.com/propeller/
Both are cheap (sub $10 per chip), tiny, and easy to use.
You'll have to use strict C and use breakouts and your own breadboard/components, but it's cheap enough to be disposable for hobbyists.
Edit:When talking about TI, i thought I'd share this subsite by TI (http://tideals.com/) for anyone who isn't savy on it yet. They don't have a deal right now though.