Get yourself connected: GrovePi+ Starter Kit When it comes to hacking hardware, there’s an easy way and there‘s a hard way. The hard way involves connecting peripherals direct to one of the standard buses supported by your Arduino, Raspberry Pi, Beaglebone or whatever. Buses like I²C, SPI, UART and 1-Wire. You’ll need to take care with your wiring: have you got the right …
Not necessarily - it might be that they'll do the electronics later, or are doing it in a different class.
In this class you learn how to control the stuff you are learning how to build over there - that's not a bad idea, the proof of your electronics is instantly available - you can see where both classes are heading...
Not everyone has to specialise in everything...
[Dodgy analogy alert]
Some people are happy to make a model out of Lego, others want to get a bit closer to reality with Meccano. And a few will start from first principles. They each have their value.
Personally speaking I'm planning to rebuild my daughters Lego Friends Dolphin cruiser as HMS Hood.
Surely the analogy is to software: a one-off kluge, versus properly re-usable code. This is properly re-usable hardware.
As we all know, the one-off kluge has its place, if it's really one-off, and if the cost of writing it to be general and re-usable is significant extra programming hours. Here. the cost of doing the hardware right is, er, about £40. An absolute steal, I'd say. Provided it works (see review).
It may also save you (say) £30 every time you accidentally fry a tiny little interface card costing £3 rather than a whole RPi costing £33. Kids (even more so than adults) learn by making mistakes, you'd best be able to afford them doing so.
Not necessarily - it might be that they'll do the electronics later, or are doing it in a different class.
Not in school. People seem to massively overestimate the standard of anything taught in school. Admitted this is 20 years ago but when I was there calculating using Ohm's law was at level 9 on the curriculum - i.e. the standard expected of an 'A' student at GCSE. In that context even simple things such as determining the value of a base resistor (part research, part calculation, part judgement) are completely off the menu.
You can blame that squarely on the National Curriculum - if everyone has to cover everything the coverage becomes so wafer-thin that it is of little practical use. When what became CDT and then D&T was half a dozen separate subjects, e.g. woodwork, metalwork, needlework, cooking, tech drawing and systems - of which you might do perhaps two - you could develop skills to a reasonable level. Now it seems the entire subject is classroom based and the pupils never pick up so much as a junior hacksaw.
Not that that is bad in and of itself - I remember when I was at school logic circuits were on a ready made board and wired with banana plugs. That is enough to teach the principles of what is being taught even if not component level design. It was only at A level electronics the breadboards and soldering irons came out.
they still don't even teach the fundamentals of AC theory either.
I blame that Thomas Edison chap. Electrocuting elephants? Who could stoop so low?
(probably the guys that kill them with LSD, if nobody minds me answering my own rhetorical question)
I think this is a great kit for someone who just wants to get the end result achieved. It maybe do lots of stuff for you, but the source code will be there if you want to learn the details. Just that fact that easier and quicker means more people will get something working and then are more likely to try something else a little more complex, and take they path more interesting to them.
> I think this is a great kit for someone who just wants to get the end result achieved.
The difficulty is that it doesn't work.
I experienced a number of IO errors, all related to I²C communications, during the time I spent with the Kit. Dexter Industries Forum posts relating to the GrovePI+’s predecessor suggest this is not uncommon but at least can be caught and managed in the core Python code.
If the kit has a buggy I²C implementation (or the software running on the Pi does) then it's of little use to non-technical people who's abilities limit them to "plug and play" components. Not only would those people not be able to produce a working outcome, but they wouldn't have the skills to diagnose or fix the problems, themselves.
And for the more advanced user: would the time needed to work through or around those bugs, or to work within the limitations imposed by them, be worth the convenience of buying this kit - especially when there are debugged hardware alternatives with known quality software available from other sources? Or even (perish the thought) Arduino solutions that have a proven track record.
Maybe it would be better to wait for version 2.0
Its predecessor had the same I2C communication problems. They should have fixed it in this version (the "+"). I wonder if the problems have anything to do with the bit the author suggests you might want to skip: have you got the right pull-up or pull-down resistor? Is there too much capacitance in the line? Or perhaps cross-talk between clock and data? Those cables are pretty long parallel lines, and they don't run the ground between the clock and data as they're supposed to (for lines longer than 10cm, according to the I2C spec). Longer cables will not only allow more cross-talk, but also increase the capacitance on each line. Either one of those could result in I/O errors.
OTOH, there could be lower level driver issues. I once worked with a serial I/O library that would timeout after 10 seconds if you didn't first wait a half a second or so before reading after writing, even though the message read didn't come for several seconds after setting up the read.
...set of modules for it, I'm particularly pleased to see the 'electricity detector'. There's no shortage of devices to allow me to turn a mains powered device on or off but very few easy ways to get a pi to react to a mains powered event. E.g. when appliance X turns on then turn on appliances Y & Z as well. This is much neater than a relay.
A few random musings.
* It slightly odd that the I²C buses and Analogue ports are numbered from 0, but the digital ports are numbered from 1.
* I hope the power pin-outs on each port are the same, across the port types, so that if you connect a device to the wrong type of port, it doesn't work, but isn't damaged.
* I²C is a bus system, each device has (from memory) a 8 bit address, which is 5 bit class and 3 bit device. Therefore there can be 4 digital ports, with 8 bit output each. Thus, if the LEDs had HEX dip selectors on them, there could be upto 32 LEDs.
* The I²C ports should allow daisy-chaining, or star-topology, subject to power constraints.
All-in-all an useful interesting module. As for it not teaching people how to access a device, why run before you can walk? Adding a UART directly to the data/address bus is the classic way to do it, but is easy to destroy both the CPU, the UART, and anything else on the buses if things go wrong. Anyway, don't modern peripherals for embedded systems come with a SDK, or at least sample code, rather than just a data sheet?
> * It slightly odd that the I²C buses and Analogue ports are numbered from 0, but the digital ports are numbered from 1.
There are two types of people:
1) Those who think that arrays should be numbered from one
1) Those who think that arrays should be numbered from zero.
A mate of mine had a Phillips Electronics building kit, where discrete components were held down onto a "bread board" with tiny coil springs. I think it made a light activated buzzer, a radio and a few other things. How things have changed..
Anything that allows kids to tinker is to be applauded, a really good teacher will spot when someone wants to go beyond the kit into DIY by suggesting "well maybe we could make one". Real enthusiasm and success for a discipline comes from guided self motivation
I had one of those Philips Electronic Engineers kits. While most of the competition used modules that plugged in, the Philips was just a means of connecting together standard electronics components. This made it easy to go 'off-piste' and do your own thing years before the solderless breadboard, as we now know it, existed.
The GrovePi+ kit reminds me more of those modular electronic kits like the Tandy/Radio Shack ones. For about half the money you can get the Philips type equivalent from Amazon including LCD display, sensors, buzzers, LEDs, servomotors, connecting wires and one of those modern solderless breadboards. It might not be as neat or quick as the GrovePi+ but I reckon the educational value and expansion potential more than makes up for this.
I had two of the Philips kits.
The earlier one used geranibum transistors, AC126 AF116, and the usual passives including a light-dependent resistor.
The later version moved to silicone, with plastic-encapsulated bc147 and bc149 mounted on 1"ish square PCBs to make them easily connectable-to with the spring things.
Somewhere in the house was an OC71 with the paint scraped off, to make it photosensitive.
Happy days. Thank you Philips (and probably Mullard).
Now in 2015, to make your Raspberry Pi 2 photosensitive, all you need is a Xenon flash. No need to scrape the paint off.
I had the 'Science Fair 150 in 1 Electronic Project Kit' from Tandy back in the late 70s, built all sorts, including a lot of things that I just made up myself. Happy days.
Later went on to become an Electronics engineer for around 10 years, although in IT these days (technical architect).
That brings back memories!
My sister worked for Philips and received a staff discount, so *everything* electrical in our house was made by them.
I had both the Electronic and Mechanical Engineers kits*. I built a light actuated car, amplifiers and all sorts of stuff. I also had Philips first portable cassette recorder, rigged to play stereo.
Let's hope that a new generation of engineers will develop from kits like the GrovePi.
*I just had to Google these.Lo and behold, there's the car that I built! <rings nursing home>
http://www.girdersandgears.com/norelco.html
http://www.radiomuseum.org/r/philips_ee_20_ee20.html
Bought a little OLED off eBay, found the data sheets, wrote a driver. Even wrote a program to clone the display to a 128x64 mono display. Useless but fun.
I have time on my hands and I think that I learn/understand more when I do it myself.
Still, this board looks like it exposes some good stuff to interface with.
This post has been deleted by its author
My electronics career started at 13 with a transistor radio kit bought from a newspaper advert. The wires were connected by 8ba nuts and bolts into a free-standing rat's nest. Several years later the audio was boosted for a loudspeaker with an expensive set of matched OC72 and two miniature transformers - still a rat's nest.
In parallel valves were used for receivers and transmitters - with one Christmas spent building a Heathkit oscilloscope.
Then it was time to etch my own circuit boards with shellac resist. UHF required GM0290 with very short leads.
Veroboard was useful for digital work like using very discrete TTL ICs to build a digital counter - but the desired Nixie tubes were unaffordable.
Skip through the Apple II peripherals and other micro building years.
The peak 25 years on was a PC ISA prototyping board that was wire-wrapped by hand - with two Xilinx 2018 FPGA devices. Worked first time - with only a home-made digital probe for hardware debugging.
Nowadays the soldering iron rarely gets used for much more than a relay - and development is concentrated on software.
There's no limit. The code is open source, the interfaces on the board are simple standards. As well as the specific modules there will be people building their own thing and connecting it up to this.
Nearly every input or output device you can think of can be found with an I2C interface. And python allows people to slice and dice the data really simply.
Look at the teensy board board being connected to different peripherals using a breadboard
https://www.pjrc.com/teensy/
Got one for about £80 from Amazon last year, should probably have spent more on it to get higher speed...
Pro: It works OK, pretty much as described, fair enough value/quality
Con: Not a huge advance over software-on-standard-PC-audio-card for audio stuff
Windows only at this price point - but TinyXP in a VM under Linux does the job after some fiddling.
When I get round to it (hah!) plan is to dedicate an ancient-but-repairable low end XP laptop to become the dedicated 'scope box. Now where did I put that laptop?
Bottom line - haven't used it much for a few months now, but have projects in hand that will surely need it.
What's the manufacturer? Model number?
Darned if I can remember - and the lab is like a tip right now. Maybe tomorrow.
I've been retired all this century, age is advancing, usual excuses....
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
