Need advice on Sensors for use with hyperinstrument
I have recently purchased the Pocket Electronics for use with a hyperinstrument that will serve as a portion of my masters thesis. The sensors that i have been looking at (i-cube) are quite expensive, so i was wondering if anyone knows of an less expensive version of these sensors:
Touchmicro v1.0 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/73)
Light v1.2 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/51)
GForce2D&Tilt2D v1.1 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/65)
The main thing is that the voltage output of the sensors must be between 0….5+
If i just needed one of each sensor, then the i-cube products would be fine, however i need 8 of the Touchmicro, one light and one GForce2D&Tilt2D.
Any suggestions? Thanks for the replies in advance.
The cost of the i-cube sensors is mostly the work that goes into assembling them. If you solder, then you can make these things much cheaper.
> Touchmicro v1.0 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/73)
This sensor uses a Force Sensing Resistor from Interlink Electronics. The design kit will get you a number of them in different sizes.
Here’s another company:
And their website has circuit diagrams.
> Light v1.2 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/51)
This sensor can be had for three bucks at Radioshack.
It is also a variable resistor (like the FSR), and will work in essentially the same circuit.
> GForce2D&Tilt2D v1.1 (http://infusionsystems.com/catalog/product_info.php/cPath/2 4/products_id/65)
This is most likely an ADXL202 from Analog devices. Building this circuit is a little more involved. It is used very frequently, so Google will tell you more than you every wanted to know.
It might be worth buying the i-cube one. Or, get the Eowave one from Cycling ’74 for $78. (I got mine from Phidgets, but they don’t seem to sell that anymore.)
A great book is "Physical Computing" by Igoe and O’Sullivan.
While it spends a lot of time dealing with microcontrollers, the sensor sections will be helpful, along with general electronics tips.
Thanks for the step-by-step help mzed. I grew up in a family of electricians and brick masons, have wired gauges in my ’66 Pontiac, etc. so i think i can handle the soldering/wireing for this project. At least i can get help fairly easy if needed.
I figured the iCube stuff was expensive because of assembly, but whenever i tried to do a google search for the sensors i always ended up finding exactly what i did not want.
Thanks again for the info. Completely thorough and helpful!
Yes, the cost of these sensors is largely the time to produce, test, and package them. Also, when you buy sensors from someone like Infusion Systems or Electrotap, they are built on custom printed circuit boards which improves performance and reliability over hand soldered boards.
I do not think that the Infusion Systems accelerometer is an ADXL202 – the 202 outputs a pulse-width modulated square wave, which would then need additional filtering and treatment before being fed into a continuous voltage input. I’m not sure what they are using, but a better match would be the ADXL311 – if you have the stomach for the cockamamie surface mount package that it comes in. You might want to buy 2 or 3 in case you screw one of them up by hand soldering it.
> I’m not sure what they are using, but a better match would be the
> ADXL311 – if you have the stomach for the cockamamie surface mount
> package that it comes in.
I second that chip
here’s a pic so you get an idea how small it is:
next to a standard DIP chip
(but don’t get confused that it has the same number, the LM311 is just
minimum external components is three caps and a resistor which fit
neatly at the bottom
> I do not think that the Infusion Systems accelerometer is an ADXL202 –
> the 202 outputs a pulse-width modulated square wave, which would then
> need additional filtering and treatment before being fed into a
> continuous voltage input.
the gluion, however, has digital inputs that can read these PWM signals
directly and measure the pulse width with a high-res counter. The
advantage is that you can run much longer wires w/o signal degradation.
I think i am going to bypass the ADXL311 and the ADXL202 for the Eowave GForce sensor, just for simplicities sake/less work/less chance for me to screw up. It does seem that the ADXL320 would be a better fit than the other two though (wider output range to match the doepfer voltage range of 0->5+). But then again, the wiring for the ADXL320 may be a bit out of my league. It has been a while since i have read wiring schematics, so i will go home and take a closer look before i decide.
thanks again for the replies
FWIW, The ADXL320 is what is used in the Electrotap Accelerometer. You are correct that it is not possible to solder this one by hand.
Check out this resource for sensor-related things: