I just hooked up a Sharp GP2D12 IR range sensor to a scope and saw that the output is very noisy. The sensor updates at 25Hz so I was thinking that if I were to add some kind of 25Hz low pass filter to the output that would clean things up.
Datasheet:
formatting link
The sensor also operates at 0-2.45v so a gain of 2 from the filter would be useful to bring it in line with the 5v range of my ADC.
I thought about this circuit
formatting link
But it is unity gain - could it be modified to give a higher gain?
You can use this page to design the analog filter:
formatting link
It allows you to set the gain to be whatever makes sense, and computes the right component values.
However, if you are interfacing with a microcontroller, then you might want to use a digital filter instead. You can easily average the output over time, and save on the external components. You still need an RC filter to eliminate high frequency noise, since a digital filter will alias frequencies higher than 1/2 the sampling frequency into the baseband if you don't filter them out. You can control the gain with a digital filter as well. There are websites that will design any kind of digital filter for you (i.e., a java app). They usually use floating point, though. For this application, I've used a simple RC filter, along with an averaging loop like this:
forever {
while (!sample_available()) { do_something_else(); }
average = (average >= 3; // do something with the average here }
This gives each sample an exponential decay.
If that simple filter doesn't work, there are lots of others to choose from. The choice of filter will depend on the application.
Ditto all of this, except that if most of the noise is coming from the detector, and if the detector is really updating at 25Hz, and if you can sample synchronously with that 25Hz update, then an anti-alias filter probably won't do much good.
--
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html
Hi, Dave. Your noise problem will certainly be reduced (and probably eliminated) with good layout practices and bypassing. You could do worse than starting out with a .01uF to .1uF ceramic cap in parallel with a 22uF electrolytic right across the power pins of the sensor.
The 0-2.5V analog output means you only lose one bit of resolution with your ADC. Might not be worth setting up a filter with a gain of two to bump it up. If you still have residual noise after bypassing, try a simple R-C filter to cut some of the higher frequency stuff:
You might check the archives of comp.robotics.misc. I recall several discussions about the amount of current that some (all?) of the Sharp IR rangers consumed.
I do know from personal experience that an IR remote control unit such as the Siemens SFH506 (admittedly a slightly different device) can go from "flaky" to "fantastic" when a 100Ohm/100uF filter is added to keep the device's voltage up.
Hope this helps...
Frank McKenney
-- Government is the great fiction, through which everybody endeavors to live at the expense of everybody else. -- Frederic Bastiat, French Economist
-- Frank McKenney, McKenney Associates Richmond, Virginia / (804) 320-4887 Munged E-mail: frank uscore mckenney ayut mined spring dawt cahm (y'all)
Ah, one other thing, those sharp sensors take a big current spike when they power the LED. Depending on your setup, you may actually be seeing the spikes on your A/D input.
The solution for this is a big cap (like 100uF) near the power inputs for the sensor, which will smooth the power supply.
Note that this only matters if you are using a small supply that can't keep up. Scope the power input to see if it makes a difference.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.