Post by thread:[OT]: Reflected IR problem

I'm trying to build an IR based object detector based on the 16F628. The prototype works fine just the way I want it to, but I have run into a problem in encasing the unit. Since it works on detecting IR reflected from an object, the transmitter LED and the receiver IR module are both to be mounted within the same case. To reduce the chances of the transmitted IR being directly incident on the receiver, I've put a piece of heat shrinkable tube on the LED, exposing just the tip. This works pretty well as a shield. Now, if I try to mount the unit in a cabinet, the plastic/acrylic bezel reflects the IR, enough to trigger the detector. Is there any material that is *much* more transparent to IR that I could use as a bezel? The only other alternate seems to be to keep the two exposed, without using a bezel on the housing. Do any of you folks have a suggestion? Regards, Anand Dhuru -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu>

> >Now, if I try to mount the unit in a cabinet, the plastic/acrylic bezel >reflects the IR, enough to trigger the detector. Is there any material >that is *much* more transparent to IR that I could use as a bezel? The >only other alternate seems to be to keep the two exposed, without using a >bezel on the housing. BTDT. It was interesting. Two problems. Specular reflection, and diffuse reflection. Specular comes from the surfaces of the window, and the only thing you can do is to arrange the emitter and detector so that the emitter's specular reflection doesn't fall on the detector. Diffuse comes from imperfect window surfaces, and dirt on the window. You can't do much about diffuse other than keeping the optics clean. What I did, was to point my emitter straight up through the window, so that it's reflection came back on itself. I added a pipe (brass pipe cut to length) just short enough not to block the view of the detector. The detector was mounted outside the reflection halo of the emitter, and had it's own shorter pipe. Beware. The LED emits significant energy through it's skirt, and the phototransistor can see through it's own skirt. When I started, I had a contrast ratio (target/no target) of 3/1 through the window. When I finished, I had achieved 7631/1 -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu>

Good morning, Anand I have several IR transceivers working here and I have never found anything that is transparent to IR. It is just like my sonar - if there is ANYTHING in front of it, it is going to reflect. Sid Weaver W4EKQ Port Richey, FL -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu>

At 11:49 AM 5/6/2003 -0400, Sid Weaver wrote: >Good morning, Anand > >I have several IR transceivers working here and I have never found anything >that is transparent to IR. It is just like my sonar - if there is ANYTHING >in front of it, it is going to reflect. Worse than that, some materials are more or less opaque to visible light, and like frosted glass to IR. We hit some white (actually off-white) abs that had about 70% transmissivity in IR, and maybe 5% in visible. Caused serious problems. The vendor subbed this in place of the stuff we spec'd in, that was opaque to IR. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu>

Anand Dhuru wrote: {Quote hidden}> I'm trying to build an IR based object detector based on the 16F628. > The prototype works fine just the way I want it to, but I have run > into a problem in encasing the unit. > > Since it works on detecting IR reflected from an object, the > transmitter LED and the receiver IR module are both to be mounted > within the same case. To reduce the chances of the transmitted IR > being directly incident on the receiver, I've put a piece of heat > shrinkable tube on the LED, exposing just the tip. This works pretty > well as a shield. > > Now, if I try to mount the unit in a cabinet, the plastic/acrylic > bezel reflects the IR, enough to trigger the detector. Is there any > material that is *much* more transparent to IR that I could use as a > bezel? The only other alternate seems to be to keep the two exposed, > without using a bezel on the housing. > > Do any of you folks have a suggestion? > > Regards, > > Anand Dhuru Separated compartments for transmitter and receiver, with independent bezel for each one? Two holes into the case, small IR transparent disks inserted into each hole? Are you sure you need just one lens in front of the TX/RX pair? Wagner -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestspam_OUTTakeThisOuTmitvma.mit.edu>

> > Separated compartments for transmitter and receiver, with independent bezel > for each one? > Two holes into the case, small IR transparent disks inserted into each > hole? Thats an interesting idea, Wagner. > > Are you sure you need just one lens in front of the TX/RX pair? > I just need is a bezel for aesthetic reasons, to cover up the IR components; I dont need a lens. And as you suggested, I could have two, if I can compartmentalize the LED and the module inside the box. I'm still wondering how the commercial units manage this (as in the automatic faucets and flushes) using just one window. Anand -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu>

> >I'm still wondering how the commercial units manage this (as in the >automatic faucets and flushes) using just one window. I told you how I managed it in a commercial unit. 30,000+ of these in the field, with no sensor problems, now that we've purged the IR-translucent "white" plastic! -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-requestKILLspammitvma.mit.edu>

Anand Dhuru wrote: {Quote hidden}> > > > > Separated compartments for transmitter and receiver, with independent > bezel > > for each one? > > Two holes into the case, small IR transparent disks inserted into each > > hole? > > Thats an interesting idea, Wagner. > > > > > Are you sure you need just one lens in front of the TX/RX pair? > > > > I just need is a bezel for aesthetic reasons, to cover up the IR components; > I dont need a lens. > > And as you suggested, I could have two, if I can compartmentalize the LED > and the module inside the box. > > I'm still wondering how the commercial units manage this (as in the > automatic faucets and flushes) using just one window. If you look carefully you'll see that they have a separator moulded into the plastic bezel. They probably also have adjusted the gain since the faucets only operate if you get within 3 inches of the sensors. And they may also detect the -change- in level. e.g. subtract out the baseline IR they see from spillover. R -- http://www.piclist.com hint: To leave the PICList KILLspampiclist-unsubscribe-requestKILLspammitvma.mit.edu>

Dave, you mentioned having mounted the IR LED and the detector in separate copper tubes. Does that mean you were using a photo diode? I am using a 3 pin detector, the one that responds to 38KHz modulated IR. It would be cumbersome to put the module in a tube. Can one get away with non-modulated IR for this task, using a photo diode in place of a 3 pin module? Regards, Anand {Original Message removed}

At 10:41 PM 5/6/2003 +0530, Anand Dhuru wrote: >Dave, you mentioned having mounted the IR LED and the detector in separate >copper tubes. Does that mean you were using a photo diode? Phototransistor. >I am using a 3 >pin detector, the one that responds to 38KHz modulated IR. It would be >cumbersome to put the module in a tube. Just a bit :) >Can one get away with non-modulated IR for this task, using a photo diode in >place of a 3 pin module? I did. The secret there is to blast the emitter, and make the detector relatively deaf. Control your sense point by controlling where the two fields of view overlap. In my app, I was sensing just above the window, which made it harder, but I didn't have to deal with direct sunlight, or strong artificial light. Still, the same principles apply. Also, camcorders see IR quite nicely, making it easier to see what's happening. -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu>

> >I'm still wondering how the commercial units manage this (as in the > >automatic faucets and flushes) using just one window. > > I told you how I managed it in a commercial unit. > 30,000+ of these in the field, with no sensor problems, now that we've > purged the IR-translucent "white" plastic! Hehe, one annoying thing is I have one shirt (black) that is so absorbent of IR that it doesn't reflect enough to be detected! It's like a "stealth" shirt!? It would be cool if it weren't for the fact that most don't offer an alternative way to flush, so I have to stand there with my hand in front of the sensor for a few seconds to trigger it, quite annoying... -- http://www.piclist.com hint: To leave the PICList spamBeGonepiclist-unsubscribe-requestspamBeGonemitvma.mit.edu>

> > Hehe, one annoying thing is I have one shirt (black) that is so > absorbent >of IR that it doesn't reflect enough to be detected! It's like a "stealth" >shirt!? It would be cool if it weren't for the fact that most don't offer an >alternative way to flush, so I have to stand there with my hand in front of >the sensor for a few seconds to trigger it, quite annoying... We found some checks that were like that too. Near 90% absorption. You'll have to wear reflective buttons on that shirt! -- http://www.piclist.com hint: To leave the PICList TakeThisOuTpiclist-unsubscribe-requestEraseMEspam_OUTmitvma.mit.edu>

On Tue, 6 May 2003 14:43:50 -0400, you wrote: >> >I'm still wondering how the commercial units manage this (as in the >> >automatic faucets and flushes) using just one window. >> >> I told you how I managed it in a commercial unit. >> 30,000+ of these in the field, with no sensor problems, now that we've >> purged the IR-translucent "white" plastic! > > Hehe, one annoying thing is I have one shirt (black) that is so absorbent >of IR that it doesn't reflect enough to be detected! It's like a "stealth" >shirt!? It would be cool if it weren't for the fact that most don't offer an >alternative way to flush, so I have to stand there with my hand in front of >the sensor for a few seconds to trigger it, quite annoying... before you try flushing, how about this interesting diversion..... http://web.media.mit.edu/~hayes/mas863/urinecontrol.html -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu>

At 12:19 AM 5/7/03 +0100, you wrote: >On Tue, 6 May 2003 14:43:50 -0400, you wrote: >before you try flushing, how about this interesting diversion..... > >http://web.media.mit.edu/~hayes/mas863/urinecontrol.html And... the creators of the above contraption used a PIC to control it. Here's a snip from their web site: The two-stage signal processing circuit uses an amplifier with a gain ranging from 10-100 and an envelope follower to curb the signal attenuation. A 16F877 PIC microcontroller receives the signals as digital inputs. The impact of a stream of liquid creates a signal that breaks the 2.5-Volt threshold necessary to send the microcontroller s digital inputs high. We chose to use digital inputs because they can be read more rapidly, and we found that the resolution provided by the digital inputs was sufficient for designing a compelling interactive activity. SOFTWARE We programmed a custom interactive game in C++ on the Windows 2000 operating system. Our software reads the state of the sensor array from the microcontroller over a serial data link at a rate of 100 samples per second. The game we chose was a variant of Whack-A-Mole, a classic carnival game. Users aim at a series of jumping hamsters, with input position on the urinal corresponding to target position on the screen above. A successful hit turns a hamster yellow, makes it scream and spin out of control, and rewards the player with ten points. The parabolic trajectories of the hamsters conceal the grid-like arrangement of sensors, resulting in a fluid transition between input and output. -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestEraseME.....mitvma.mit.edu>

> "The parabolic trajectories of the hamsters conceal the grid-like > arrangement of sensors, resulting in a fluid transition between > input and output" "fluid" transition - haha. Are they taking the picc or what ? Maybe change the subject line to "Stupid Tricks With Piss" -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestmitvma.mit.edu>

>> >I'm still wondering how the commercial units manage this (as in the >> >automatic faucets and flushes) using just one window. >> >> I told you how I managed it in a commercial unit. >> 30,000+ of these in the field, with no sensor problems, now that we've >> purged the IR-translucent "white" plastic! > > Hehe, one annoying thing is I have one shirt (black) that is so absorbent >of IR that it doesn't reflect enough to be detected! It's like a "stealth" >shirt!? It would be cool if it weren't for the fact that most don't offer an >alternative way to flush, so I have to stand there with my hand in front of >the sensor for a few seconds to trigger it, quite annoying... I always thought these were passive sensors looking for body heat. apparently not :) -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> >I always thought these were passive sensors looking for body heat. >apparently not :) Generally, passive sensors have a segmented mirror, and a white window, which is transparent at 10 microns. Active reflection sensors have some sort of lensing, and dark red windows which are transparent at 1 micron. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 11:49 AM -0400 5/6/03, Sid Weaver wrote: >Good morning, Anand > >I have several IR transceivers working here and I have never found anything >that is transparent to IR. It is just like my sonar - if there is ANYTHING >in front of it, it is going to reflect. What about needing to evenly diffuse IR? I need to have an omnidirectional source relative to a plane using a single LED. You know those flashlights the ground crew uses to direct an airliner into the gate? How they have a sort of tube or cone clipped on to make a nice torch effect? I've wondered if I could insert the LED into a small solid tube of styrofoam or some such to spread the IR out evenly in all directions...or at least all directions on a plane. What material would efficiently diffuse the IR this way? Or is there a better way (other than using multiple LEDs)? Phil -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> >What about needing to evenly diffuse IR? A diffuse tube, or reflector will work, if you don't need it perfectly even. If so, this will be difficult. LED's have pretty "lumpy" beam patterns as well, don't believe the smooth curves in the datasheet. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 1:59 PM -0500 5/7/03, Dave VanHorn wrote: >>What about needing to evenly diffuse IR? > >A diffuse tube, or reflector will work, if you don't need it perfectly even. >If so, this will be difficult. >LED's have pretty "lumpy" beam patterns as well, don't believe the smooth >curves in the datasheet. OK, but what material would work best for a "diffuse tube"? (Since some materials are opaque to visible light and transparent to IR and so on I am thinking this is a non-trival question...) thanks, Phil -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Philip Galanter wrote: {Quote hidden}> At 11:49 AM -0400 5/6/03, Sid Weaver wrote: >> Good morning, Anand >> >> I have several IR transceivers working here and I have never found >> anything that is transparent to IR. It is just like my sonar - if >> there is ANYTHING in front of it, it is going to reflect. > > What about needing to evenly diffuse IR? > > I need to have an omnidirectional source relative to a plane using a > single LED. You know those flashlights the ground crew uses to > direct an airliner into the gate? How they have a sort of tube or > cone clipped on to make a nice torch effect? I've wondered if I > could insert the LED into a small solid tube of styrofoam or some > such to spread the IR out evenly in all directions...or at least all > directions on a plane. > > What material would efficiently diffuse the IR this way? Or is there > a better way (other than using multiple LEDs)? > > Phil Think about a small cylinder in acrylic. A broken window blinder twister can produce several of those... ;) Make a small hole in one side to insert the LED, now sand the cylinder completely so it turns to be all opaque. Sand the LED face too, so it will be difuse and will spread light all around. Probably, probably, you will have all the cylinder iluminated. It worths the try. You can try without sanding the LED, but I guess it will only iluminate the top of the cylinder. Of course for a larger cylinder I would use several high intensity LEDs, mostly the rectangular ones used in the new car's rear blinkers and lamps substitutes. Wagner. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> >(Since some materials are opaque to visible light and transparent to >IR and so on I am thinking this is a non-trival question...) Opaque to 10 micron IR, yes. but 1 micron is close enough to visible that it's hard to filter out. Anything dealing with remote controls and such is near IR, at 1 micron. Thermal is 10 micron. Then there's that part inbetween :) Frosted glass would work. Sections of glass tubing exposed to flourine gas. You can sand it if you're only doing one. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads