Post by thread:[EE] Proximity Sensors

Has anyone here done much work with proximity sensors? I am looking to trigger an event when someone stands in front of (or walks past) a specific area in a retail environment. The sensing must be done from one side only (no beam break methods). The sensing range would need to be approx 200mm - 1000mm (8" - 40"). The horizontal detection width would need to be approx 500mm (20") at end of the range I guess. There would be multiple sensors, spaced at approx 600mm - 900mm (24" - 36") intervals. There is nowhere to mount the sensors apart from near the floor or *maybe* on the wall higher up. Ideally the sensors need to stand up to a bit of abuse from knocks and maybe a splash from a cleaners mop. Typical machinery proximity sensors would be easy to mount, but seem to be mostly very short range. Ambient light can change somewhat so shadow detection would not be suitable I suspect. I had thought about ultrasonic "range finders" (as used in robotics) but they would probably interact when use this close together. They also wouldn't stand up to the abuse they'd cop being mounted down low. I think typical PIRs are going to be too coarse, but maybe there are better ones than what I've seen so far? David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On 20-Feb-12 23:42, David Duffy (AVD) wrote: > Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). Sharp IR proximity sensors? <http://www.schukat.com/schukat/schukat_cms_en.nsf/index/warengruppe?OpenDocument&wg=A2425&refDoc=CMSB5525456731BA303C125711E004F62C5> Djula {Quote hidden}> > The sensing range would need to be approx 200mm - 1000mm (8" - 40"). > The horizontal detection width would need to be approx 500mm (20") at > end of the range I guess. There would be multiple sensors, spaced at > approx 600mm - 900mm (24" - 36") intervals. > > There is nowhere to mount the sensors apart from near the floor or > *maybe* on the wall higher up. Ideally the sensors need to stand up to > a bit of abuse from knocks and maybe a splash from a cleaners mop. > > Typical machinery proximity sensors would be easy to mount, but seem to > be mostly very short range. Ambient light can change somewhat so shadow > detection would not be suitable I suspect. > > I had thought about ultrasonic "range finders" (as used in robotics) but > they would probably interact when use this close together. They also > wouldn't stand up to the abuse they'd cop being mounted down low. > > I think typical PIRs are going to be too coarse, but maybe there are > better ones than what I've seen so far? > David...

On 2/20/2012 2:42 PM, David Duffy (AVD) wrote: {Quote hidden}> Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). > > The sensing range would need to be approx 200mm - 1000mm (8" - 40"). > The horizontal detection width would need to be approx 500mm (20") at > end of the range I guess. There would be multiple sensors, spaced at > approx 600mm - 900mm (24" - 36") intervals. > > There is nowhere to mount the sensors apart from near the floor or > *maybe* on the wall higher up. Ideally the sensors need to stand up to > a bit of abuse from knocks and maybe a splash from a cleaners mop. > > Typical machinery proximity sensors would be easy to mount, but seem to > be mostly very short range. Ambient light can change somewhat so shadow > detection would not be suitable I suspect. > > I had thought about ultrasonic "range finders" (as used in robotics) but > they would probably interact when use this close together. They also > wouldn't stand up to the abuse they'd cop being mounted down low. > > I think typical PIRs are going to be too coarse, but maybe there are > better ones than what I've seen so far? > David... > Multiple sensors can be used if you control them all. Just use time division multiplexing - only one on at a time. For the other requirements, I would look at IR transmitters/receivers. What is the penalty for false positives

Did you check doppler modules ? -----Mensaje original----- De: spam_OUTpiclist-bouncesTakeThisOuTmit.edu [.....piclist-bouncesKILLspam@spam@mit.edu]En nombre de David Duffy (AVD) Enviado el: Monday, February 20, 2012 2:42 PM Para: PicList Asunto: [EE] Proximity Sensors Has anyone here done much work with proximity sensors? I am looking to trigger an event when someone stands in front of (or walks past) a specific area in a retail environment. The sensing must be done from one side only (no beam break methods). The sensing range would need to be approx 200mm - 1000mm (8" - 40"). The horizontal detection width would need to be approx 500mm (20") at end of the range I guess. There would be multiple sensors, spaced at approx 600mm - 900mm (24" - 36") intervals. There is nowhere to mount the sensors apart from near the floor or *maybe* on the wall higher up. Ideally the sensors need to stand up to a bit of abuse from knocks and maybe a splash from a cleaners mop. Typical machinery proximity sensors would be easy to mount, but seem to be mostly very short range. Ambient light can change somewhat so shadow detection would not be suitable I suspect. I had thought about ultrasonic "range finders" (as used in robotics) but they would probably interact when use this close together. They also wouldn't stand up to the abuse they'd cop being mounted down low. I think typical PIRs are going to be too coarse, but maybe there are better ones than what I've seen so far? David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

Look at this item 120782820889 sold by eBay. Says detection 20 meters... -----Mensaje original----- De: piclist-bouncesKILLspammit.edu [.....piclist-bouncesKILLspam.....mit.edu]En nombre de David Duffy (AVD) Enviado el: Monday, February 20, 2012 2:42 PM Para: PicList Asunto: [EE] Proximity Sensors Has anyone here done much work with proximity sensors? I am looking to trigger an event when someone stands in front of (or walks past) a specific area in a retail environment. The sensing must be done from one side only (no beam break methods). The sensing range would need to be approx 200mm - 1000mm (8" - 40"). The horizontal detection width would need to be approx 500mm (20") at end of the range I guess. There would be multiple sensors, spaced at approx 600mm - 900mm (24" - 36") intervals. There is nowhere to mount the sensors apart from near the floor or *maybe* on the wall higher up. Ideally the sensors need to stand up to a bit of abuse from knocks and maybe a splash from a cleaners mop. Typical machinery proximity sensors would be easy to mount, but seem to be mostly very short range. Ambient light can change somewhat so shadow detection would not be suitable I suspect. I had thought about ultrasonic "range finders" (as used in robotics) but they would probably interact when use this close together. They also wouldn't stand up to the abuse they'd cop being mounted down low. I think typical PIRs are going to be too coarse, but maybe there are better ones than what I've seen so far? David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

part 1 1647 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable) > I think typical PIRs are going to be too coarse I've found PIR isn't too bad, and use it for detection at a variety of ranges and angles. A bare PIR sensor has quite a short range, a couple of hundred mm, which can be substantially increased by a PIR Fresnel lens. The style and size of the lens is important wrt detection, as the plastic itself is the front-end filter, removing all but heat infra- red, and particular lens patterns are used for particular targets eg http://www.infraotica.com/index_files/Page380.htm You might have noticed my post about curtain lenses a couple of days ago ( [OT] PIR curtain lens ). In the right conditions fairly reliable detection is many many metres. Note that PIR sensors contain two elements and the output is the difference, greatest when the movement is perpendicular to the orientation of the elements (ie a target moving horizontally will produce the best output from vertically aligned elements) dV/dT can be used to differentiate between a moving object and a changing environment (drafts, area heating/cooling etc). A low-cut filter is often sufficient to remove < 3-4Hz. Can be done with op-amps or in software to determine velocity Compare the two attached schematics. pir_180 is very low-power, motion_detector is from a security light. Both have high gain at LF Another method I've tried is RF interference, although not practical for my battery-powered applications. Similar to how a Theremin works. An antenna is fed with RF and any disturbance to the field is detected Pretty much immune to almost everything except movement Joe part 2 9011 bytes content-type:image/gif; name="motion_detector.gif" (decode) part 3 6495 bytes content-type:image/gif; name="pir_180.gif" (decode) part 4 181 bytes content-type:text/plain; name="ATT00001.txt" (decoded base64) -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at mailman.mit.edu/mailman/listinfo/piclist

David www.suntekstore.com/goods-14001970-Wireless+Module+Doppler+Radar+Micr owave+Motion+Sensor.html This link has some data and schematics. Look at " document download ". Cheers Rodolfo -----Mensaje original----- De: EraseMEpiclist-bouncesspam_OUTTakeThisOuTmit.edu [piclist-bouncesspam_OUTmit.edu]En nombre de Nily Enviado el: Monday, February 20, 2012 3:05 PM Para: Microcontroller discussion list - Public. Asunto: RE: [EE] Proximity Sensors Look at this item 120782820889 sold by eBay. Says detection 20 meters... -----Mensaje original----- De: @spam@piclist-bouncesKILLspammit.edu [KILLspampiclist-bouncesKILLspammit.edu]En nombre de David Duffy (AVD) Enviado el: Monday, February 20, 2012 2:42 PM Para: PicList Asunto: [EE] Proximity Sensors Has anyone here done much work with proximity sensors? I am looking to trigger an event when someone stands in front of (or walks past) a specific area in a retail environment. The sensing must be done from one side only (no beam break methods). The sensing range would need to be approx 200mm - 1000mm (8" - 40"). The horizontal detection width would need to be approx 500mm (20") at end of the range I guess. There would be multiple sensors, spaced at approx 600mm - 900mm (24" - 36") intervals. There is nowhere to mount the sensors apart from near the floor or *maybe* on the wall higher up. Ideally the sensors need to stand up to a bit of abuse from knocks and maybe a splash from a cleaners mop. Typical machinery proximity sensors would be easy to mount, but seem to be mostly very short range. Ambient light can change somewhat so shadow detection would not be suitable I suspect. I had thought about ultrasonic "range finders" (as used in robotics) but they would probably interact when use this close together. They also wouldn't stand up to the abuse they'd cop being mounted down low. I think typical PIRs are going to be too coarse, but maybe there are better ones than what I've seen so far? David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On 21/02/2012 8:57 AM, Marcel Duchamp wrote: > On 2/20/2012 2:42 PM, David Duffy (AVD) wrote: >> Has anyone here done much work with proximity sensors? I am looking to >> trigger an event when someone stands in front of (or walks past) a >> specific area in a retail environment. The sensing must be done from >> one side only (no beam break methods). >> > Multiple sensors can be used if you control them all. Just use time > division multiplexing - only one on at a time. For the other > requirements, I would look at IR transmitters/receivers. > > What is the penalty for false positives? That's a fair point about multiplexing the sensors. I was envisaging a central controller board that processes up to 8 sensors. Larger installations would have multiple controller boards, or maybe an expander board. There is no serious consequence of a false positive. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On 21/02/2012 8:55 AM, Djula Djarmati wrote: > On 20-Feb-12 23:42, David Duffy (AVD) wrote: >> Has anyone here done much work with proximity sensors? I am looking to >> trigger an event when someone stands in front of (or walks past) a >> specific area in a retail environment. The sensing must be done from >> one side only (no beam break methods). > Sharp IR proximity sensors? > > <http://www.schukat.com/schukat/schukat_cms_en.nsf/index/warengruppe?OpenDocument&wg=A2425&refDoc=CMSB5525456731BA303C125711E004F62C5> > > Thanks for the link. I might get a couple of those style sensors to experiment with. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

Thanks for the link. I checked them out but they have a min range of 3 metres, which will cause triggering by people way outside the target range. David... On 21/02/2012 9:42 AM, Nily wrote: {Quote hidden}> David > > www.suntekstore.com/goods-14001970-Wireless+Module+Doppler+Radar+Micr > owave+Motion+Sensor.html > > This link has some data and schematics. Look at " document download ". > > Cheers > > Rodolfo > > -----Mensaje original----- > De: RemoveMEpiclist-bouncesTakeThisOuTmit.edu [spamBeGonepiclist-bouncesspamBeGonemit.edu]En nombre de > Nily > Enviado el: Monday, February 20, 2012 3:05 PM > Para: Microcontroller discussion list - Public. > Asunto: RE: [EE] Proximity Sensors > > > Look at this item 120782820889 sold by eBay. > > Says detection 20 meters... > > -----Mensaje original----- > De: TakeThisOuTpiclist-bouncesEraseMEspam_OUTmit.edu [RemoveMEpiclist-bouncesTakeThisOuTmit.edu]En nombre de > David Duffy (AVD) > Enviado el: Monday, February 20, 2012 2:42 PM > Para: PicList > Asunto: [EE] Proximity Sensors > > > Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). > > The sensing range would need to be approx 200mm - 1000mm (8" - 40"). > The horizontal detection width would need to be approx 500mm (20") at > end of the range I guess. There would be multiple sensors, spaced at > approx 600mm - 900mm (24" - 36") intervals. > -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On 21/02/2012 9:40 AM, IVP wrote: >> I think typical PIRs are going to be too coarse > > I've found PIR isn't too bad, and use it for detection at a variety of > ranges and angles. A bare PIR sensor has quite a short range, a > couple of hundred mm, which can be substantially increased by a PIR > Fresnel lens. The style and size of the lens is important wrt detection, > as the plastic itself is the front-end filter, removing all but heat > infra- > red, and particular lens patterns are used for particular targets Hi Joe, I might revisit the PIR modules. Finding something sturdy enough as an off the shelf product is the ideal scenario. We *could* build sensors, but I'd rather not just yet as it's just an idea with no commitment to bulk quantity yet. If I could buy a ready to go sensor with a cable tail for under $50 in small quantity that would suffice. If the idea gets up, then we can look at custom made sensors to get the price down. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On Tue, Feb 21, 2012 at 9:09 AM, David Duffy (AVD) <davidEraseME.....audiovisualdevices.com.au> wrote: > I might revisit the PIR modules. Finding something sturdy enough as an > off the shelf product is the ideal scenario. We *could* build sensors, > but I'd rather not just yet as it's just an idea with no commitment to > bulk quantity yet. If I could buy a ready to go sensor with a cable > tail for under $50 in small quantity that would suffice. If the idea > gets up, then we can look at custom made sensors to get the price down. Maybe you can try to look at some photoelectric sensors -- look for "Diffuse Mode" sensors. Sensing range of 1000mm is not a problem. The mutual interference part can be a problem but your requirement may not require such a feature as the distance between the sensors are sufficiently apart. For example, you can search here, not so sure about the "US$50 in small quantity part" as I no longer follow the product closely. www.pepperl-fuchs.us/usa/en/classid_47.htm Disclaim: I used to work for Pepperl+Fuchs until 2006. Other vendors include Banner Engineering, SICK, Omron, SUNX, etc. -- Xiaofan

On 2/20/2012 5:09 PM, David Duffy (AVD) wrote: > red, and particular lens patterns are used for particular targets > Hi Joe, > I might revisit the PIR modules. Finding something sturdy enough as an > off the shelf product is the ideal scenario. We *could* build sensors, > but I'd rather not just yet as it's just an idea with no commitment to > bulk quantity yet. If I could buy a ready to go sensor with a cable > tail for under $50 in small quantity that would suffice. If the idea > gets up, then we can look at custom made sensors to get the price down. > David... > Zilog makes a cheap one supposedly good to 5 meters: 269-2513-ND for about $6.00 If it senses too broad, you can shutter it a bit to narrow the view. Think "barn doors" on photo lighting equipment. Sparkfun has some also

On 21/02/2012 11:46 AM, IVP wrote: >> I might revisit the PIR modules > You could try one (security sensors, especially exterior, are often > at giveaway prices, w/o bulbs) and window the lens with paint or > paper to limit its view Panasonic have some small PIR modules starting around the $15 mark that might do if can reduce the range. I tried an ultrasonic ranging module (SRF02) today and it works quite well. That may be a solution if ceiling mounting them becomes an option as they should be away from abuse and dust up there. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

Possibly the Doppler radar can be restricted to a shorter range by attenuating the signal. Perhaps a wire mesh in front of the unit. There is also a pot that claims to adjust the range..... gus {Quote hidden}> On Feb 20, 2012, at 6:03 PM, David Duffy (AVD) wrote: > > Thanks for the link. I checked them out but they have a min range of 3 > metres, which will cause triggering by people way outside the target range. > David... > > On 21/02/2012 9:42 AM, Nily wrote: >> David >> >> www.suntekstore.com/goods-14001970-Wireless+Module+Doppler+Radar+Micr >> owave+Motion+Sensor.html >> >> This link has some data and schematics. Look at " document download ". >> >> Cheers >> >> Rodolfo >> >> -----Mensaje original----- >> De: EraseMEpiclist-bouncesmit.edu [RemoveMEpiclist-bouncesEraseMEEraseMEmit.edu]En nombre de >> Nily >> Enviado el: Monday, February 20, 2012 3:05 PM >> Para: Microcontroller discussion list - Public. >> Asunto: RE: [EE] Proximity Sensors >> >> >> Look at this item 120782820889 sold by eBay. >> >> Says detection 20 meters... >> >> -----Mensaje original----- >> De: RemoveMEpiclist-bouncesspam_OUTKILLspammit.edu [RemoveMEpiclist-bouncesTakeThisOuTspammit.edu]En nombre de >> David Duffy (AVD) >> Enviado el: Monday, February 20, 2012 2:42 PM >> Para: PicList >> Asunto: [EE] Proximity Sensors >> >> >> Has anyone here done much work with proximity sensors? I am looking to >> trigger an event when someone stands in front of (or walks past) a >> specific area in a retail environment. The sensing must be done from >> one side only (no beam break methods). >> >> The sensing range would need to be approx 200mm - 1000mm (8" - 40"). >> The horizontal detection width would need to be approx 500mm (20") at >> end of the range I guess. There would be multiple sensors, spaced at >> approx 600mm - 900mm (24" - 36") intervals. >> > > -- > ___________________________________________ > David Duffy Audio Visual Devices P/L > Unit 8, 10 Hook St, Capalaba 4157 Australia > Ph: +61 7 38235717 Fax: +61 7 38234717 > Our Web Site: http://www.audiovisualdevices.com.au > ___________________________________________ >

Ok, if you have enough light to see the targets, you can always detect them using low cost cameras. By the way, you can use to detect and log ( if the latter is needed). Rodolfo -----Mensaje original----- De: EraseMEpiclist-bouncesspamspamBeGonemit.edu [RemoveMEpiclist-bouncesKILLspammit.edu]En nombre de David Duffy (AVD) Enviado el: Monday, February 20, 2012 5:04 PM Para: Microcontroller discussion list - Public. Asunto: Re: [EE] Proximity Sensors Thanks for the link. I checked them out but they have a min range of 3 metres, which will cause triggering by people way outside the target range. David... On 21/02/2012 9:42 AM, Nily wrote: > David > > www.suntekstore.com/goods-14001970-Wireless+Module+Doppler+Radar+Micr {Quote hidden}> owave+Motion+Sensor.html > > This link has some data and schematics. Look at " document download ". > > Cheers > > Rodolfo > > -----Mensaje original----- > De: piclist-bouncesSTOPspamspam_OUTmit.edu [spamBeGonepiclist-bouncesSTOPspamEraseMEmit.edu]En nombre de > Nily > Enviado el: Monday, February 20, 2012 3:05 PM > Para: Microcontroller discussion list - Public. > Asunto: RE: [EE] Proximity Sensors > > > Look at this item 120782820889 sold by eBay. > > Says detection 20 meters... > > -----Mensaje original----- > De: KILLspampiclist-bouncesspamBeGonemit.edu [EraseMEpiclist-bouncesEraseMEmit.edu]En nombre de > David Duffy (AVD) > Enviado el: Monday, February 20, 2012 2:42 PM > Para: PicList > Asunto: [EE] Proximity Sensors > > > Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). > > The sensing range would need to be approx 200mm - 1000mm (8" - 40"). > The horizontal detection width would need to be approx 500mm (20") at > end of the range I guess. There would be multiple sensors, spaced at > approx 600mm - 900mm (24" - 36") intervals. > -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

----- Original Message ----- From: "David Duffy (AVD)" <@spam@david@spam@spam_OUTaudiovisualdevices.com.au> To: "PicList" <spamBeGonepiclistKILLspammit.edu> Sent: Monday, February 20, 2012 5:42 PM Subject: [EE] Proximity Sensors > Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). Ultrasonics polled in order so they don't interfere with each other? -- Bob Ammerman RAm Systems

> Ok, if you have enough light to see the targets, you can always detect > them > using low cost cameras. By the way, you can use to detect and log ( if the > latter is needed). I read down looking for the camera suggestion and was surprised to not see it - but Rodolfo saved the day. Canon A series camera until recently could be controlled using software loaded on the SD card and written using the flexible powerful and free CHDK aka Canon Hack Development Kit. http://chdk.wikia.com/wiki/CHDK Every known and available feature of the cameras can be controlled plus you can get them to do things outside their apparent specs. You can write programs for the camera to run or set up various boundary conditions including detecting movement within a predefined area. The camera could deliver an image or just a trigge Various builds and supported models http://mighty-hoernsche.de/ Features http://chdk.wikia.com/wiki/Features NB: Motion detection - Trigger exposure in response to motion, fast enough to catch lightning Scripting - Control CHDK and camera features using ubasic and Lua scripts. Enables time lapse, motion detection, advanced bracketing, and much more. Many user-written scripts are available on the forum and wiki. USB remote - Simple DIY remote allows you to trigger exposures or scripts. Manual: http://chdk.wikia.com/wiki/CHDK_User_Manual More Russell McMaho

May I suggest you use COTS PIR sensors from the alarm industry? They are small, cheap, modular, and readily available. Many can be set to close a contact immediately on motion, or after 2 or 3 'beams' have been passed by. 12V at 50 mA for power makes them easier to use than most 'security light' types. Many have 'range' pots so you can dial down their sensitivity, and by aiming them correctly, at the floor at 3' distance, you prevent false triggers from targets farther away than you desire. Many also come with interchangable lenses with differing numbers of focal elements, giving you more or fewer detection beams. You can mask out the beam pattern with opaque paint to get exactly the sensing box you desire. Given their range, you could mount them on the ceiling in a store, to get them out of harms way, while giving you exactly the trigger box you desire. They also have the advantage of KISS, where a contact closure triggers your playback/ animation, so no fancy controller required. Rober

part 1 1281 bytes content-type:text/plain; charset="gb2312" (decoded base64) After some discussion on another list (forum), the idea of using vehicle parking sensors came up. This seems ideal as the sensors are quite cheap, already sealed and designed for simple through panel installation. The basic idea seems to be to send out a 40KHz ring pulse and time it's return. Has anyone ever interfaced these off-the-shelf sensors to a PIC? David... On 21/02/2012 8:42 AM, David Duffy (AVD) wrote: > Has anyone here done much work with proximity sensors? I am looking to > trigger an event when someone stands in front of (or walks past) a > specific area in a retail environment. The sensing must be done from > one side only (no beam break methods). > > The sensing range would need to be approx 200mm - 1000mm (8" - 40"). > The horizontal detection width would need to be approx 500mm (20") at > end of the range I guess. There would be multiple sensors, spaced at > approx 600mm - 900mm (24" - 36") intervals. > -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________ part 2 181 bytes content-type:text/plain; name="ATT00001.txt" (decoded base64) -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at mailman.mit.edu/mailman/listinfo/piclist

On 29/02/2012 9:45 PM, IVP wrote: >> Has anyone ever interfaced these off-the-shelf >> sensors to a PIC? > > Popular in robotics. Google ultrasonic robot Yeah, already played with the robot ones. They are open transducers. The sealed ones need a lot more drive level it seems. I guess my question is really two-fold; A) Does anyone have a reasonable reference design for interfacing to the sealed parking type sensors? B) Is there some basic code out there that shows what's required the ping and receive? There are lots of Google hits for unsealed Tx-Rx pair system, but not much for the sealed all in one units. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

> Yeah, already played with the robot ones. They are open transducers. > The sealed ones need a lot more drive level it seems. > > I guess my question is really two-fold; > > A) Does anyone have a reasonable reference design for interfacing to the sealed > parking type sensors? > > B) Is there some basic code out there that shows what's required the ping and > receive? > > There are lots of Google hits for unsealed Tx-Rx pair system, but not much for the > sealed all in one units. > David... Sounds like time to mount the 'scope in the car, and do some manoeuvring .... -- Scanned by iCritical.

part 1 2050 bytes content-type:text/plain; charset="gb2312" (decoded base64) I bought one China-made parking sensor with 4 transducers to reverse engineer. Very interesting design. Uses a Cypress 8051-based PSOC with some programmable analog building blocks. I drawn parts of its schematic, I can send it if you are interested. It uses canned trimmable transformers, similar to radio IF transformers, but with just two two-pin windings. The analog functions of the PSOC seems to be used for filtering. It uses one CMOS analog multiplexer to scan the transducers. There is just one analog front-end What I liked most in that device is that the same transducer sends and receives the pulses, and they are waterproof. It cost USD25.00 and comes with four transducers. It is cheaper than buying just the transducers from Digikey. Best regards, Isaac Em 29/2/2012 08:30, David Duffy (AVD) escreveu: > After some discussion on another list (forum), the idea of using vehicle > parking sensors came up. This seems ideal as the sensors are quite > cheap, already sealed and designed for simple through panel installation. > > The basic idea seems to be to send out a 40KHz ring pulse and time it's > return. Has anyone ever interfaced these off-the-shelf sensors to a PIC? > David... > > On 21/02/2012 8:42 AM, David Duffy (AVD) wrote: >> Has anyone here done much work with proximity sensors? I am looking to >> trigger an event when someone stands in front of (or walks past) a >> specific area in a retail environment. The sensing must be done from >> one side only (no beam break methods). >> >> The sensing range would need to be approx 200mm - 1000mm (8" - 40"). >> The horizontal detection width would need to be approx 500mm (20") at >> end of the range I guess. There would be multiple sensors, spaced at >> approx 600mm - 900mm (24" - 36") intervals. >> > > > > > ----- > No virus found in this message. > Checked by AVG - http://www.avg.com > Version: 10.0.1424 / Virus Database: 2113/4837 - Release Date: 02/28/12 part 2 181 bytes content-type:text/plain; name="ATT00001.txt" (decoded base64) -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at mailman.mit.edu/mailman/listinfo/piclist

On 29/02/2012 10:03 PM, .....alan.b.pearcespam_OUTstfc.ac.uk wrote: >> Yeah, already played with the robot ones. They are open transducers. >> The sealed ones need a lot more drive level it seems. >> >> I guess my question is really two-fold; >> >> A) Does anyone have a reasonable reference design for interfacing to the sealed >> parking type sensors? >> >> B) Is there some basic code out there that shows what's required the ping and >> receive? >> >> There are lots of Google hits for unsealed Tx-Rx pair system, but not much for the >> sealed all in one units. >> David... > Sounds like time to mount the 'scope in the car, and do some manoeuvring .... I have an after market 4 sensor system on the way. :-) Hopefully, being cheap they don't use any esoteric parts. The PIC code should be easy enough now that I've found a bit more info on their general operation. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

Em 29/2/2012 08:52, David Duffy (AVD) escreveu: > On 29/02/2012 9:45 PM, IVP wrote: >>> Has anyone ever interfaced these off-the-shelf >>> sensors to a PIC? >> Popular in robotics. Google ultrasonic robot > Yeah, already played with the robot ones. They are open transducers. > The sealed ones need a lot more drive level it seems. > > I guess my question is really two-fold; > > A) Does anyone have a reasonable reference design for interfacing to the > sealed parking type sensors? > > B) Is there some basic code out there that shows what's required the > ping and receive? > > There are lots of Google hits for unsealed Tx-Rx pair system, but not > much for the sealed all in one units. > David... I measured several parameters of the one I bought from China (frequency, pulse width, number of pulses, etc.). I can send the results if I can find them... Isaac

On 29/02/2012 10:10 PM, Isaac Marino Bavaresco wrote: > I bought one China-made parking sensor with 4 transducers to reverse > engineer. > Very interesting design. > > Uses a Cypress 8051-based PSOC with some programmable analog building > blocks. > I drawn parts of its schematic, I can send it if you are interested. > It uses canned trimmable transformers, similar to radio IF transformers, > but with just two two-pin windings. I found one "hack" article and it showed the transformers you mention. Not sure how available (in small qty - say 100) they are. {Quote hidden}> The analog functions of the PSOC seems to be used for filtering. > > It uses one CMOS analog multiplexer to scan the transducers. There is > just one analog front-end > > > What I liked most in that device is that the same transducer sends and > receives the pulses, and they are waterproof. > It cost USD25.00 and comes with four transducers. It is cheaper than > buying just the transducers from Digikey. Crazy isn't it! David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

On 29/02/2012 10:13 PM, Isaac Marino Bavaresco wrote: > I measured several parameters of the one I bought from China (frequency, > pulse width, number of pulses, etc.). I can send the results if I can > find them... That would be great Isaac. Only if it's no bother though. David... -- ___________________________________________ David Duffy Audio Visual Devices P/L Unit 8, 10 Hook St, Capalaba 4157 Australia Ph: +61 7 38235717 Fax: +61 7 38234717 Our Web Site: http://www.audiovisualdevices.com.au ___________________________________________

Em 29/2/2012 09:16, David Duffy (AVD) escreveu: > On 29/02/2012 10:13 PM, Isaac Marino Bavaresco wrote: >> I measured several parameters of the one I bought from China (frequency, >> pulse width, number of pulses, etc.). I can send the results if I can >> find them... > That would be great Isaac. Only if it's no bother though. > David... No problem, it will be a pleasure. What I want is to design my own system for R/C use but I don't have much time for hobby now, so the project is dormant. Besides, as I'm mostly a digital guy, I'm struggling with the filter part. I can calculate and design active filters, but I'm not sure what are the necessary parameters for them, and I don't have time to do much experimentation. Perhaps we could cooperate in the design. Best regards, Isaac

At 13.03 2012.02.29, you wrote: >> Yeah, already played with the robot ones. They are open transducers. >> The sealed ones need a lot more drive level it seems. >> >> I guess my question is really two-fold; >> >> A) Does anyone have a reasonable reference design for interfacing to >the sealed >> parking type sensors? >> >> B) Is there some basic code out there that shows what's required the ping and >> receive? >> >> There are lots of Google hits for unsealed Tx-Rx pair system, but >not much for the >> sealed all in one units. >> David... > >Sounds like time to mount the 'scope in the car, and do some manoeuvring .... Nothing specific to You: The wife does the manoeuvring, and the husband uses the oscilloscope. or The husband does the manoeuvring, and the wife uses the oscilloscope. What will work best? *g*