ultrasonic motion detector

[mikeb]

i built a ultra sonic motion detector from a article from the poptronics website. the article says the unit should measure motion from 4 to 7 meters away but im only able to measure motion up to a distance of about a inch to a inch an a half. i was wanting to know if i increase the current flow thru the transmitting transducer would that likely increase the range of the unit.
the oscillator is built around the 4049 ic an transmitts at 40khz, i was thinking of adding maybe a darlinton amp at the output of the oscillator which would interface the oscillator to the transmitter. if theres someone with knowledge on how i can increase the range of this unit to at least 12 inches id much appreciate it.

[rshayes]

The signal should be inverse fourth power with distance (inverse square from transmitter to target times inverse square from target to receiver). Your range (1.5 inches) compared to the expected range (160 inches) is shorter than expected by a factor of over 100. This implies that something is wrong by a factor of about 100 million (assuming inverse fourth) or at best 10 thousand (assuming inverse square).<p>I suspect that the problem is in the transducers. Many ultrasonic transducers are mechanically resonant, and have a relatively narrow bandwidth. If the oscillator frequency does not match the transmitting transducer's resonant frequency, very little energy will be transmitted. If this frequency does not match the receiving transducer's frequency, even less energy will be received.<p>If the transducers are the piezoelectric type, I would try measuring their resonant frequencies. The impedance of the transducer will be generally capacitive, with a dip at the resonant frequency. If you can vary the oscillator frequency it should be possible to locate this dip. Adding a series resistor in the 1K range might make the dip more visible. If the signal level is limited, you may be able to test the receiving transducer using the transmitting circuit. Ideally, the two transducers should match, and the oscillator should be tuned to the same frequency.

[Edd]

Wow, its sort of tricky to track down some of their stuff since their demise…thank goodness for aliasing!.
Probably this do be the one …..and possibly the only one that they ever had….right?<p>http://www.web-ee.com/Schematics/Motion ... tector.pdf<p>The transmitter certainly seems to take care of full utilization of all of the hex’s stages. One should be able to track the 40khz right on thru the buffer and output stages with merely a meter..much less a scope. Also something just wants to question your xtal’s integrity for the unit. If you have a well stocked Junque box you might call in a 32,768 xtal from an old watch/electronic wall clock, etc for an exploratory/temporary test sub for the 40 Khz unit at a somewhat mismatched/reduced output level.
On the receiver portion the unit might be tested with one of the OLD generation of ultrasonic TV remote xmitters if you happen to have or know of a contact that still has one that you might borrow. In a crude manner, you also might jingle a set of keys in front of the rec transducer unit if you have a scope to monitor on down the lineup of the 324 outputs at # 1,7,8 & 14. They certainly were able to throw out upper harmonics and cause some false function triggering on some of my old remote control TV’s of that vintage.
Also I must have 24 or so virgin new 40 KHZ units in their egg crate packaging medium. I believe that these units have either xmit OR rec intercomatibility, you’re certainly free to some, if that proves to be the nature of your problem. I had been thinking of a passenger side of the vehicle detection for blind side bandits or for lane changing insurance….alas… no developmental time at the present.<p>On your last para’s comment …..that final stage should be saturating and fully driving that stages transducer, and a 9 V supply level should put out one hefty ultrasonic signal. Also not knowing what equipment you have, but if you happen to have a sine/sq wave osc or pulse generator you could drive the xmit transducer directly at the 40khz and bring it up on the receiver transducer as per Steve. Of the two units , somehow, I’m suspecting the xmitter unit to be your problem. Also on that 12 in range as max ? and considering that you have been using a big enough target to reflect back the signal, you potentially might have to cant both of the the units inward at an angle to optimize reflected intercoupling at that close of a distance, if the target is of a small profile.<p>73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled-Spam*Cookies*Crumbs)
<p>[ March 29, 2004: Message edited by: Edd Whatley ]</p>

[jwax]

I'm all for ultrasonics applications, but if it's motion detection you want, infra-red detectors are cheap, directional, and available in all types of ranges.

[Mike]

i'd like to build one of those too. Do you have a higher quality copy of the pcb than in the datasheet?

[Edd]

Mike:
You can at least be grateful that PCB layout was presented in Acrobat format, plenty of editing flexibility. No problem to transport to WP/Word –Paint and clean it up to the extent that it satisfies you.
I found a 1x scale dimension occurring for me at 88.9% using micro adjustable zoom tool on a test print out to a HP 6L LaserJet ($8.68 at a thrift store..”non-working”…..piece of cake). When one zooms on up for an inspection into the approx1600% mag region, the trace lines seem to have been made by an errant CNC micromill tooling attempt, with resultant trace edge aberrations. Personally, I would just drop in some IC overlays on top of the two existing IC outlines, as they are tending to be distorted undersize, just a skosh. Of more import to me would be the absence of hole references, so I would be more into transporting in a quad cluster of pixels to produce an etch dimple for component lead holes marking/and/resultant drilling ease, later on, after etch.
lexpro:
Perchance did you buy the whole chicken kit , complete with all the feathers, for that project, or just get the exotica, like the ultra s transducers and the xtal, and is the xtal a canned quartz unit , or possibly a ceramic resonator? I checked my u s transducers and they were R units ,but I could easily work around that.Better for them to be that instead of T units.<p>73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled-Spam*Cookies*Crumbs)
<p>[ March 29, 2004: Message edited by: Edd Whatley ]</p>