Post by thread:ultrasonic ranging

I am working on a project that needs to find the distance to objects. Currently I am using using a PIC16C84 as my controller. I'm using 40kHz transmit and recieve transducers from polaroid. I am driving the transmit transducers by attatching one of it's leads to RA1 and the other to RA2. In my code RA1 is always the opposite of RA2 and I switch them at 40kz 20 times, to make an outgoing train of pulses. My reciever transducer has one pin to ground and one to a 2 transister amp (2n2222's). I am having problems with range, I fear recier amp is not strong enough, but if I increase it's gain, I detect ghost (false) signal returns. Any suggestions to improve my design? I want a range of 2-15 feet accurate within only a foot. Also,does anyone know of a way to narrow by field of detection. My current design detect objects off to the side.

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 11:32 AM 7/12/96 -0400, you wrote: >I am working on a project that needs to find the distance to objects. >Currently I am using using a PIC16C84 as my controller. I'm using 40kHz >transmit and recieve transducers from polaroid. I am driving the >transmit transducers by attatching one of it's leads to RA1 and the >other to RA2. In my code RA1 is always the opposite of RA2 and I switch >them at 40kz 20 times, to make an outgoing train of pulses. My reciever >transducer has one pin to ground and one to a 2 transister amp >(2n2222's). I am having problems with range, I fear recier amp is not >strong enough, but if I increase it's gain, I detect ghost (false) >signal returns. Any suggestions to improve my design? I want a range of >2-15 feet accurate within only a foot. Also,does anyone know of a way to >narrow by field of detection. My current design detect objects off to >the side. There was an article on a PID pong device in a Circuit Cellar Ink from about a year ago that detailed using the polaroid transducer to sense distance. The author could possibly give you a dirth of information as he addressed ghost images very briefly. Sorry I can't be more help. Terry Ewing -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBM8hkjnhdHoVbUC3kEQKsOACg6trrK5TwrESxUstfD1xlyvWbuSYAoOuR J6Ta2XYDnt+VBPyAC1FfYSvO =TeOP -----END PGP SIGNATURE-----

Are you working with the indoor or the ruggedized Polaroid transducer? The indoor (the round black mesh-looking one) has a lot better gain, but doesn't like being splashed alot. The outdoor is elipsoid-shaped and has a more focused pattern. In any case, you have to be careful of: The ringing of the input signal after sending the sonar pulse (You must blank for x amount of time.) targets that will bounce back an even stronger signal. (sometimes) signals from other transducers firing in the same frequency. broadband sonar noise -- Try shooting white noise ala air brakes or an air hose nearby and watch a real strong echo. Both frequency and voltage on the output transducer are critical with rather narrow bands. I've got a patent in that field and had so solve a lot of stupid little problems to get the devices to work reliably. (PIC C57 which also powered the RS-485 network.) You can get more accurate than a foot -- Ours is running .6 inches reliably. I'd like to get more detailed with code, etc. but I'll bet our customer would get reasonably upset. Tom ---------- From: pic microcontroller discussion list on behalf of Frank B. Scalzo Sent: Friday, July 12, 1996 11:32 AM To: spam_OUTPICLISTTakeThisOuTMITVMA.MIT.EDU Subject: ultrasonic ranging I am working on a project that needs to find the distance to objects. Currently I am using using a PIC16C84 as my controller. I'm using 40kHz transmit and recieve transducers from polaroid. I am driving the transmit transducers by attatching one of it's leads to RA1 and the other to RA2. In my code RA1 is always the opposite of RA2 and I switch them at 40kz 20 times, to make an outgoing train of pulses. My reciever transducer has one pin to ground and one to a 2 transister amp (2n2222's). I am having problems with range, I fear recier amp is not strong enough, but if I increase it's gain, I detect ghost (false) signal returns. Any suggestions to improve my design? I want a range of 2-15 feet accurate within only a foot. Also,does anyone know of a way to narrow by field of detection. My current design detect objects off to the side.

part 0 2274 bytes ---------- From: Tom Mariner[SMTP:.....Tom_MarinerKILLspam@spam@MSN.COM] Sent: Sunday, July 13, 1997 8:41 AM To: PICLISTKILLspamMITVMA.MIT.EDU Subject: Re: ultrasonic ranging Are you working with the indoor or the ruggedized Polaroid transducer? The indoor (the round black mesh-looking one) has a lot better gain, but doesn't like being splashed alot. The outdoor is elipsoid-shaped and has a more focused pattern. In any case, you have to be careful of: The ringing of the input signal after sending the sonar pulse (You must blank for x amount of time.) targets that will bounce back an even stronger signal. (sometimes) signals from other transducers firing in the same frequency. broadband sonar noise -- Try shooting white noise ala air brakes or an air hose nearby and watch a real strong echo. Both frequency and voltage on the output transducer are critical with rather narrow bands. I've got a patent in that field and had so solve a lot of stupid little problems to get the devices to work reliably. (PIC C57 which also powered the RS-485 network.) You can get more accurate than a foot -- Ours is running .6 inches reliably. I'd like to get more detailed with code, etc. but I'll bet our customer would get reasonably upset. Tom ---------- From: pic microcontroller discussion list on behalf of Frank B. Scalzo Sent: Friday, July 12, 1996 11:32 AM To: .....PICLISTKILLspam.....MITVMA.MIT.EDU Subject: ultrasonic ranging I am working on a project that needs to find the distance to objects. Currently I am using using a PIC16C84 as my controller. I'm using 40kHz transmit and recieve transducers from polaroid. I am driving the transmit transducers by attatching one of it's leads to RA1 and the other to RA2. In my code RA1 is always the opposite of RA2 and I switch them at 40kz 20 times, to make an outgoing train of pulses. My reciever transducer has one pin to ground and one to a 2 transister amp (2n2222's). I am having problems with range, I fear recier amp is not strong enough, but if I increase it's gain, I detect ghost (false) signal returns. Any suggestions to improve my design? I want a range of 2-15 feet accurate within only a foot. Also,does anyone know of a way to narrow by field of detection. My current design detect objects off to the side.

On Monday, July 14, 1997 12:32 AM, pic microcontroller discussion list On Behalf Of Andy Tucker wrote: {Quote hidden}> ------ =_NextPart_000_01BC8FED.AC0EBF40 > Content-Type: text/plain; charset="us-ascii" > Content-Transfer-Encoding: 7bit > > Do you have a source for the outdoor transducer? > TIA > Andy Tucker > > ---------- > From: Tom Mariner[SMTP:EraseMETom_Marinerspam_OUTTakeThisOuTMSN.COM] > Sent: Sunday, July 13, 1997 8:41 AM > To: PICLISTspam_OUTMITVMA.MIT.EDU > Subject: Re: ultrasonic ranging > > Are you working with the indoor or the ruggedized Polaroid transducer? The > indoor (the round black mesh-looking one) has a lot better gain, but doesn't > like being splashed alot. The outdoor is elipsoid-shaped and has a more > focused pattern. > > In any case, you have to be careful of: > > The ringing of the input signal after sending the sonar pulse (You must blank > for x amount of time.) > targets that will bounce back an even stronger signal. (sometimes) > signals from other transducers firing in the same frequency. > broadband sonar noise -- Try shooting white noise ala air brakes or an air > hose nearby and watch a real strong echo. > Both frequency and voltage on the output transducer are critical with rather > narrow bands. > > I've got a patent in that field and had so solve a lot of stupid little > problems to get the devices to work reliably. (PIC C57 which also powered the {Quote hidden}> RS-485 network.) You can get more accurate than a foot -- Ours is running .6 > inches reliably. I'd like to get more detailed with code, etc. but I'll bet > our customer would get reasonably upset. > > Tom > > ---------- > From: pic microcontroller discussion list on behalf of Frank B. Scalzo > Sent: Friday, July 12, 1996 11:32 AM > To: @spam@PICLISTKILLspamMITVMA.MIT.EDU > Subject: ultrasonic ranging > > I am working on a project that needs to find the distance to objects. > Currently I am using using a PIC16C84 as my controller. I'm using 40kHz > transmit and recieve transducers from polaroid. I am driving the > transmit transducers by attatching one of it's leads to RA1 and the > other to RA2. In my code RA1 is always the opposite of RA2 and I switch > them at 40kz 20 times, to make an outgoing train of pulses. My reciever > transducer has one pin to ground and one to a 2 transister amp > (2n2222's). I am having problems with range, I fear recier amp is not > strong enough, but if I increase it's gain, I detect ghost (false) > signal returns. Any suggestions to improve my design? I want a range of > 2-15 feet accurate within only a foot. Also,does anyone know of a way to > narrow by field of detection. My current design detect objects off to > the side. > > > > Yes, both Polaroid and Murata Erie have ruggedized units. The Polaroid is a single unit and the Murata device has separate send and receive transducers. In both cases you will have to whack them with a lot more power to get the range you want, but they have much better water / shock / temperature specs. Also, the units we used a few years ago had a much narrower resonant frequency band than the indoor types. We also experienced some unit to unit frequency drifts. Tom