Newsgroups: comp.robotics
Path: brunix!uunet!gatech!udel!news.intercon.com!psinntp!dg-rtp!webo!kitkat!pmalenfa
From: pmalenfa@kitkat.webo.dg.com (Paul Malenfant)
Subject: Re: Ultrasonic Rangefinde
Sender: usenet@webo.dg.com (Usenet Administration)
Message-ID: <1993Apr16.215926.15425@webo.dg.com>
Date: Fri, 16 Apr 93 21:59:26 GMT
Reply-To: pmalenfa@kitkat.webo.dg.com (Paul Malenfant)
References:  <gate.TsiV2B1w165w@toz.buffalo.ny.us>
Organization: Data General Corporation, Westboro, MA
Lines: 33

In article <gate.TsiV2B1w165w@toz.buffalo.ny.us>, cyberman@toz.buffalo.ny.us (Cyberman) writes:
|> {Michael Fulbright} said
|>    "Ultrasonic Rangefinders -"
|>       to <All> on 04-07-93  18:01
|>  MF> 1 Pair of Ultrasonic transducers (an transmitter and a receiver)
|>  MF> 1 NE555 timer chip
|>  MF> 1 NE567 Tone detector
|>  MF> 2 741 op amps
|> 
|>     I suggest you use a 32,768hz crystal for your clock source or
|>     a ceramic oscillator.  These are VERY stable and cheap. In
|>     fact you can get 38khz crystals.  You can use a CMOS invertor
|>     for buffering it's output.  Anyhow 555's are notorious for
|>     drift.

I built a similar circuit to that described above (from the Robotics
Bonanza book). It uses a 3.056MHz crystal and divides it to get 40k.

I have two problems with it that maybe someone could help.

First, the range is very short, <4 ft reliably.  Maybe the transmitter
is too quiet, or the receiver is not sensitive enough.  Should I
try a better transmitter/receiver pair than the $3.50 pair I got?

Second, the cpu does not get realistic/reliable timing numbers.
They are always much longer than the 2ms/ft that sound travels.
It is interfaced to the cpu through an optoisolator, but I don't
know how it gets number 10x what they should be.

-- 

Paul Malenfant
pmalenfa @ kitkat.webo.dg.com (508-870-6460)
