Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!usc!cs.utexas.edu!news.sprintlink.net!news.indirect.com!seeker
From: seeker@indirect.com (Stan Eker)
Subject: Re: RSTM printer and TEST modes info
Message-ID: <D8ponK.BAA@indirect.com>
Sender: usenet@indirect.com (Internet Direct Admin)
Organization: Internet Direct, indirect.com
Date: Wed, 17 May 1995 07:26:08 GMT
References: <D8nxAz.F2v@indirect.com>
X-Newsreader: TIN [version 1.2 PL2]
Lines: 52

To continue the saga of gutting the Radio Shack Electronic Tape Measure,
here's the latest...

The 50KHz burst used to ping the transducer is only about 500uS long, and
can be reduced to 400uS without compromising the accuracy overmuch.  That
only moves the minimum measurable distance in about a half inch, not much of
an improvement.  The _displayed_ numbers will change, but the delta time
from beginning of pulse to first recognizable echo (measured with a port pin
from your favorite micro) is easily modified.  Wasted effort, there.

BTW - I was dain-bramaged in the last message.  You only need a SINGLE pin
to monitor the distance: U3 pin 3.  When it goes high, a ping has started. 
When it goes low again, you've got the first return echo.  Subtract out
1.8mS of the time (in code, not the hard way) and you have the TOF of the
echo.  If it's 0 or nearly so, it's sub-1.7 feet.

The big part of the measurement limitation is because the electrostatic
transducer is driven UNDAMPED, or nearly so.  It rings on for roughly 1.3mS
after the end of the transmission ping.  Grumble.  I'll be playing with
various forms of damping over the next week or two, but if anyone out there
has prior design experience with these Polaroid transducers, I'd appreciate
a few hints or sage advice.

A simple passive damping method would reduce the ringing, but would also
probably drastically reduce the far echo reception, not a good thing.  I'm
not sure if it's advisable to just dump the energy across a transistor or
two (by shorting it out) to reduce the ringing in the transducer; I don't
want to reduce it's life just to shorten the min. distance.  Looks like it's
time for another trip to ASU's engineering library!

A note in Johann Borenstein's `Where Am I' report suggests there's a note
about exactly this problem (near echo reception with Polaroid xducers) in a
1991 Polaroid product literature flyer titled "Electronic Ranging System". 
If anyone has a copy (and is willing to find and distill the reference),
that'd cut days or much more off the info hunt.  Thanx, if you can...


One other good reason for improving the damping of the transducer:  in
playing around, I noticed several times that hard returns at about an inch
or two from the transducer caused it to read 30 FEET.  Imagine this:

 transducer to 68HC11: "We're 30 feet away from the zone, Captain"

 68HC11 to drive motors: "Full speed ahead, Mr. Mabuchi!"

 contact sensors to 68HC11: "Captain, she canna' take much more of this!"

 (repeat from start)

It's probably best to make sure that THIS scene doesn't make it into reruns,
especially if you don't have any wisker sensors.

