home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.whtech.com
/
ftp.whtech.com.tar
/
ftp.whtech.com
/
Geneve
/
9640news
/
MBP
/
WEATHER
< prev
next >
Wrap
Text File
|
2006-10-19
|
6KB
|
126 lines
#2 Cypress Drive
Wichita, KS 67206
June 22, 1986
Mr. William M. Lucid
6005 Elaine Street
Speedway, IN 46224-3032
Dear William,
I've looked high and low for the design details on our
weather station but haven't been able to find them. My MBP
partner Jerry has moved to another house and currently has almost
all of the old MBP documents in storage. Therefore, I can't give
you all of the details on how we made all of the weather stuff
work. I can however give you some theory on how it worked that
might help and I have included a listing of the XBASIC subroutine
that we used in the TIBBS program.
The wind direction indicator was built by having a vane that
turned with the wind. We built it from plastic Easter eggs. The
vane was mounted to the egg and the entire egg was allowed to
rotate by placing a hollow rivet in the bottom of the egg and a
rivet in the top of the egg. The upper rivet was also hollow and
we placed a small ball from a ball bearing in the upper rivet for
the center shaft to ride on. We used a brush cut down from a
generator brush as a wiper that sent voltage down to the ADC.
The brush rubbed on a printed circuit board that had been
segmented into 8 sectors. The N sector was connected to ground
through a 1K 1/4 watt resistor. The NW sector was connected to
the 5 volt supply through a similar resistor. Each of the
remaining segments was connected to its neighbor with a similar
resistor. What we had then was a voltage divider ladder with
large pads that we could touch with a brush and get 8 different
voltages. It sounds like a hard thing to do and it wasn't easy
to build but we couldn't find a continuous pot without a large
dead area in it. At any rate, all we had to do was read the
voltage and we knew the wind direction. The thing wasn't
entirely free of trouble because the contact resistance between
wiper and the pcb segments was always changing. We evently used
a scheme that said if the voltage is above that on the sector to
the left and below that for the sector to the right then it must
be in the current sector. That worked well and the unit
functioned well on the Wichata TIBBS for over 4 months from the
time we put it up until we took it down. The reason for the
resistors between the ground and the first segment and 5 volts
and the last segment was that the brush overlapped the segment
edges and we didn't want the 5 volts to see a direct path to
ground (ie through the brush).
The anemometer was a little more difficult. If you're not
familar with electronics you may want to get some help on this
one. What we had was another Easter egg with cups on it that was
free to turn. The rotor had magnets on it that closed a read
switch. One side of the read switch went to 5 volts and the
other to the trigger of a one shot that fired each time the
switch closed. The output from the one shot then charged a
capacitor that had a leak path to ground. The more often the one
shot was fired, the more often a measured charge was dumped to
the capacitor and thus the higher the charge on the capacitor.
The high side of the capacitor then went to one of the ADC
channels. We calibrated the unit by putting it out of the window
of the car and driving down the street. The darn thing is
amazingly linear. That also worked very well for the whole time.
Reed switches are somewhat difficult to find.
With the above explanations, I think that you will now be
able to understand the subrountine listing included with this
letter. Line 7020 addresses the anemometer ADC channel and
converts the voltage to mph. You will see that the equation for
voltage versus speed was not quite linear. In line 7030 we check
to see if the indicated velocity is below 1 mph. If it is, we
set up AQ$ to indicate a calm wind and skip down to get the
temperatures. Line 7040 sets up BQ$ with velocity and then
determines the wind direction by reading a different ADC port.
DW$ is defined elsewhere when the BBS program is initialized as:
DW$(1)="NE"
DW$(2)="E"
DW$(3)="SE" etc.
Line 7070 gets the outside temperature and line 7080 gets the PEB
temperature. AQ$ contains the wind direction and velocity, BQ$
contains the outside air temperature, and CQ$ contains the PEB
temperature when we return from the subrountine. We used LM335
sensors with an amplifer with the gain of 4 and some offset to
get better temperature resolution. I can't remember (or find
data on) the amplifier right now but it was one of the few op
amps that will work near the rails on a 5 volt supply.
If you get absolutely despirate for the design details,
leave me another message and I'll drive to Jerry's and we'll go
through his storage until we find it.
Thanks for your support of MBP! We haven't had time to work
on that much in the last year. There is a new download on
Wichita TIBBS that shows how to hookup analog joysticks to the
MBP card for use with TI-ARTIST. The program is written in
assembly level language and shows how to use a delay in the
program to allow the ADC time to do its thing. You might be
interested in that...
Yours truly,
William Byrne
7020 AQ$= Wind is "::CALL PEEK(-31078,aq)::CALL PEEK(-
31072,AQ)::AQ=INT(-.03252*AQ+.0017667*AQ*AQ)
7030 IF AQ< THEN AQ$=AQ$&"calm."::GOTO 7070
7040 BQ$=STR$(AQ)::CALL PEEK(-31088,AQ)::CALL PEEK(-
31072,AQ)::FOR DQ=1 to 7
7050 Next DQ::DIR$="N"
7060 AQ$=AQ$&DIR$&" at "&BQ$&" mph"
7070 CALL PEEK(-31082,AQ)::CALL PEEK(-31072,AQ)::BQ$=" Outside
temp. is "::AQ=INT(109.7597-.4897*AQ)::BQ$=BQ$&STR$(AQ)&" F"
7080 CQ$=" My temp. is "::CALL PEEK(-31086,AQ)::CALL PEEK(-
31072,AQ)::AQ=INT(118.0-.5556*AQ)::CQ$=cq$&STR$(AQ)&" F"::RETURN
p.s. I assume that your weather station will also have pressure.
I've never been able to get one of the Motorola tranducers you
suggested. The May 1, 1986 issue of EDN (Electronic Design News)
has some interesting infomation on transducers.
Note: This letter was re-typed by William M. Lucid and uploaded
FYI to owners of the MBP card.