To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 486
485  |  487
Subject: 
Re: Multiplexing Input Circuit
Newsgroups: 
lugnet.robotics
Date: 
Thu, 12 Nov 1998 15:59:07 GMT
Original-From: 
Larry Coffin <lcoffin@pointinfinity.SPAMLESScom>
Viewed: 
2334 times
  
I think Sven thought of this first.  Anyway I threw a circuit together
last night that uses the SetSensorType to switch between 4 inputs. I put
it on a page at:
http://ourworld.compuserve.com/homepages/Michael_Gasperi/mux.htm
It seems to work well, but startup is a problem.  It starts out on input
1 but if you stop the program it can get lost.  I need to add a reset
circuit.  Any suggestions?

Looks good!

How about using the third input port to assert the reset line and
set the counter back to zero?

Is your counter set up as a binary counter? From your diagram it
looks like there are times that more than one sensor would be active -- say
on count 3 both O0 and O1 would be active, right?

Do you need to have bit 5 connected to the reset line? I would
think you could simply continue counting -- the other lines might go high
but the four lines you are using should just wrap back around.

I've been using the MAX337 and the MAX336 CMOS MUXs -- the 337 is a
dual 8 channel mux, the 336 is a single 16 channel mux. See the specs at
http://209.1.238.250/arpdf/1704.pdf (from
http://www.maxim-ic.com/efp/AnalogSM.htm). They cost about $6.50 each
through Arrow electronics (http://www.arrowamericas.com). With the counter
in binary mode you could run these MUXes to cycle through 8 or 16 sensors
(or more with multiple muxes).


From your diagram it looks like your sensors are sharing a direct
connection to the RCX input port. Did you have any trouble with orientation
of the contact on the RCX? When I tried the same setup (through my MUX
though) I found that I would get bad sensor readings when it was connected
in one orientation and good readings when rotated 180 degrees. I assumed
that there was something going on that had one sensor continue to feed a
signal back to the RCX even after it was not supposed to be selected.

---Larry


+-----------------------------------------------------------------------+
  Larry Coffin                                lcoffin@PointInfinity.com
  Point Infinity                           http://www.PointInfinity.com
  Nantucket, Massachusetts                 508-325-6677 * fax: 325-6676
+-----------------------------------------------------------------------+

Micro Credo:
Never trust a computer bigger than you can lift.


-



Message has 2 Replies:
  Re: Multiplexing Input Circuit
 
(...) The counter is a CMOS decade counter that has only one output high for each count. There is only one analog switch on at a time. The counter is reset when it hits the value 5 so it jumps back to 0 to start over. The analog switch is (...) (26 years ago, 12-Nov-98, to lugnet.robotics)
  Re: Multiplexing Input Circuit
 
(...) The setup I was using was bi-directional too, I'm not sure exactly why I was seeing it, but I remember getting a high reading for a non-pressed touch sensor when the next one over (the previous one in the cycle?) was pressed. If I reversed the (...) (26 years ago, 12-Nov-98, to lugnet.robotics)

Message is in Reply To:
  Multiplexing Input Circuit
 
I think Sven thought of this first. Anyway I threw a circuit together last night that uses the SetSensorType to switch between 4 inputs. I put it on a page at: (URL) seems to work well, but startup is a problem. It starts out on input 1 but if you (...) (26 years ago, 12-Nov-98, to lugnet.robotics)

6 Messages in This Thread:




Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact

This Message and its Replies on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR