🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.electronics
Article #8151

Re: Design a bar graph

#8151
From: Norman Lin
Date: Sat, 03 Nov 1990 05:50
51 lines
3381 bytes
tindle@ms.uky.edu (Ken Tindle) writes:

>I have three binary lines, for eight possible states (yeah, oh wow).  How
>can one light seven LED's from this, with zero having all LED's off and
>seven lighting all seven of them, in bar graph fashion?  A dot display is
>easy, but bar is tougher.  As a final touch, one should tax each digital
                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^
>line with but one standard TTL load. 
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

If I understand this restriction, this pretty much rules out any solution
involving a demultiplexer, since one would need several OR gates connected
to the demultiplexer output, each OR gate having one input from the current
output line and another from the previous (i.e. line 4 would go high either
if line 4 was selected from demultiplexer input 010 OR if line 5 was high,
since we want a bar graph, line 5 -> line 4 which similarly -> line 3 ->
line 2 -> line 1, so we get a continuous 1-2-3-4-5 bar output).  Thus,
the only solution I see (and you are right, it is a bit surprisingly,
and, I might add, annoyingly, difficult!) is to use some sort of an arithmetic
unit with binary or perhaps BCD output, since only this way can you control
all 7 lines independently and simultaneously (as needed for your bar graph)
while still only "taxing each digital line with but one standard TTL load."
Stated in the following manner, the problem may perhaps be more easily solved:

   Given the binary number 000, what do I do to get the binary number 000 0000?
                           001                                        000 0001
                           ...                                        ... ....
                           110                                        011 1111
                           111                                        111 1111

I see no trivial solution to this, but I do believe looking at the problem
this way may help in finding its solution.

Of course the brute-force least-parts solution is to write out the entire
truth table and use a PLA... :-)

>I can think of two ways to implement this, but they both seem like too many
>parts have to be used.

What were your methods?

>--------------------------\ /-----------------------------------------------
>INTERNET:tindle@ms.uky.edu | "Could you please continue the petty bickering?
>BITNET: tindle@ukma.bitnet |  I find it most intriguing."   ---    Data, 
>Ken Tindle - Lexington, KY |  Star Trek, The Next Generation, "Haven"
>--------------------------/ \-----------------------------------------------
-- 
\    /\/\    /\/\        +-----------------------------+    /\/\    /\/\    /./
\\  /./\\\  /./\\\  /====|Norman Lin/norlin@129.15.20.2|===/./\\\  /./\\\  /./
\\\/./  \\\/./  \\\/./===| norlin@uokmax.ecn.uoknor.edu|====/  \\\/./  \\\/./
 \/\/    \/\/    \/\/    +-----------------------------+        \/\/    \/\/
Technical Details
Message-ID: <1990Nov3.055034.23343@uokmax.ecn.uoknor.edu>
Path: arc1-ca.pugleaf.net!utzoo!utgpu!watserv1!watmath!att!att!rutgers!apple!uokmax!norlin
References: <16307@s.ms.uky.edu>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!utgpu!watserv1!watmath!att!att!rutgers!apple!uokmax!norlin
From: norlin@uokmax.ecn.uoknor.edu (Norman Lin)
Newsgroups: sci.electronics
Subject: Re: Design a bar graph
Message-ID: <1990Nov3.055034.23343@uokmax.ecn.uoknor.edu>
Date: 3 Nov 90 05:50:34 GMT
References: <16307@s.ms.uky.edu>
Distribution: na
Organization: Engineering Computer Network, University of Oklahoma, Norman, OK
Lines: 51