Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
Michael Karas
11/07/11 09:52
Read: 396 times
Beaverton Or
USA


 
Msg Score: +1
 +1 Good Answer/Helpful
#184579 - Do not let inputs float...
Responding to: Arif Deshmukh's previous message
Arif Deshmukh said:
Hi,
Thanks for replies.

I want to drive four Common Anode 7-segment display.
I tested on 74LS245, its out is at logic "1" when its input is kept floating or not connected to "1" or "0", so all the segment of display always on.
So I am looking for such buffer IC whose output strictly follows the input. I mean when input is 1 or 0 then only output is 1 or 0 and if input is floating output should be at high impedance.
Suggest me any other solution of this problem if possible.

Bye


Arif,

You should re-read the posting that Kai made to this thread. Do not let inputs float!! As a derivative comment to that - Do not let floating (or open) inputs be a design criteria that you try to depend upon for your electronics implementation.

Even in days when I still used "LS" logic it was an unacceptable practice to leave inputs open. These days when CMOS type parts are the prevalent choice it is EVEN MORE IMPORTANT to not let inputs be open of float. Kai gave many of the reasons why. The most worrisome ones being those that could blow out the input gates of the part due to the 10^^12 ohm input impedance on this type of part.

Do note that on some parts the chip designer has taken care to provide some pull-up or pull-down resistors so that unused inputs can be left open in some instances. Take for example the MAX232 chip (http://www.maxim-ic.com/datasheet...p/id/1798). It has 400K ohm pullups on its inputs on the CMOS input side so it is generally safe to leave these open if there is NOTHING connected to the package pin. Do note that if there is any copper connection to the open input of such chip with the on-board 400K resistor or if the part usage scenario is such that there is a lot of noise or switching activity on adjacent copper connections then it would be a wise idea to provide for a lower impedance drive to the pins using the output of some other gate or a pullup resistor with a much lower value. Size selection of course is design specific and depends on a number of factors including power source, whether there is another driving source that must compete with the pullup and whether you are trying to pull in the opposite direction than the on-chip pull-up/pull-down.

Michael Karas


List of 14 messages in thread
TopicAuthorDate
Non Inverting Buffer      Arif Deshmukh      11/05/11 21:32      
   3 states?      Andy Neil      11/06/11 00:53      
   are You sure?      Stefan KAnev      11/06/11 03:18      
      Describe the goal - not the step      Andy Neil      11/06/11 06:46      
         bidirectional driver      Jan Waclawek      11/06/11 06:56      
   Redesign your circuit!      Kai Klaas      11/06/11 06:04      
   You've described..        Oliver Sedlacek      11/07/11 04:54      
      Goal      Arif Deshmukh      11/07/11 06:07      
         "its input is kept floating" do you mean      Erik Malund      11/07/11 06:39      
         let's get down to low-level details      Richard Erlacher      11/07/11 07:21      
         Common Anode 7-segment display      Stefan KAnev      11/07/11 08:34      
            No need to float the signals      Per Westermark      11/07/11 08:56      
         Do not let inputs float...        Michael Karas      11/07/11 09:52      
      Absolutely Right.      Michael Karas      11/07/11 10:10      

Back to Subject List