Jump to content
rebeltaz

can a 7448 drive two LEDs by itself

Recommended Posts

I am building a(nother) clock with a custom 7-segment display. Each "segment" will be made up of two high brightness LEDs. The LEDs I am looking to use right now are blue (although I may use white) with a voltage rating of 3.2-3.8v and 35mA maximum continuous current and 70mA peak current. The two LEDs will be driven in series with the segments wired in common-anode configuration.

 

If I power the 7447 with 9v, do you think it would be able to drive the display alone, or should I buffer that with a ULN2003? I tend to try and over-engineer things, so I wanted to get a second opinion before I finalized the design.

 

thanks guys :)

 

-[edit]-

Sorry... I meant 7447 in the subject line, not 7448.

Share this post


Link to post
Share on other sites

What does the datsheet say is the maximum amount of current the 7447 can sink?

Now how about the ULN2003?

 

Which one will handle two LEDs at full brightness?

 

My money is on the ULN2003s.

Share this post


Link to post
Share on other sites

first off, please note the 7447 maximum Vcc is about 6 volts, from it's 5 volt TTL heritage.     i suspect that 9 volts could burn your 7447.   

i quickly add that i've pushed TTL and HC logic well over their Vcc/Vdd max on many occasions to get more speed (out of HC).

 

secondly, a 7447 with 5v --->    this means that turning on 2 series blue LEDs, whose combined total voltage drop is somewhere around 6 to 7ish volts (batch dependent) might not happen if you stick to 7447 Vcc specs.     so - your decision #1 is how to turn on those 2 LEDs reliably, for which you could try boosting the 7447 Vcc on your breadboard.    this in fact means the 7447 drivers power dissipation per driver pin can be very small, and this is an important, practical side-effect.

    Example:   Vcc = 7.0v   Vled = 3.2v   Vsat 7447 = 0.1v     so the 7447 output transistor drops only 0.5v at its given current.

As a practical matter, you have to adjust Vcc on the breadboard so that the 2 blue LEDs turn on.

 

However ---  here's an idea you may want to consider ---   'it the spirit of experimentation' ;)  - and to understand how volume production handles this situation  ;)  ;)  

 

Try pulsing the LEDs.   you'll likely be doing this as part of the clock digit scan routine anyway.   If you were using only ONE LED, you would connect the MSP430 output pin directly to the LED itself, anode or cathode depending on your design.

 

 

 

TIMING (one LED driver pin  -or-  7447 RBIpin5.   or 4bits if you are sending bcd to 7447)   or etcetera:

======

 

    __________                       _________________________        _________.....                ____

___|  dim LED |_____________________|     bright LED          |______|                      ....___|

.  A          B                     A                         C      A               D             A

 

 

--> timing A to A is constant, let's call it the 'scan rate'

--> timing A to B produces a dim LED as it is short

--> timing A to C produces a bright LED as it is 2x longer than A-B

--> BtoA   CtoA  and   DtoA     are required to prevent 'ghosting'

 

 

 

The current into an LED that you measure with a DC connection will be higher than the current used by the LED in a pulsed situation.

This is what consumer electronics folks figured out decades ago to cut cost in display units.

 

It is better design.  As a plus, one can become more familiar with programming duty cycles with microprocessors for all sorts of things :  LEDs, stepper / other motors,  even switching power supplies have been I/O pins, a timer ISR, inductor and transistor for ages now.

 

Last point, which is certainly known by you as an experienced clock maker making a(nother) one --- your could read the ambient room light level with a 'light dependent resistor', average it's value, then change the LED on-times accordingly.

Share this post


Link to post
Share on other sites

Well... loking at the datasheet for the 7447, it looked to me like it can handle 15v, or am I not reading that correctly?

 

Yeah, the LEDs will be pulsed since the display is going to be matrix scanned, but I've never been a fan of dimming LEDs based on ambient light. Besides that, these LEDs are going to need to shine through a semi-opaque colored plastic, so they will need to be run as brightly as possible.

post-28620-0-67648500-1413133380_thumb.png

Share this post


Link to post
Share on other sites

I thought about the 4511, which DOES show an operating voltage between 3-20v, but it's "high-output-sourcing-capability" is only 25mA... 10mA below the nominal operating current on a single LED.

post-28620-0-51051300-1413134874_thumb.png

Share this post


Link to post
Share on other sites

 I think your design should work without any issues. The only thing you need to ensure however is if the IC can sink the current spike which is produced when all LEDs are lit on at once. Since you

Share this post


Link to post
Share on other sites

 I think your design should work without any issues. The only thing you need to ensure however is if the IC can sink the current spike which is produced when all LEDs are lit on at once. Since you

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×