seven segment driving

Status
Not open for further replies.

zeroskj1988

New Member
i want to drive 12 digit seven segment. i am using common anode (positive) seven segment. i connect segment to directly to pic16f877a's portb. since due to deficiency of pin. i use cd4514 demux to drive common anode of the seven segment. since output current of cd4514 is very much less. seven segment light dull. can anybody suggest me remedy for this.
whether uln is an option
 
That device can be powered up to 15v! May be you need to drive the chip with 12 volts... You can still command it with 5 volts.. that way you can get up to 8mA per segment..
 
Last edited:
Hi zero,

Multiplexing of several digits always produces a dim display unless you use somewhat high current drivers. The reason for this is because each segment is only on for a fraction of the total time.
For example, if your segment is on for 1/7 of the time and the max current from the driver is 20ma, that means the segment only sees 20ma divided by 7 which equals roughly 3ma average current, and it's the average current that produces the light intensity not the peak current of 20ma. That would mean the display is being grossly under driven and so wont be very bright at all.
The driver BTW has to provide enough current in both the anode and cathode, so that could mean both upper and lower drivers have to be changed.

The only solution using the current LED display type is to increase the peak drive current. 70ma for example would produce 10ma in each LED, and 140ma would produce an average of 20ma per segment in the above example. This assumes that the data sheet says it allows 140ma or higher peak.
If your display segments are not connected together internally you could use direct drive, but that requires a lot of chips.

You may be able to go to a higher brightness LED type digit also, but that would of course mean buying all new display digits.
 
Last edited:
I was going to vote for integrated muxed drivers, but the NXP SAA1064 is the only one available. It'll do it, but writing the I2C interface is a push versus direct driving the high side with P Channel MOSFETS and the low side with a ULN2008 each from a uC port pin.
A problem with high volt sourcing and uC port sinking is you easily go Way past port tolerence.
Unless there isn't much else the uC has to do, I'd figure on dedicating a separate small uC to it and using its' serial port for data transfer. Good Hunting... <<<)))

P.S. the MM5450 (now from Micrel) is viable, but illustrates what a shame the Toshiba TB62709 is gone.
 
Last edited:
Also consider the MM5450 from NatSemi.

The 5450 will drive four seven segment digits, you will need 3 of them.
The micro will need five pins to drive them, Data, Clock, Enable1, Enable2 and Enable3.

JimB
 
Hey Jim... I thought they stopped making them.. I must try looking harder. We used to use these but found it difficult / dearer to get hold of... we moved to a serial unit in the end..

They are a good chip though
 
I dont know Ian, they are in my mind at the moment because I am at the moment using a couple which I bought a few years ago.

They started life on a project which I recently re-worked and is now using an LCD display.
I am now re-using the 5450s for something which is just using four 7 seg LEDs.

JimB

On edit: I just looked at the RS website, they have them but under the Micrel name rather than NatSemi.
 
Last edited:

I am not sure, but you can get some idea from this.......
**broken link removed**
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…