## 7 segment display current limiting resistors

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Ken1
Posts: 121
Joined: Sat Dec 31, 2005 1:01 am
Location: Sandy Lake, Manitoba
Contact:

### 7 segment display current limiting resistors

I am using a 7447 display driver IC to drive a 7 segment LED display. How do I determine the value of the current limiting resistors for each segment?

JPKNHTP
Posts: 488
Joined: Wed Jun 29, 2005 1:01 am
Location: Missouri
Contact:
-JPKNHTP
-God Bless

Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.
And if they are strobed, you can divide by the time factor.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:
Most of my LED projects driven from a 5 volt decoder, I have used a 330 ohm resistor per segment . I generally use hi-efficiency red and they are multiplexed. This has always given me a satisfactoy display. There are many variables involved though such as LED color and size and ambient light. The brightness chosen is an arbitrary value by the individual user.And of course its always a trade off between brightness versus power conservation for battery operated equipment. As long as you stay within the LED's limits, you can pick any current you want to suit your application. Since you do not mention the type you are using, only you have the specs for them. Read the specs and apply ohms law as mentioned. Its really one of the simplest calculations involved in cicuit design.

jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:
There are also other matters to consider. Are the seven segment displays common anode or common cathode.

Is the version of the chip you are using a open collector or does it have internal pull up resistors? I believe the 7447 can only SINK current and has a open collector output. Thus you would use a pull up resistor and bias the LED normally being sure you do not exceed the sink capacity of the 7447 when the bias resistor is shorted to ground. I do not think you will be sourcing current from the chip to the 7 segment display.
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:
Hi there,

As Chris mentioned an important factor in the selection of a current
limiting resistor is if the display will be multiplexed. If it is not,
then you simply follow Ohm's Law. If, on the other hand, it WILL
be multiplexed then you need to know the duty cycle so that you
can figure the average current in each LED segment.
The duty cycle here is defined as the ratio of the time on to
the total time: DC=Ton/Ttotal .
You can then multiply the DC times the value you get using Ohms
Law.

For the same example stated elsewhere, lets say we have a 5v
power supply, the 7447 drops 0.5 volts when on, and the LED
segments each drop 1.7 volts when on. This means the power
voltage V=5-1.7-0.5=2.8 volts.
Now assuming you are shooting for 10ma LED current (20ma for
higher brighness, 10ma for longer life), the resistor value would
be 2.8/0.010 ohms, or 280 ohms if the LED was on all the time
(ie your not multiplexing). If you are multiplexing however, and
you are multiplexing four (4) digits, then you need to divide that
resistor by 4: 280/4=70 ohms. See how simple that works?
The only thing left to do is figure the peak current and if that
increases the voltage of both the LED when it's on and the
7447 voltage drop when one of it's outputs is turned on, then
recalculate the voltage V and the resistor value.
Note that for 4 digits where you want brightness approximately
equal to 10ma for each LED you need 40ma for each LED when
they are multiplexed.

I've succesfully multiplexed an array of 8 red LED's for my
(now somewhat old) frequency counter. If you like i can look
up all the values i used (1/4 inch high red LED's).

Oh yeah, note also that the scan frequency has to be such that it
refreshes each LED fast enough to make it look constantly lit up.
I think i used 60 Hz as my rule of thumb, so each digit gets refreshed
60 times per second. Going lower means the display might appear
to be blinking, and going higher means the LED brightness drops
off more even though the duty cycle remains the same.
LEDs vs Bulbs, LEDs are winning.

Ken1
Posts: 121
Joined: Sat Dec 31, 2005 1:01 am
Location: Sandy Lake, Manitoba
Contact:
Hi,
The display is not strobed. It's 3 digits driven by 3 7447's (DM7447AN is printed on the chips) and the power supply is 5V. The LED's I have are common anode, red in color and have an If rating of 20mA according to the datasheet. They are NAR 161 LED's, Digikey part #404-1124-ND. What I am actually doing is converting a digital display that used 7 segment incandescent lamp digits. The incandescent displays are obsolete. One end of each of the 7 incandescent bulbs in the display is connected directly to the 5V supply and the other end of each bulb is connected directly to the 7447 driver.

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:
Hi Ken,

Ok then you dont have to use the duty cycle in the calculation.

R=(Vp-Vd-Vled)/I

where

R is the resistor value (per segment) in ohms
Vp is the power supply voltage in volts
Vd is the chip (7447) drop when 'on', in volts
Vled is the LED segment drop, in volts
I is the current through the LED segment, in amps

Example
Your 5v power supply, and say the drop in the 7447 is 0.4 volts and
the LED drop is 1.7 volts, and current is 15ma, we then have:
R=(5-0.4-1.7)/0.015
which works out to
R=193 ohms

You could use a 180 ohm or 220 ohm resistor.
The wattage rating is under 1/4 watt.
LEDs vs Bulbs, LEDs are winning.

Dean Huster
Posts: 1263
Joined: Wed Dec 05, 2001 1:01 am
Location: Harviell, MO (Poplar Bluff area)
Contact:
Just don't make the mistake of trying to save a few cents and some PCB space by using a single current-limiting resistor (per display digit) in series with the common lead and none in series with the segment leads. It'll work, but every digit will have a different display brightness because of a differing number of segments turning on, changing the current draw through the resistor. I've seen it done that way and it makes the display look stoooooopid!

Dean
Dean Huster, Electronics Curmudgeon
Contributing Editor emeritus, "Q & A", of the former "Poptronics" magazine (formerly "Popular Electronics" and "Electronics Now" magazines).

R.I.P.

Ken1
Posts: 121
Joined: Sat Dec 31, 2005 1:01 am
Location: Sandy Lake, Manitoba
Contact:
Hi,
Thanks everypne for your replies! I completed the project with 7 180 ohm resistors for each digit and it works perfectly. The only difficulty I had was placing the resistors. The displays are not attached to the board but rather sit alongside the board at a right angle. As the pinout of the LED displays are different than the old incandescent displays, I had to criss cross the resistors over and under each other to connect them to the correct pins on the LED displays.

philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:
HI again,

Yes, some pics would certainly be nice at this point

BTW, Dean, yeah, i used one resistor per digit a long time ago
just to see what it would look like, and wow, was that funny
looking ha ha. Looked like something was 'wrong' with the display
so that it appeared that something had gone wrong with the electronics.
With my Frequencey counter (that has 8 digits) i went to multiplexing
because i just couldnt see using eight 7 segment decoders and
56 (yes count em, fifty six) resistors he he. Instead i just needed
one seven seg decoder and 7 resistors, a 555 for scan freqency, and
eight quad NAND gates (or NOR gates, dont rem now), a counter,
and a 4 to 8 line decoder. The frequency was set somewhere around
8 times 60 Hz (480 Hz).
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:
Beyond three digits-multiplexing is the only way to go,unless you want to drive your self nuts wiring in all those little suckers!

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:
Hello again,

Robert: You just reminded me...now i remember another motivation
for going to multiplexing in the case of my 8 digit freq counter...

I had purchased three or four 3-digit seven seg led displays,
where each 3 digit LED group had only 7 segments, 1 dec point,
and 1 common cathode (leads). This meant the 3 digits in each
with that in mind. I used three groups of 3 each for a total of
nine digits, but only used 8 of them. I would have had to purchase
more LED's all single units in order to drive each one separately.
Still, im sure i would have went with multiplexing anyway to save
on resistors and wiring, and plus it made it possible to have to run
only 15 leads (no decimal point) from the main count board to the
display board. Not multiplexing would have meant running something
like 64 wires from main board to display
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:
Mr Al
I notice in your reply that you omitted the decimal point in your counter. That would give it an upper limit of 100 Mhz (without assumption). My last two LED counter projects were of one with 2 four digit displays and the other with 2 three digit displays. These were all multeplexed units for obvious reasons. The real head ache was that the decimal points (which were required in both units) were also multeplexed with no obvious pro vision to light them independantly.I actually had to incorporate some homebrew decoders into the circuit to light them at the proper time, and what a pain in the a*s that was. These were recently purchased from Digi-key and I Don't have the manufacturer at the moment. Have you run into this situation at all and if so, what was your "fix".

### Who is online

Users browsing this forum: No registered users and 23 guests