Why won't the IC work???

Status
Not open for further replies.

Electronics4you

Member
Hi there,

I have just bought a simple ULN2003AN to drive a small light bulb from a PIC. I did a simple test first.

Pin 1 == 5V
Pin 8 == GND
Pin 9 == 12V

Pin 16 is the output, but it did not reach 12V as expected when powering the input (pin 1) with 5V. What seems to be the problem here?
 
The ULN 2003 has 7 inverting open colector buffers.

When an input pin is high (5V), the output is connected to ground (0 V)

When the input is low (0 V) the output isn't conneced.


You must connect one side of your lamp ro the ULN's output and the oher side to your supply (+ 12V)
 
The problem is that your expectations are wrong. The data sheet plainly states that this is a darlington transistor array. You would not expect the output to go to 12V since (as the data sheet schematic shows) it's a transistor collector with the emitter to ground. The output acts as a switch to ground. For the test you need a load such as a resistor or light bulb connected between 12V and the IC output (500ma maximum current). Then +5V on the input will connect the output to ground and allow current to flow through the load.

(Isn't it amazing what you can learn from a data sheet?)
 
Last edited:
Electronics4you said:
Okay, didn't notice that about the inverting function and got it to work, but without the COM pin. Why's it there?
Click to expand...
Again referring to the data sheet, the logic diagram shows that the COM pin is connected to the output clamp diodes connected to each collector. The COM pin can be connected to the +12V to clamp any transients above the supply voltage due to an inductive load (such as a relay coil) that could otherwise blow the output transistor.
 
Thanks for the advice. Now finally I tried to use the ULN2003AN as a LED matrix driver just by grounding the GND pin and then place the chip in serial with the GND-lines for the display for amplification. It works fine for one LED, but when I set up the speed to power up all LEDs the illumination from each LED is very low. Is it the wrong chip for this purpose?
 

hi,
What is the LED scanning rate and dwell time [ON time] for each LED.?
How many LED's.
 
No, this chip should work fine. Darlington amps can sink quite a bit of current. What frequency are you outputing to the LEDs? Can you provide a schematic of what you are trying to do?
 
It's a 7x32 (224) LED display and they are controlled separately.
The display just must not blink so the display should update like 15 times a second.
 
Last edited:
Electronics4you said:
Only one LED lights at the time, so they are only ON for 1/224 of the frame time of 1/15 s. So the LED is ON for ~300µs
Click to expand...

That ON time is short, I have found 1 to 3mSec gives a good light.

It is also possible to increase the LED's brightness by increasing the +12V supply to the LED's.
The ULN will work upto about 50V.!

Dont connect the ULN diode common pin, leave it o/c.
 
Last edited:
That all makes sense and I get a good bright light from the display at that rate, but how about large commercial LED displays. Do they increase the voltage or do they only control a small matrix with one chip?
 
Electronics4you said:
That all makes sense and I get a good bright light from the display at that rate, but how about large commercial LED displays. Do they increase the voltage or do they only control a small matrix with one chip?
Click to expand...

Allot of times they use line drivers to meet the required fanout. Bus drivers like '244 and latches like '373 are good for that.
 
How would it help to just buffer the data.
That'll just delay the display data and decrease the frame rate.
The trick must be parallel data to control a whole row (column) at a time.
It'll require more current from the PSU but not much.
Can this be right???

By simply using a PIC and make a duty cycle of etc. 1% ON (equalling 100 LEDs shifting each second) the LED will dim. The LED power must be increased to a higher level??
 
Last edited:

hi,
As you say it would not help to buffer the data, as you are already using ULN drivers.

The requirement is to get more current thru the LED while its ON, [increase the LED drive voltage]
or increase the dwell ON time for the LED.

If you examine the LED spec, it should give a pulse current rating and the permitted pulse duration.

To avoid flicker effect with the eye, the scan should be greater than 25Hz per LED.

Also check the scanning with a scope, make sure any program delays you have are not causing long delays
between the ON times for the individual LED's
 
Last edited:
Status
Not open for further replies.

Similar threads

  • Question
The very simple circuit that the boost converter voltage from 1.5 to 12 does not work in the Proteus software...why????
Replies
12
Views
3K
electronium
F
  • Solved
Unidentified IC
Replies
5
Views
1K
rjenkinsgb
G
  • Question
Why use this Capacitor
Replies
16
Views
2K
aliarifat794
A
D
  • Question
Why in Analog and Digital Filters the phase loss is not so important?
Replies
11
Views
3K
Tony Stewart
A
  • Question
Latch LS374 works for sometimes and doesn't work at all next clocks
2
Replies
26
Views
2K
Tony Stewart
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…