I have a very simple Arduino project which lights 6 leds with a 500ms delay in between.
It works fine when the LEDs are powered directly from the Arduino pro mini itself.
However I want to increase the number of LEDs in each output - a small parallel string on each.
I've tried to use a Toshiba ULN2803APG Darlington array with the output of the Arduinos.
I have common ground / common positive for the project. I have used a couple of different power supplies both 5v to the raw pin.
However I cannot get the ULN2803APG to work correctly. I have tested ULN2803APG separately using a jumper from +5v to trigger the inputs, it all works fine. However with the Arduino outputs connected, the LEDs all light up at the same time, and then become brighter one at a time, then go out.
In the attached image, circuit 2 works fine. Circuit 1 does not work.
It's your wiring (implementation).
The 2803 outputs aren't like 4000-series, 7400-series logic gates (though that is a common expectation / misunderstanding).
They are open collector, the load is placed between +V and the respective output.
The outputs don't provide a HIGH or a LOW, they make for a conduction path to Ground (Open, Closed).
In the first schematic, I have individual LEDs with a resistor from +5v into the individual ULN2803APG output pins.
I though that when the corresponding input is high ( from the Arduino output pin ) the current will flow from +5v through the resistor / LED into the ULN2803APG output pin and down to GND on pin 9 ?
Ref the 2803 possibly being counterfeit. I don't know. It was sold as Toshiba from a UK seller.
It does work normally if you just probe the input pins with +5v. I've checked that on each pin.
[edit]
Forgot to mention, I've fed 9v from bench power supply into RAW now. Also the black leads on the LED cathodes have a resistor in the leg already. I've also tested this on a standard Genuine Arduino R3 with the same results.
I've seen where those power strips (rails) do not have continuity their whole length.
Keep the connections to Ground and +5, +9 on the same half - end as it were.
(Didn't want to further confuse the issue, but I think the LEDs should run from the 9V (w/ different resistors yes) and not the ProMini's rather limited 5V.)
Well I would like to thank everyone for their help.
I've learnt some things today.
Turns out the circuit I made in the first picture works. I've had it running on USB and also from standalone 9v to Vin ( on the Geniune R3 ) and to raw on the pro mini clone.
However, the breadboards I have are warped.The chip was working in a some positions and not in others. (as suggested by larryd).
Can anyone recommend a good breadboard for the UK ? or is it better to try and fix the Chinese ones I have.
Also the solid core wire I'm using is listed 0.60 not 0.64 if that makes a difference. It does seem a tiny bit loose on the R3 headers ?
I use 0.59mm hook wire a lot, shouldn't be a problem but dupont header sockets can get stretched with lots of use, in which case I'd invest in some strips of dupont header pins and solder wires to them.