Shop OBEX P1 Docs P2 Docs Learn Events
Propeller project board and ULN2803 — Parallax Forums

Propeller project board and ULN2803

LoopyonionLoopyonion Posts: 24
edited 2013-07-30 01:04 in Propeller 1
Hi,

Im trying out the ULN2803 in respect to controlling an LCD counter. Its basically a 12v powered counter, but has 3 x 5V outputs of which are grounded in order to increment/decrement/reset the number.

The spin code is a simple OUTA command, and written always ON, on pin 0.

If i ground the counter 5v's to the negetive supply, then it functions perfectly, if i run the program, and dab the 5v counter input to the activated ULN2803 output, then the counter will increment. However.. when i dab the decrement, or reset to the same pin, nothing. Yet it works to ground.

My hunch, the 0V sink the ULN2803 is providing, is not quite an 0 ohms resistance, hence the counter not acting on it. The schematic of the counter must require a solid 0V, close to 0 ohms to see a signal.

Please forgive my lamens terms, and drawing below. .

board overview.jpg


The reason i am going with the ULN2803 is space required on my board will be limited, (as ever), and i will be requiring to ground out 5v incoming feeds in order to send information from the prop, to other controllers.

Is there a way, other than a relay, which would give a closer 0V 0 ohms?

When i set the mutimter to continuity, and place it between the ground, and activated ULN2803 output, its not near 0 ohms, and no continuity buzzer from mr Fluke. If this reading is required, i can add it as soon as it stops chucking it down with rain lol...

the load itself is near 25mA, measured with a fluke multimeter, when the counters 5v input is grounded.

Maybe I cant see the wood for the trees, and have missed something obvious :)

ps, the +ve on the ULN2803 is not connected, as they are non inductive loads.
1024 x 851 - 158K

Comments

  • phatallicaphatallica Posts: 64
    edited 2013-07-29 07:43
    1) It is not a good idea to try to measure continuity or resistance on a live (powered) circuit. You will not get a good reading and you may damage the circuit, the multimeter, or both.

    2) Please provide information regarding the counter, so we can see the specification for the inputs.

    3) With low current (~25mA), the ULN2803 output should pull down to about 0.7V. Since this is a darlington transistor output, a lower output could be obtained with a single NPN transistor or even a logic-level MOSFET. However, it is not yet clear that this is the problem. The counter specification will help to confirm.

    ~ph
  • JonnyMacJonnyMac Posts: 9,107
    edited 2013-07-29 08:25
    Keep in mind that you could be dropping up to 1.6V across the outputs of the ULN2803 (depends on current); this suggests that you'll want to bump your supply or -- as phatallica points out -- use an n-channel mosfet. There is a chip called the TPIC6A595 that acts like a 74x595 shift register but has n-channel outputs good for 350mA each (with all running). That would cut your pin count do everything in one package.

    http://www.ti.com/lit/ds/symlink/tpic6a595.pdf
  • AribaAriba Posts: 2,690
    edited 2013-07-29 10:35
    Have you connected the Ground with the counter:
    board overview.jpg


    If the counter input has an input threshold voltage under 3V, and a pullup resistor greater 2 kOhm, then you don't need any level shifter, just connect the Prop output the the counter input (and the Ground line too!).

    Andy
    1024 x 851 - 95K
  • LoopyonionLoopyonion Posts: 24
    edited 2013-07-29 11:50
    I see the hole in my plan:-

    JonnyMac - the key in your comment, is logic level. This, in ignorance, I didn't think of. I assumed that the GND was GND, but the counter ( of which there's no spec/docs - god bless eBay lol) requires this.

    I'm going to look into MOSFET like the array JM linked to, or experiment with individual ones.

    The counter ran previously on a MOSFET shield for arduino, so I know it works.

    And i did try the GND to both supply etc, and it wasn't that. I've run small relays off the ULN2803 and all outputs function.

    This has prompted me to leave the counter for now, and run an larger 8 segment led off it.

    Thanks to all for your help ;)
  • JonnyMacJonnyMac Posts: 9,107
    edited 2013-07-29 12:15
    As another pointed out, you could get more/better help if you provided more details -- perhaps a link to the board you're using.
  • LoopyonionLoopyonion Posts: 24
    edited 2013-07-30 01:04
    Hi,

    Thanks again for the help.. I cant give any details on the counter, as is an ebay item ( from china) that is simply a motorcycle gear indicator. I just needed to count 6 stages of an operation, and a reset. This was about 3£, and did the trick with the arduino. ( plus its waterproof lol)..

    The indicator was designed to use magnetic reed switched inputs, and on my last project, i used an arduino mosfet shield to switch it. Lessons learned :- not all fet/mosfet do the same tasks. Obviously, i need to look more closely at the individual specs of components. Foolishly assuming the ULN2803 would provide the correct ground needed for the counter.

    The only bit of data that i measured was the 10ms pulse time that the counter needed to see, in order to read an input, and the minimal mA it used while this happened.

    For now, i have shelved this part, and gone with my larger 8 segment LED, this works well with the ULN2803.

    However I do need to look at grounding another 5v supply with the prop, to signal another controller, and assume that i can use the 5v to 3.3v method, then output the prop pin low to signal? Obviously, i need to look at the amperage involved before testing this, i will get the appropriate information, and come back.

    Ariba:- yes did try that also :)
Sign In or Register to comment.