Shop OBEX P1 Docs P2 Docs Learn Events
High current load to prop — Parallax Forums

High current load to prop

stilgarstilgar Posts: 47
edited 2012-09-06 01:11 in Propeller 1
I have a problem that I have been working on for awhile now. I am hoping that
you guys(gals) can help.

I need to control 16, 9 volt, 100mA filament lamps in a scanner type situation (back and forth).

A couple restrictions..
Max output per I/O at 15mA. (up to 4 I/O on at same time)
PCB space is limited (1.25" x 3.0" (3.175cm x 7.62cm))

I have tried several transistors, arrays, buffers, mosfets, I had thoughts about opto-couplers, but cant find any 8 channel. in order to get the required number it wont fit in the required space.

thanks
stilgar

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-04 12:52
    Have you looked at the ULN2803? I'm not sure how well it would handle incandescents, though, since their instantaneous cold turn-on current can be quite high.

    -Phil
  • stilgarstilgar Posts: 47
    edited 2012-09-04 13:19
    I tried one, but could not get it to work, Do you have a schematic that shows how to connect the ULN2803 as a simple switch, the datasheet is confusing.

    thanks
  • jmgjmg Posts: 15,185
    edited 2012-09-04 14:12
    stilgar wrote: »
    I need to control 16, 9 volt, 100mA filament lamps in a scanner type situation (back and forth).

    How many of these lamps need independent control ?
    Are you switching a 9.1V regulated supply ?

    This device has lower drop than the older darlington variant
    http://www.ti.com/product/uln2003v12

    or the TIPIC6A/B/C/259/595/596 series save pins.
  • stilgarstilgar Posts: 47
    edited 2012-09-04 14:30
    I need 16 independent control. the power unit is 2, 9 volt batterys connected in parallel.
  • Mark_TMark_T Posts: 1,981
    edited 2012-09-04 14:36
    stilgar wrote: »
    I tried one, but could not get it to work, Do you have a schematic that shows how to connect the ULN2803 as a simple switch, the datasheet is confusing.

    thanks

    ground to ground, inputs to microcontroller, outputs to bulbs, other side of bulbs to +9V. You will lose some brightness due to the darlingtons.
  • jmgjmg Posts: 15,185
    edited 2012-09-04 15:05
    stilgar wrote: »
    I need 16 independent control. the power unit is 2, 9 volt batterys connected in parallel.

    Are these serious batteries, or the jellybean PP9 ? - How long do you expect them to last ?
    For independent control, the serial Power MOS devices in the TIPIC6A/B/C/259/595/596 series, let you choose Current/Price and they have low losses, and use few host pins.

    Check your lamp peak current, with a new battery.
  • Mark_TMark_T Posts: 1,981
    edited 2012-09-04 16:58
    jmg wrote: »
    Are these serious batteries, or the jellybean PP9 ?

    I think you mean PP3, PP9's are quite large (they would not be nice to drop on your toes!). PP6's are intermediate in size.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2012-09-04 17:22
    stilgar wrote: »
    I need 16 independent control. the power unit is 2, 9 volt batterys connected in parallel.
    This sounds all wrong. First off you are using incandescent lamps when LEDs can be far brighter, more reliable, and easier to drive. Second, not only are you using PP3s, the most inefficient battery on the earth, but you are connecting them in parallel. Don't use 9V batteries, at least use six AA cells as they are far more effective, cheaper, and last much longer. Why the lamps? Is this just a quick "muck around"?

    BTW, 100ma is NOT high current
  • jmgjmg Posts: 15,185
    edited 2012-09-04 17:52
    Mark_T wrote: »
    I think you mean PP3, PP9's are quite large (they would not be nice to drop on your toes!). PP6's are intermediate in size.

    Oops, yes, the PP3, I try to avoid them like the plague, so am rusty on the exact name..
    I have seen long life Lithium batteries claiming 1200mAH in a PP3 case... but they are not cheap.
  • kwinnkwinn Posts: 8,697
    edited 2012-09-04 21:24
    I would suggest looking at the TPIC6595. Two dip chips would fit in the space you have, typical Rds on is 1.3 - 2 ohms, can easily handle the current, and they can be controlled with 4 prop pins.
  • Mark_TMark_T Posts: 1,981
    edited 2012-09-05 15:05
    kwinn wrote: »
    I would suggest looking at the TPIC6595. Two dip chips would fit in the space you have, typical Rds on is 1.3 - 2 ohms, can easily handle the current, and they can be controlled with 4 prop pins.

    At standard operating temperatures (around 2500K) tungsten filaments will have about 12 to 14 times the resistance they have when at room temperature - thus the inrush current when switching on from cold will be about 12 to 14 times the steady-state operating current. So for a 100mA bulb that implies 1.2 to 1.4A at switch on (briefly).

    The TPIC6595 has a pulsed current maximum rating of 1.5A. That's rather close for comfort, I'd change the wording to "can just handle the current" rather than "easily handle the current". Also you might need to consider decoupling capacitors as switching many bulbs simultaneously will mean large current spikes that those batteries might fail to provide, depending on the internal resistance.

    For a high-temperature bulb such as tungsten-halogen, the ratio from hot to cold might be more like 20 - see the table on this page: http://hypertextbook.com/facts/2004/DeannaStewart.shtml
  • kwinnkwinn Posts: 8,697
    edited 2012-09-06 01:11
    Mark_T wrote: »
    At standard operating temperatures (around 2500K) tungsten filaments will have about 12 to 14 times the resistance they have when at room temperature - thus the inrush current when switching on from cold will be about 12 to 14 times the steady-state operating current. So for a 100mA bulb that implies 1.2 to 1.4A at switch on (briefly).

    The TPIC6595 has a pulsed current maximum rating of 1.5A. That's rather close for comfort, I'd change the wording to "can just handle the current" rather than "easily handle the current". Also you might need to consider decoupling capacitors as switching many bulbs simultaneously will mean large current spikes that those batteries might fail to provide, depending on the internal resistance.

    For a high-temperature bulb such as tungsten-halogen, the ratio from hot to cold might be more like 20 - see the table on this page: http://hypertextbook.com/facts/2004/DeannaStewart.shtml

    All good points to consider. The peak current is is very brief, is reduced by the internal resistance of the output transistors, and can be reduced further by adding a small inductor in series. I have used the TPIC for driving a bulb filament (12V 0.1A) in the past and the only time one failed was when the bulb burned out.
Sign In or Register to comment.