Shop OBEX P1 Docs P2 Docs Learn Events
Thermal info on P8X32A - how warm does a propeller chip get? — Parallax Forums

Thermal info on P8X32A - how warm does a propeller chip get?

WossnameWossname Posts: 174
edited 2012-02-06 17:31 in Propeller 1
My Propeller has been powered up and running for an hour or so and I just noticed that it feels slightly warm to the touch. Is that normal? Nothing else on the (perf) board is warm (apart from the voltage regulators but that's ok since they are running off 12v input).

All voltages are OK into the chip, only one cog is running and it's driving 4 LEDs through 470R resistors off 4 separate pins in a slow loop (1/8 second per loop).
I've had a look in the datasheet for the Prop chip and it seems only to include environmental temperature and how it affects the chip. I've never noticed the chip heating up before and I've made several propeller proto-boards before now.

I'd like to know how warm does this chip normally get just under it's own steam (so to speak) during normal operation.
I'd estimate the chip is probably 5 or 6 degrees C warmer than my cold hands (it's snowing outside! :) ).

I'm not overly concerned about this but I'm curious nevertheless.

Comments

  • Dr_AculaDr_Acula Posts: 5,484
    edited 2012-02-06 03:19
    I suspect it is because it is driving leds. There is an internal resistance value and some clever person here will probably chime in with what that value is and thus you can work out V=IR for those pins W=IV and work out actual milliwatts. But ballpark figure driving leds I'd expect a little warmth. Maybe take off the leds and see what temp it runs at?
  • LeonLeon Posts: 7,620
    edited 2012-02-06 03:20
    Unused inputs left floating can increase current consumption considerably. What have you done with your unused pins?
  • WossnameWossname Posts: 174
    edited 2012-02-06 04:58
    Yeah they are all floating since this is a work-in-progress design. I didn't know that about floating inputs being a cause of increased power drain, interesting.

    So ideally you'd pull an unused input pin low (or high??) or tie it directly to one of the supply plains? Or set them as outputs even.

    Hmm.
  • LeonLeon Posts: 7,620
    edited 2012-02-06 05:01
    Try measuring the current consumption as it stands, and with the unused pins configured as outputs, and see if there is a difference. There should be, but I've never actually tried it for myself.

    With the unused pins as inputs, factors such as board cleanliness and humidity will have an unpredictable effect, which isn't desirable.
  • RaymanRayman Posts: 14,845
    edited 2012-02-06 10:05
    I've only seen a Prop get noticeably hot when something was very, very wrong.
    You might want to make sure you have all Vdd and Gnd pins connected and nothing is shorted out somewhere.
    The datasheet says 1W max power dissapation. If the prop were thermally isolated, it could conceivable get a little hot..
    The Prop itself draw drawas about 10mA for one cog at 3.3V. So, that's about 33 mW.
    The outputs have about 30 Ohm self-resistance and can output up to 300mA. That would give about 2.7W.
    That is actually enough to be noticeable, now that I think about it...
  • Dr_AculaDr_Acula Posts: 5,484
    edited 2012-02-06 14:44
    Ballpark calcs -led drops 2.2V so that leaves 1.1V
    27 ohm internal resistance, 470R so V=IR, I=V/R = 1.1/497 = roughly 2mA

    W=IV = about 2mW dissipated in the chip per pin driving a led.

    x4 leds is about 8-9mW. Not enough to get warm I'd guess. And less than the current consumption of the chip itself 33mW as calculated by Rayman above.

    Is this a homebrew design or one of the pre-made boards?

    If it is homebrew, are there bypass caps etc?
  • RaymanRayman Posts: 14,845
    edited 2012-02-06 16:29
    I don't quite follow your calculation Dr_Acula....
    Most LEDs are full brightness at 20 mA. Four would be 80 mA.

    Ooops, never mind, I see the 470 Ohms in the top post now...
  • Dr_AculaDr_Acula Posts: 5,484
    edited 2012-02-06 17:31
    ... which deepens the mystery of why this chip is warm. I'm wondering if a photo might help us?
Sign In or Register to comment.