Shop OBEX P1 Docs P2 Docs Learn Events
How many watts would need to be drawn by a chip in what unit time to cause fire? - Page 2 — Parallax Forums

How many watts would need to be drawn by a chip in what unit time to cause fire?

2»

Comments

  • WBA ConsultingWBA Consulting Posts: 2,934
    edited 2014-04-13 00:12
    Funny thought... I was reading the Intel spec for the Core i7 chip and it is spec'd out to pull up to 145 amps. Yes, you read that right. Idle power consumption is 12 watts, with a typical peak of 135 watts. No wonder my laptop doubles as a lapwarmer.
  • Heater.Heater. Posts: 21,230
    edited 2014-04-13 02:34
    So who is going to do the experiment ?

    We can simulate a chip with one or more surface mount resistors that approximate the correct area/volume.

    Then hit it with ever increasing blasts of current until something blows, or goes on fire.

    Years ago some guys in our lab were doing this with a via hole. They wanted to know how much current this nice big via hole could tolerate. It was smouldering away stinking out the place for quite a while. Circuit board end up all black and charred looking but I don't recall any fire or how many amps they were up to.
  • evanhevanh Posts: 15,915
    edited 2014-04-13 03:20
    The fibreglass just scorches and smokes while the soldermask flakes off and the copper separates, the circuit eventually fails. More likely the case or surrounding components will ignight rather than the PCB itself.

    EDIT: Solder can melt of course. Depends on situation as to what happens next. IC could fall off board. Shorts are very likely. Even little explosions and splatters.
  • Heater.Heater. Posts: 21,230
    edited 2014-04-13 03:24
    True. A realistic experiment calls for the circuit board to have a good coating of household dust, cat hair and the like. Like the inside of my PC.
Sign In or Register to comment.