Actions
Engagement
  • Author Author: pchan
  • Date Created: Date Created
  • Last Updated Last Updated:
  • Views 1475 views
  • Likes 5 likes
  • Comments 8 comments
Related
Recommended

Ben Heck's Magic Smoke Preventor Episode -- Episode 249

image

element14's The Ben Heck Show

Join the Ben Heck team every week for amazing hacks! Watch them build and mod community-inspired projects using electronics!

Back to The Ben Heck Show homepage image

Connect with TBHS
Featured Bonus Materials
See All Episodes

 

 

Ben and Felix produce a laser guided infra-red grid array sensor, connected via i2c to a micro controller with mechanical relays to disconnect the power to a device that causes fire! They integrate a Hitatchi HDD44790 based LCD display to create a menu and interface for calibration with an enclosure designed with Autodesk Fusion 360. If advanced technology is almost distinguishable from magic then an electronics component that sets itself on fire has “let the magic blue smoke out” and killed the magic.

  • in reply to beacon_dave

    According to Ben: "the trigger point was selectable. Either any one cell exceeds the trigger temp, or the average of all cells do."

  • Given the number of fires being reported as having being caused by charging Li-Ion batteries then this project could perhaps be adapted to monitor designated charging areas. For lead acid batteries, perhaps it could additionally be coupled with a hydrogen gas detector to additionally monitor for excessive gassing during the charging process.

     

    Other areas could be for monitoring electrical distribution boards to determine if any fuses are starting to run hot indicating a circuit overload scenario.

     

    Could also be used for combined fire and security applications giving a better indication of how a fire is spreading and where any people still in the building are, assisting evacuation.

  • in reply to rouverius

    I'm not sure that the Omron D6T would be the best sensor for that. If you look at the specification sheet:

    https://www.omron.com/ecb/products/pdf/en-d6t.pdf

    then it suggests that its object detection temperature range is 5-50degreesC along with no icing or condensation. Water boiling over will be up in the 100degreesC range, and generally creates condensation in the process.

     

    The specification sheet and application note

    https://www.omron.com/ecb/products/sensor/special/mems/pdf/AN-D6T-01EN_r2.pdf

    are a bit vague as to what happens outside of that temperature range. The application note suggests each cell outputs 16bit signed data in tenths of a degree however I suspect the accuracy will be significantly less than the +/- 1.5degreesC and +/-3degreesC stated in the specification sheet.

     

    Everything around the cooking pot area is likely to be above 50degreesC though due to the heat source and heat conduction/convection so I'm not sure that you would be able to get a clear indication with the sensor that boiling over is occurring as opposed to just some steam venting off or things getting hot.

     

    Perhaps a digital camera would be a better choice here as it would be able to compare two images for a substantial difference as the pot boils over and the foam runs down the side. Would need to exclude the normal motion of the pot lid and steam venting however.

  • They say a watched pot never boils. But what about a pot monitored by thermal sensor?

    Could it be modified to work in the kitchen to alert against boil overs?

  • This would be great for thermal protection for your computer just point it at your computers exaust!

  • A little bit more involved, but a good upgrade/addition would be a focusing lens for the sensor attached mechanically to the laser pointers.

    As you focuses the sensor into the target area, the laser points expand/contract accordingly to give you an accurate reference.

  • Great Episode.

     

    Who doesn't like watching resistors die with lots of magic smoke.

     

    DAB

  • How many cells of the Omron D6T grid array were showing a significant temperature increase in those resistor tests ?

     

    Just wondering how much heat was being detected in the smoke itself (e.g. several cells activated), or if it was all from the resistor itself (single cell activated).

     

    This could have a significant impact on the placement of the sensor for it to be effective, as you could have scenarios where there is line of sight to loads of smoke, but no line of sight to the source of the heat (e.g. components on an inverted circuit board.) 

     

    In addition, if the sensor was placed at a greater distance then you may need to switch to a 'any single cell' detection mode as it may otherwise take a substantial increase in temperature for the average across all 16 cells to be high enough to trip the alarm threshold.