Basic "Heat Safety" Alarm

      Basic "Heat Safety" Alarm

      Following a post by @KineticIsEpic in this thread about a kigger's thermal safety in harsh summer weather conditions, I thought I would dig out my older creative self and design a battery-powered thermal alarm with a single probe to be placed up against the area of the body most likely to overheat. Here is my schematic and an example working PCB render of it:
      thermalalarm.png

      This circuit is simply a CA3140E single operational-amplifier reconfigured as a voltage comparator in which it turns its output on when the voltage on one input exceeds another (red probe is greater than the blue one).

      An on-board variable resistor is present to allow adjustment.

      Notes
      • Power supply is 4.2-4.5v (or three AA/AAA batteries, rechargeables will work but will offer shorter runtime)
      • Components have been positioned to allow the op-amp chip to be soldered directly to the board without the need of a socket, reducing the overall surface profile of the circuit
      • Standby "off / non-triggered" current is around 1.09mA; "on / triggered" current is around 1.49mA without load
      • Though optional, a back-EMF suppression diode (D1) is included to protect the switching transistor from reverse voltages created from inductive components when turning on-to-off; D1 is not required for resistive components (i.e. LEDs)
      • An extra hole has been included on the PCB design to allow higher power TO220-standard transistors (e.g. TIP41A) if preferred
      At a later date, I will attempt to design a more sophisticated system using a PICAXE 08M programmable chip which can allow a handler to wirelessly monitor (more accurately) up to 3 points of a kigger at the same time.
      We have an electronics engineer in the house! This looks really nice, however the reason I said my idea was bad is because, after some discussion, I've concluded that mask temp (or even a reading from the skin) is probably useless for accurately monitoring body overheating; the only way to do that well would to have a thermometer in you mouth or similar. Still worth a shot considering that I don't have any evidence either way, both the original idea and the viability of it were 100% speculation.
      Your signature here

      KineticIsEpic wrote:

      We have an electronics engineer in the house! This looks really nice, however the reason I said my idea was bad is because, after some discussion, I've concluded that mask temp (or even a reading from the skin) is probably useless for accurately monitoring body overheating; the only way to do that well would to have a thermometer in you mouth or similar. Still worth a shot considering that I don't have any evidence either way, both the original idea and the viability of it were 100% speculation.

      I don't think the idea, on the whole, is "bad" at all and I don't see "100%" speculation behind your drawing board, I just don't think we need the granularity you're thinking of; we're not really after numbers, are we? We're just looking for utmost basic differential measurements. Digital may be the standard today, but even obsolete analogue circuitry still have their place and there are no substitutes for them.

      I wouldn't put any sensor in the mouth let alone near it, I'd say just above the eyebrows because that is normally where (from what I believe) most of one's body heat is given off and the top of the mask is where all the heat is likely to accumulate; sensor positioning may prove difficult with some skinsuits as some have whole-face apertures whereas some just have the eyes and mouth (or even nose), which would be a better choice.
      Made a few corrections to the design:
      • Component referencing cleaned up
      • Component connection: the variable resistor (VR1) in the original PCB design had all three pins connected when the schematic only designated two
      • Increased the value of the fixed resistors to reduce energy consumption slightly and improve accuracy slightly
      • External component connections (C1-C3) are now on one side of the board
      • Modified PCB tracking to accomodate more sizes of calibration resistor (VR1)
      • Amended the silk screen (white) to indicate and illustrate the required use of right-angle connectors
      thermalalarm-rev-b.png
      Took me a while, but I have managed to redesign the circuit diagram and board layout to address a previously-unidentified issue of rapid-switching:

      Resistors 4 and 5 should result in a temperature threshold of around 32-36C with this reduced when the output is on.

      FanCTRLv2.png

      Design Notes
      • To simplify board tracking, the thermal resistor has been reconnected to "pull down" the voltage applied to one input as the temperature it is exposed to increases; the input to determine the on-off threshold, whilst the output is on, is now "tugged" upwards (which results in the thermal threshold being lowered) to prevent a rapid-switching loop caused by power supply voltage fluctuations as a result of the controlled load (max. 1A without a heatsink).
      • Also, the board is now slightly smaller, with components now more tightly-packed.
      • The NPN 2N2222A (0.8A/50v) transistor has been replaced with a PNP TIP42A* 6A/60v power transistor (often paired with a TIP41A on some lower-power audio amplifiers) aided by an NPN BC547B with a 1K resistor to stop a short circuit effect.
      • In the graphic attached to this post, the back-EMF suppresion diode (D1) is shown off-board, but I have manually added two extra solder points adjacent to the output connector to allow for an on-board vertically-mounted one if this is preferred. Back-EMF suppression diodes must be connected in reverse-polarity.
      • R6 (2.2M) had to be moved to the board's backside due to the locations of the output pin and the input pin that requires "tugging", this is not likely to be the case if the CA3140E is not soldered directly to the board and a push-in socket is used.
      • Power consumption whilst idle at 6v*: 1.06mA
      Disclaimer: this circuit was intentionally designed without a power indicator LED for power-saving reasons.

      * For improved reliability and reduced heat emission.
      ** USB-level voltages (~4.8-5.5v) should still work.

      I am making this design available for anyone's use under a CC BY-NC-ND 4.0 License.