[nSLUG] [OT] Any electronic hardware wizards here?
daniel at bonmot.ca
Thu Nov 6 07:33:25 AST 2008
Funny, I was just thinking the other day how weird it was that the
"double" bed size dual control electric blanket that I bought a
quarter century ago was still working fine, and I've used every
night, every winter since then.
I've been mentally putting away the pennies for when it fails - not
the control, I don't really expect that to ever fail - but the wires
somewhere in the blanket, of the classic flexed "wire disease,"
probably where they join the three-prong connector at the foot of the
The control is pure mechanical bimetallic spring with its own tiny
heater, so that it does essentially a pulse-width modulated by room
temperature; the spring heater makes the spring cycle on and off;
when the room is cold, the spring cools off more quickly and spends
more time in "on" mode. Turning the "hotness dial" mechanically
changes how warm the bimetallic spring has to be before it clicks off.
I've marvelled my whole life that this control method, with no sensor
in the blanket or the bed, works so well.
So - starting from basics: you're sure it's the control that has failed?
If this is the case it seems to me that it should be very easy to
build an equivalent electronic one, using:
* 4.5v to 16v DC power supply (probably the most tedious part of the
* thermistor exposed to room temperature,
* a pot for the "hotness dial"
* a venerable 555 to do the timing, and
* a relay to switch the power for the blanket.
There are two tricky parts: (a) picking the thermistor that has a
range for your expected room temperature (say, from 12 to 20 Celcius)
and a "hotness dial" pot and (b) compensating for the nonlinear
output of the thermistor.
EE friends - do you have any favourite plans to share, for a
thermistor-555 pulse-width-modulation circuit?
Halifax, Nova Scotia, Canada
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the nSLUG