Take the 2-minute tour ×
Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It's 100% free, no registration required.

Many electric stovetops are operated by an infinite switch, which pulses the current on and off for intervals determined by the dial setting. The actual control is effected via a bi-metal switch that moves as it is heated. I was wondering whether the temperature of the heating element has any effect on the timing of the switch. In other words, if I put a pot of cold water on the stove, thereby lowering the resistance of the element, will that speed up the heating of the bi-metal in the switch, causing it to stay on longer? Or is the bi-metal heated by a completely separate circuit?

share|improve this question
add comment

1 Answer

Very revised answer:

Very annoying.
While my original answer to the original question was largely correct - ie the actual load in the element has no effect on timing, it was substantially wong in another area. I'll go back and edit everything in due course, but for now, original answer is left intact below.

I originally could not find a diagram of the internal connections of an "infinite switch" anywhere. Based on what is written elsewhere it sounded as if the heating coil connected across AC mains. However, this diagram which Dave found on this very interesting fire investigation site tells a different story.

More detail below, but a surprising conclusion is that
Doubling the element wattage will, for a given setting.

  • reduce power by about 30% at high heat settings,
  • reduce power output by 70% at low settings.
  • reduce power output by 50% at medium settings.

You asked about affect of load (eg pot) on element power.
Notionally pot does not load element BUT I'd expect for a given power that element would get hotter due to loss of radiation path and some convection losses and pot reflection.
If element is eg Nichrome this will not effect it. It will NOT have NTC so if it is mildly PTC (positive temperature coefficient of resistance) then res will rise with covering. Power in element = V^2/R so power will go down which will TEND to increase power level from the infinite switch as above. This should be a small effect and depends on Relement being PTC.

The above lower ower wity more Watts element effect suggest manufacturers may want to make switches to suit element powers.

The electronic circuit below dos not depend o an element in series with the element, unlike the mechanical one.


The switch's heater element is IN SERIES with the range element.When the heater is cold the switch is on and the range element will be turned on. The heating energy will be governed by the element current which will depend on the element' Wattage. So doubling the wattage doubles the current.
HOWEVER the series heater will only dissipate a small percentage of the main voltage and the voltage drop across the heater element will be proportional to current if the heater resistance remains constant. As a consequence heater watts will rise with the square of the current. So doubling element wattage increases heater power by 4 times. If on time is approximately proportional to heater power then doubling element wattage will reduce on time by a factor of 4. Off time can be expected to be close to constant as temperature at turnoff will presumably be similar and cooling time should be largely unrelated to heating rate. So, imagine that a controller was running at 66 time units on, 34 time units off. Doubling element wattage can be expected to reduce on time to about 16 time units. So duty cycle goes from 66/99 = 2/3 = 0.666 to 16/5= = 0.32 or 2:1. That sounds very good.
BUT if originally running at say 20:80 = 0.20 it would change to 5/85 = 0.06 or 30% of original.
At the other extreme, of running at 90:10 = 0.90 it would change to 70 % of original.
ie Doubling the element wattage will
reduce power by about 30% at high heat settings,
reduce power output by 70% at low settings.
reduce power output by 50% at medium settings.

This means that a higher Wattage element will reduce output for a given setting and have an increasing date of getting hot a the top end.

enter image description here


No - the temperature of the heating element or an item being heated has no effect on these controls. Here in New Zealand one name they are called is a "simmerstat" or an "energy regulator"- they will probably have other names elsewhere.

They are what is known as "open loop" controls. Only the dial setting controls the on/off ratio - the load or target object is NOT "sensed". As the dial is turned the on/off or mark-space ratio changes.

Controllers do exist that take account of the target temperature. This is usually termed a thermostat if you are heating air. These also may use a bimetallic strip but it is controlled by the target air temperature.

Here is an elecronic simmerstat control schematic. Note that they could sense load power but have chosen not to. OM1895 IC - diagram from here.

10 page application note

enter image description here

share|improve this answer
Thanks. But why, indeed, would the change in resistance not affect the rate at which the bimetallic strip heats up (albeit slightly)? –  Dave Oct 27 '11 at 17:37
As I understand what is done, the "heater" is driven by a small heating coil powered from the AC mains. While the mains MAY sag slightly when the load is switched on they are not meant to do so perceptibly. If it did affect the result it would be a weak 2nd order effect and not designable. Load current dependance coukd be added if desired but I'm not aware that it is. –  Russell McMahon Oct 27 '11 at 17:43
This schematic is way more complex than the infinite switches I've seen. All that's inside of them is a plastic cam and a few metal strips. Not even anything that might function as a heater. –  Dave Oct 27 '11 at 17:52
Oh, I see from this site (electrical-forensics.com/ElectricRanges/ElectricRange.html) that there is in fact a small heater on one of the strips. –  Dave Oct 27 '11 at 17:57
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.