Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Dimmer or router speed control as resistance heater level control

  1. Nov 10, 2007 #1
    I'm trying to figure out how to safely use a dimmer or router speed controller as an inexpensive heat level control for a (nichrome coil) resistance heater.

    I've read that you can use a dimmer switch as a heat control if you don't run it at more than 80 percent of its watt rating continuously. (I'm not sure what the issues are there... why would it be rated 25% higher than its continuous power handling ability?)

    I've also seen people using a router speed control as a heat control for small heaters. My understanding is that a router speed control is similar to a dimmer, turning the power off at a certain phase in each AC half-cycle to vary the average power. (But maybe chopping the falling edge of the half cycle rather than the rising edge, or something like that...?)

    Does it make sense to use a router speed control as an average voltage control for a simple resistive load? I would guess it's designed with an inductive load in mind, but basically a switched inductive load (brushed motor rather than inductive), and I don't know whether a straight resistive load would be a problem.

    BTW, the heaters in question do have a small startup surge, with the resistance going up 10 or 15 percent in the first second or so. (But less of a surge than for a light bulb; the filament of a light bulb burns much hotter.)

    I'm interested in using a router speed control, because Harbor Freight sells a 15-amp model that goes on sale every couple of months for $10-$15; a dimmer that could handle the same amps would be much more expensive. (The usual wall dimmers are 600- or 1000-watts, and even the 1000-watt models are several times as expensive as the 600-watt model, and more expensive than the router speed control.) I'd like to be able to run at least 6 amps @ 120V through whatever I use, maybe 12 amps. (Two 720 watt coils in parallel.)

    Would that work, and would it be safe?
  2. jcsd
  3. Nov 10, 2007 #2


    User Avatar
    Science Advisor
    Gold Member

  4. Nov 10, 2007 #3
    Thanks. I think those are way out of my price range and not easy to apply to what I'm doing---heating sheet plastic with mostly IR for a very cheap DIY vacuum former.

    In general you want to control the heating rate rather than the oven temperature, and then use exposure time to control how hot the plastic actually gets given that rate of heating. For a simple system, it's easy to do a pretty good job if you get the heat level about right and then wait until the plastic sags enough to show that it's sufficiently hot and stretchy. If you actually measure temperature, you need more sophisticated sensors that can measure the temperature of the surface of the plastic itself, rather than the oven air.

    That ends up costing way more than a $15 router speed control, and would be much more complicated to set up, so if a $15 "15 amp" router speed control can actually handle a 12.5 amp resistive load, I'll be really happy. If I could safely use it up to 15 amps, that'd be even better.
  5. Nov 10, 2007 #4


    User Avatar
    Science Advisor
    Gold Member

    "If you actually measure temperature, you need more sophisticated sensors that can measure the temperature of the surface of the plastic itself, rather than the oven air."

    In applications like this one can measure the temp at an array of points (temp. map) then select one of the points as your reference; where you put the controllers probe (thermocouple).

    "That ends up costing way more than a $15 router speed control, and would be much more complicated to set up..."

    Yes but with a couple of displays (ref. temp and time) you might be able to get a more repeatable process hence saving you time in the process.
  6. Nov 25, 2009 #5
    A electric range uses an "infinite switch", a bi-metalic switch/heater combination with a shaft-mounted cam preload against the switch. Once characterized to the appliance and input voltage, they are repeatable (it's always "4" for bacon), but no temperature setting or feedback with these. However, they're cheap and can be had up to 240 VAC/ 15A
  7. Dec 1, 2009 #6
    I doubt the need to derate a light dimmer for the purpose, unless the concern is how long the heater takes to warm up and increase its resistance. Incandescent lamps initially draw roughly 10x steady-state current if switched on at the peak of a line cycle (possibly more than this for quartz lamps due to higher operating temp), and surge duration depends on the mass of the filament. The thermal time constant of the thyristor and its heatsink are much longer than the surge into a light bulb. Both dimmers and speed controllers switch power ON at a variable point in the cycle and OFF at the next zero crossing of the current.

    Speed controllers typically monitor the motor speed by sampling the back-EMF to compensate for changes in mechanical load. The resistance heater will always have zero back-EMF. The control may work anyway, but you might find that all the setpoint range is squeezed into one end of the control's rotation. I like markknorr's idea, and would suggest going to your nearest recycling center - you might be able to get a couple of free ones (in case one's a lemon).
  8. Dec 2, 2009 #7
    Thanks very much for the info, folks.

    An infinite switch will not work for my application. I didn't think it would, but I did try it, and it didn't.

    The problem is that I need to heat plastic sheets to just the right temperature, then do something nontrivial with it immediately, in real time. (Vacuum forming plastics with narrow ranges of usable temperatures---too low, and it's not stretchy enough, but much higher, it burns.)

    Seconds matter, and infinite switches have an obnoxious tendency to switch on and off at really awkward times, over a longer timescale than I can easily use. (That's mostly okay for some kinds of heaters, with fairly massive elements that smooth out the temperature changes, but it will not work for the long thin resistance wires I use.)

    The problem gets much worse when I use multiple heating elements with separate controls. Then I end up with parts of the plastic hot enough, but parts of it often not, because that element happened to switch off at an awkward time. That gets so awkward that it's easier to just flip a toggle switch on and off by hand, more quickly, to moderate the heat. (And that is a royal pain.)
  9. Dec 2, 2009 #8
    About sensors and repeatiblity...

    I do have a cheap infrared thermometer from Harbor Freight, that measures the actual temperature of the plastic plenty well, but I don't have an automatic thermostat thing set up to use it---getting the right parts to set up an infrared thermostat system is way more expensive than buying a mass market prepackaged thermometer.

    And really, using the amount of sag can work better than infrared sensing of temperature.

    There are variations from manufacturer to manufacturer and even from batch to batch of the plastic, so that the ideal forming temperature is. (Depending on how just how much recycled scrap "regrind" plastic was used in making that type or that batch, etc.)

    Once your setup is heating the plastic evenly, and at about the right rate, using the amount of sag as a measure of formability is actually more repeatable than using the temperature. It's a direct measure of how stretchy the plastic has gotten and whether it's ready to form---it automatically takes care of slight variations in the plastic's formability range.

    That's why I use the infrared thermometer to fine-tune the setup, but eyeball the sag (relative to a sightline and reference marks) to get repeatable formability.

    I keep meaning to set up a $5 laser pointer so that I can adjust the beam height under the plastic, and then just notice when the plastic breaks the beam; that should be as precise as anything but a very complicated multisensor IR thermostat system, and require less adjustment. (E.g., changing the forming temperature setting for each batch of plastic.)
  10. Dec 2, 2009 #9
    oops... that should have said "There are variations [...] from batch to batch [...] so that the ideal forming temperature is variable."
  11. Dec 2, 2009 #10
    I see... Looked at a blog that mentioned RC enthusiasts use the Harbour Freight product (43060) on the primary side of a stepdown transformer to control a stainless steel wire heater for cutting styrofoam. The device might be a lot less sophisticated than I expected. If you do use it, you might want to try putting a light bulb across the load to give you instant feedback on the power change as you adjust the control, just in case the control's response is highly nonlinear.
  12. Dec 3, 2009 #11
    NH EE,

    That post about the plug-and-play hot wire cutter power supply (using a big wall wart plugged into that Harbor Freight item) was probably from me... that ~$20 setup is about ten feet behind me as I type this. :-)

    I've never needed to crank the power way up on the hot wire cutter, and have been a bit worried about using it for the vacuum former heater---can it do near its rated amps under a purely resistive load?---which is why I came here to ask actual EE's. (I also wanted to know if I'm giving good advice to people making hot wire cutters---am I just lucky my articular speed controller hasn't burned out?)

    Your lightbulb suggestion is a good and timely one. While I was logging in here, I was thinking what a drag it is that I don't have a voltmeter I trust to measure the voltage of a funny-shaped AC waveform. Then I saw your comment.

    A lightbulb would do the absolute-value and integration thing trivially. Visual response to light inttensity is logarithmic, so it wouldn't be very precise, but it'd be lots better than nothing.

  13. Dec 3, 2009 #12
    What I got from the hot-wire and transformer post is that the controller is not putting out pulsating DC. That's one technique that has been used for monitoring the actual motor speed, by using the remnant field flux to use the motor for a tachometer between current pulses. Thus my comment that it may not be as sophisticated as I expected.

    As for supplying current to a purely resistive load, the only concern I'd have is that the current risetime when the thyristor triggers (turns on) is faster with a resistive load. Thyristors have a di/dt rating as well as an on-state current rating. Without knowing the actual component used in the speed controller, I can't say whether that's a problem or not. If the thyristor fails, it will most likely short out and give you full power regardless of control setting.

    Gut feeling is I wouldn't hesitate to try it at 6 amps. When you add the second heater, will you want a single control for both, or dual controls?

    re: light bulb - don't forget the color changes too as the RMS voltage varies, so you wouldn't be relying only on perceived brightness. Best thing about using a bulb is probably that you don't have to take your eyes off the work. If you wanted something repeatable, an average-responding voltmeter would still be useful even though it's not giving the true RMS reading. You could preset the control to give a particular meter reading based on experience. I'd use a mechanical meter rather than a digital one for this. Easier to read at a glance and quicker response to changes.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook