I've got some 4KW 240 volt lamp dimmers from China.
What I want is a cheap electric range current regulator. (For safety I plan to replace the potentiometers with switched ones and add dual pole contactor - so "off" really is off)
The controls I'm replacing use a bimetallic switch with an extra contact to switch off both sides of the 240 line in the off position via a cam on the rotating part.
It would be cool and useful if I could also add a dual color LED to show the approximate power setting from across the room. My only idea so far is to treat it like a resistance divider with the common leg of the LED tied to the junction between the dimmer and heating element connection. I would use capacitive reactance to lower the voltage and a diode so the LED isn't reverse biased.
Presumably this should give a relatively smooth transition from green to red as the control calls for more power.
Does this sound like it aught to work? Is there a better way?
(Horsing the range in and out is no fun. Controls are due to arrive today..)
More about LEDs. From the control's point of view, a heating element is just a load resistor.
Specifically I'm wondering what effect, if any, the steep rise of the triac firing will have on the capacitor used to limit current/voltage to the LED, or any effects from putting what amounts to a small (.25uf) cap across the load and control (would that influence the firing angle?)
Note that the heaters may not survive continuous operation and may require a maximum duty cycle for safe operation. You have to make sure to not exceed the heaters' ratings, and if the original controls imposed a maximum duty cycle - so must yours. If the original controls had another means of overtemperature protection, e.g. a thermostat, then you have to make sure that this protection still works in your setup.
Ratings / derating of the lamp dimmers. There may be some numbers printed on them, and they may indicate a rating of 4 kW, but whether these controls can actually handle that power - and whether they can handle it for any extended period of time - is not a given.
Any number advertized in bold letters is likely to be overstated by a factor of 2 at the least. Besides, you have not made clear, in what temperature range these ratings are supposed to apply. If that's not given, then most likely the ratings only apply at a nominal room
that the controls won't handle 4 kW for hours in the thermal environment of a stove's control bay. They may handle 1 or 2 kW, but for anything higher you'd better pay close attention to the environmental specification of these controls. And since most specifications are egregious lies anyway - better open one of them and have a look at the triac, its thermal management and its overtemperature protection. Do a quick datasheet sanity-check of the triac and a similarly sized heatsink, just to be sure. Remember, since you probably imported them yourself, their safe operation will be your responsibility.
Use of capacitors to "feed" LEDs from mains, particularly from triac phase-controlled mains. This is a horribly bad idea that smells of an electrical fire. Forget about is as quickly as you can.
Capacitors, even "X" rated ones, are usually not made to withstand continuously pulsed loads. "Normal" capacitors won't even withstand a typical mains surge, which can have 2 to 4 kV. Applying a pulse train from a low-impedance source (mains) via a low-impedance switch (triac) to a low-impedance load (capacitor during the pulse risetime) will blast that load (capacitor + LED) with sharp current spikes that can have peaks as high as hundreds of amps (for microseconds). This will fry the LEDs really fast, and it will also fry the capacitor after some unspecified time unless the cap is rated for such continuous impulse loads. Failing caps connected to mains with hardly any current limiting tend to short out and go up in flames.
If you absolutely must use capacitors, then make sure that you: a) use capacitors with at least an "X1" rating for 240 V mains and with
b) use additional resistors in series with the caps c) size the resistors so that the current spike, when switching the peak of the sinewave into a discharged capacitor, will be limited to a reasonable value - preferably below 100 mA. In practice this means a resistor of at least 3.5 k resistance - better 5 k. d) size the cap to limit the RMS current to a reasonable value - preferably at 10 mA or below. e) use at least 2 resistors in series for single fault tolerance f) size the resistors such that they can withstand at least 1 kV pulses, for each resistor (assuming you have 2 in series) g) don't underestimate power dissipation in resistors (ca. 0.5 W) h) don't forget the freewheeling diode across the LED to protect it from reverse voltage.
Miscellaneous points to consider
Isolate the LEDs properly. A mains-operated LED has no business "sticking out" accessible to people. The LED's plastic encapsulation is not a safety insulation either. It can break off and it's not rated for mains surges. Put the LEDs into properly insulated holders intended for mains operation (like e.g. those used for neon lamps).
Make sure that your potentiometers are properly isolated too - and that their isolation is rated for mains operation. For a safety grounded metal case that also implies at least a BASIC insulation from potentiometer circuits to the case.
Isolate your wiring properly. That includes appropriate temperature ratings of the wiring isolation used! Isolate all auxiliary circuits (relays, capacitors, resistors, circuit board if any) properly too. That also includes clearances, creepage distances and choice of materials with respect to flammability and operating temperature. Given your intended environment, that also includes appropriate placement of those auxiliary components with respect to heat sources.
Always remember that it's you who is responsible for your (and your family's) safety as far as any electrical work is concerned.
Additionally, LEDs driven from half-wave-rectified mains can flicker rather annoyingly. Apparently the human eye is sensitive to the duty cycle and the risetime of light pulses to some extent, so LEDs that light up and extinguish essentially instantly flicker more than e.g. discharge lamps that have more significant time constants. Unlike the points noted above, this is obviously not safety-relevant, but you may still want to avoid the flicker anyway.
If you don't want the flicker, consider adding another diode and an RC circuit to feed the LED with a more steady DC current. Note that since the LED is a current-driven device, the RC circuit will become a "CR" circuit. It is "inverted" - with the R in series with the LED and the C "feeding" the R+LED combination. This is opposite to a "normal" RC circuit where the C is connected in parallel to the (voltage-driven) load.
The lamp dimmers that you intend to use as heater controls may have fuse ratings specified. These may sometimes be denoted with a "fuse" symbol and an amperage rating next to it (like e.g. "-[====]- 16A"). The fuse ratings indicate that the control can only be used safely on a circuit that is fused with a fuse of the rating indicated. Since the fuses (or circuit breakers) in your house wiring are likely to have higher ratings than the dimmer controls call for, you may have to install appropriate fuses in series with these dimmers in order to satisfy their specified fuse ratings. If those dimmers already have internal fuses installed, then they are less likely to specify requirements for additional fuses. But if they neither have fuses of their own nor specify fuse ratings for the installation, then their maker was probably less concerned about safety. In this case, it's probably prudent to install fuses based on the specified maximum power ratings. In any case, always install fuses with interrupt ratings appropriate for the circuits they are in (like HRC fuses for mains circuits).
In the US, a standard electric stove circuit is 240 V, 50 A (12 kW). Most houses will have a thermal-magnetic circuit breaker for this, but some older houses (before roughly the early 1960s) may still have fuses.
The original poster might be better served by using thermistors or thermocouples at each surface element to drive a comparator that drives the red-green LEDs, and a regular transformer power supply to power the comparators and LEDs.
P.S. Electronic solutions are all well and good, but for the hostile thermal environment of a stove control bay I think something simpler should be better. Even transformers might end up operating right at their thermal cutout limits, without even being loaded much.
My preference would be 2 neon lamps, one orange and one green (actually the green ones are with mercury vapor and phosphor coating but they look just like neon lamps). They are available with identical sizes and similar brightness levels. Here's an example:
Connect current limiting resistors as indicated, making sure to use at least 2 in series for surge and single fault tolerance, then the lamps can be connected across the dimmer (green) and across the heating element (normal neon) just like the OP intended to connect the LEDs. That should be simple and provide a reasonable degree of passive safety (it's difficult to induce a fault that would take out the lamp circuit in an unsafe way).
I do appreciate your erring on the side of caution, and the time and effort you spent. but... This isn't my first foray into electronics and I've had one 5 amp SSR with slow pulse width modulation working one element and another on a 10 A SSR with a small programmed controller chip. With tricolor LEDs (and working off a low voltage supply)
I was looking for something I could implement with a little less effort.
The 4 KW controller looks like it may actually do 4 KW - heat sink and aluminum chassis and the elements I'm intending to use them are 800 watts and 600 watts.
As for running LEDs with a diode, cap and small surge limiting resistor - been doing that for some time now. The circuit is in the Siemens Optoelectronics Data Book 1990 App note 6.
Safety is my primary concern, and my designs do reflect this. "This ain't my first rodeo" is the idiom we use.
Ah, alright then. You can't tell experience from a short usenet post, and since high temperatures, electronics, and surges don't mix well, I thought it better to call your attention to some caveats.
Hey, at least some of that info may turn up useful for some future novices searching or reading some usenet archive somewhere :)
If you want a simple solution, why not 2 "neon" lamps with different colors? Orange and green are common, and some other colors seem to exist too (a photo on the the wiki page shows a blue one). It's hard to beat resistors in terms of simplicity.
Or maybe not. The voltage has to exceed the firing threshold of the gas, but that still may not be a bad idea. One neon on for minimum power, two with some overlap for the midrange and the other on for maximum power.
With my IC controller, the leds I have come on red when on and green when off and can see the relative pulse width from anywhere in the room. For maximum power there is no proportioning and I turn on the red and blue together for magenta.