2 rectifier diodes in parallel to dim a 120v 15A heater?

[Raj]

i just remember reading an article about using non contact ir paired up with triac and microcontroler (microcontroller isolated via opto coupler triac) as thermostat

could easy write a program if you want one.

[Jay112]

I got a 20A zero-crossing SSR, and I tested it for my first time today. Everything works correctly when I have a 60W light bulb hooked up to it: I can turn the bulb on or off, and have it flash at different frequencies with PWM.

But when I hook up the IR heater to it the heater never turns on.

When I remove the rectifier diodes (the ones we installed earlier in this thread) that were being used to dim the heater down to about 50%, then the heater works as expected with the SSR: I can have it turn on and off, and anywhere in-between. When I put the diodes back in, the heater doesn't work with the SSR. If I have the diodes in and I skip the SSR and plug the heater directly into the mains, then it works as expected with the diodes dimming it about 50%

Does anyone know why the heater won't work with the SSR when the rectifier diodes are being used?

This is the schematic I tried to draw. There wasn't an SSR symbol available and since I didn't know what to use I just used a symbol for a switch.

[Andy Watson]

If by "SCR" you mean a thyristor (as opposed to a triac), the clue is in the name; Silicon Controlled Rectifier.

[Jay112]

If by "SCR" you mean a thyristor (as opposed to a triac), the clue is in the name; Silicon Controlled Rectifier.

Thanks for responding, but there are no SCRs in my circuit. Just a zero-crossing SSR, a half wave rectifier diode, and the resistive load of the heater.

[Circlotron]

SSR = solid state relay.
Probably the SSR needs to see both half cycles for it's zero crossing detect circuitry and other internal low power stuff to operate. The diode is cutting off one half cycle so it is going strange internally. Does anything change if you reverse the diode?

Edit -> did you mean it worked okay with the diode in circuit on the globe but not the heater?

[Andy Watson]

Thanks for responding, but there are no SCRs in my circuit. Just a zero-crossing SSR,

Oops, sorry, my mistake (the afluence of inccohol is strong tonight!). But, a "zero-crossing" SSR will require a zero-crossing - which the series diode prevents.

[Jay112]

Thanks guys!

I just tried reversing the diode, and it still didn't work.

Also I hadn't tried the diode + lamp before, and I just tried that now. It had the same result as the heater: the lamp doesn't turn on at all when the diode is in the circuit, no matter which direction the diode is pointing.

It sounds like you guys got it right that the SSR needs to see the voltage crossing past zero. I wonder if there's any way to make a tiny "bounce" past zero, even with the rectifier diode in place. Any ideas?

[Andy Watson]

I wonder if there's any way to make a tiny "bounce" past zero, even with the rectifier diode in place. Any ideas?

You might be able to fool the SSR by arranging for some negligible leakage across the diode - perhaps a low wattage lamp? It doesn't need to be much - just enough to the trip the zero-crossing detection in the SSR.

[Ian.M]

You need enough of an unrectified load on the SSR for it to see a full AC waveform. It would *probably* work if you added an 18K 1W resistor from the SSR switched Live output to Neutral, unless the SSR requires a minimum load current >6.6mA.

However for the reasons other posters have listed above, diodes to reduce the power of a high current load are a dumb idea that can get you in trouble with your electricity supply company, so I wouldn't recommend leaving the diodes for half power in there.

If you are implementing cycle skipping control, you need a line frequency reference to your control circuit so you can avoid skipping half cycles, and only skip whole cycles, also to avoid DC imbalance.  One approach is to use the line frequency input to clock the PWM through a D type flipflop before feeding it to the control terminals of your zero crossing switching  SSR.

You can derive a suitable line frequency signal by tapping one of the AC outputs from the transformer to the bridge rectifier of a conventional linear PSU, and feeding it through a resistor with a Zener clamp (to 0V) to reduce it to your desired logic level.  That gives you a somewhat assymetric line frequency near squarewave, with fairly poor rise and fall times.  Clean it up with a Schmitt trigger input buffer before you do anything else with it.

[Jay112]

Thanks for the help and ideas, everyone!

Just to see if it would work, I just tested your ideas of using a small lamp/resistor across the SSR. I found a 3-watt 74 Ohm resistor, and it worked! I calculated that it was probably drawing about 1.2mA of current.

@Ian.M, I appreciate your advice about safety, and my ultimate goal is to be able to recreate the original circuits that came with these heaters but which have broken. Ultimately I'm hoping to put together a board that uses a zero-crossing switch (with adjustable pot) to fire a triac that powers the heater.