Controlling AC Loads

[adaminc]

Hello all,

First, I shall tell you what I want to be able to do, essentially just a basic on/off and "dimming" circuit, I will eventually be controlling it with a microcontroller.

I am having some problems figuring out how to easily control the power going to any sort of load that would be on a 120VAC 60Hz circuit. I know you can use the tried and true method of using TRIAC's and burst or phase control. But I remember reading that these methods are really only good for resistive loads that naturally smooth out the power they are receiving. I could be wrong though. Please let me know if I am.

I have thought about reverse phase switching with FETs or IGBTs, but I do not know much about it, whether or not it is suitable for inductive loads. Then there is the method of converting AC to DC then inverting back to AC, which I would rather not do as its inefficient.

Any ideas would be appreciated.

 

[crutschow]

If you're not using the "dimming" circuit to dim a resistive lamp than what are you trying to dim?

 

[adaminc]

Not really sure, this is more of a general purpose works in all situations kinda thing. I want to make 1 design, then use it all over the house for automation.

 

[Analog]

For AC, you have only two choices, either phase control via TRIAC or some sort of DC to AC inverter that can vary the voltage coming out, which is much more complicated.

 

[ericgibbs]

hi,
These opto's for triac control give the required isolation and zero phase point switching in one package.

 

[adaminc]

It seems to me that if I want to be able to on/off and dim any load, resistive or inductive, then I will need to do the AC/DC/AC method, so i figure I will just make 2 devices, an on/off for inductive loads, and an on/off/dim for resistive loads.

To Analog, there is another method called reverse phase control, where you allow the power to flow, then cut it off at the appropriate time, so you cant use TRIAC's but instead use transistors (FETs or IGBTs).

I have been thinking of trying this method out. Supposedly its more efficient, and you can prevent filament noise in lights.

 

[adaminc]

Do any of you have any idea how to do Sinewave Dimming? Supposedly its the newest and greatest thing, it seems all they are doing is changing the amplitude, like you would with a variac, except they use IGBT's and PWM or something.

 

[Analog]

It seems to me that if I want to be able to on/off and dim any load, resistive or inductive, then I will need to do the AC/DC/AC method, so i figure I will just make 2 devices, an on/off for inductive loads, and an on/off/dim for resistive loads.

To Analog, there is another method called reverse phase control, where you allow the power to flow, then cut it off at the appropriate time, so you cant use TRIAC's but instead use transistors (FETs or IGBTs).

I have been thinking of trying this method out. Supposedly its more efficient, and you can prevent filament noise in lights.

How is it more efficient?

 

[Ubergeek63]

For AC, you have only two choices, either phase control via TRIAC or some sort of DC to AC inverter that can vary the voltage coming out, which is much more complicated.

There is another...a FET in a diode bridge can chop the AC.

 

[Leftyretro]

I think it's a pretty daunting task to make an good Variac equivalent out of solid state devices that can deal with reactive loads as well as pure resistive loads. I'm sure it can be done but most likely not on a real general purpose cost effective basis.

First you would have to define an upper current limit that you wished to have, as that would be the biggest factor to determine size, costs and heat management issues. Also Triac type dimmers can be a big source of EMI (noise) generation if not well designed to deal with that.

Lefty

 

[Ubergeek63]

I think it's a pretty daunting task to make an good Variac equivalent out of solid state devices that can deal with reactive loads as well as pure resistive loads. I'm sure it can be done but most likely not on a real general purpose cost effective basis.

As I have said before: Rectify, Buck ratiometric, and Anti-Rectify.

It is still noisy without further attention, but it is not particularly difficult.

That makes it general purpose and not extremely expensive as long as you are not looking at high current loads.

$20-$30 in parts I would guess, with most of the cost being in the inductor and output "bridge".

AC->bridge rectifier->AC referenced buck regulator->power output bridge (to invert every other half cycle)

 

[ericgibbs]

there is another method called reverse phase control, where you allow the power to flow, then cut it off at the appropriate time, so you cant use TRIAC's but instead use transistors (FETs or IGBTs).

Hi,
Dimming tungsten lighting is fairly easy and in common usage.

Using a Triac, the firing angle of the triac is set to trigger at different angles [points] over an half mains cycle.

IC's are available that detect the zero crossover point of the phase and then trigger the triac later in the half cycle,
as set by an external control.

The later its triggered, the dimmer the light.

If you Google for light dimmers you will get lots of hits.

 

[Gayan Soyza]

Not really sure, this is more of a general purpose works in all situations kinda thing. I want to make 1 design, then use it all over the house for automation.

AC Method
In my home light dimmer the circuit I have configured for resistive loads (triac output to AC bulb).The circuit measures the both side AC wave forms
& controlling the TRIAC.But the same circuit I never use for inductive loads it need some additional components (never tried yet for inductive loads).

The technic that I use Eric explained.

DC method
You could also use the PWM.Once I have replaced the TRIAC & placed a MOSFET & connect the bulb via a bridge rectifier.This also worked fine very well.In this case you need to know when to turn "off" the TRIAC instead of "on".

 

[Hero999]

You can use a pulse transformer which can be made using a ferrite bead and some PVC insulated wire.
https://www.electro-tech-online.com...-transformer-using-a-ferrite-bead-core.36019/