TRIAC Flickering problem

aaronik19 · Joined: 25 Apr 2011 Posts: 297

Dear All,

I am trying to make a new triac switch using an optocoupler and TRIAC. Please refer to the attached picture.

http://postimg.org/image/x3ok52vkd/

I am trying to interface the PIC micro to control the duration of the on and off. But first I am trying the power hardware before I move to the software and when I apply 5V to the optocoupler the lamp switch on 100% ... so far so good!

But when I remove the 5V, the lamp starts to flicker fast and when I measure the voltage from teh output of the TRIAC to neutral I found 163VAC. The lamp power is 50W and the TRIAC is 1A.

Can someone tell me what coulc be the problem. From my knowledge when you switch of the input of the optocoupler the lamp must switch off. right? Thanks for your help and support as always.

Ttelmah · Joined: 11 Mar 2010 Posts: 19381

Nothing to hold the TRIAC off. You need a resistor from the gate to the bottom, to prevent 'self commutating' (which is what you are seeing). It also depends on the power factor of your load. Though a simple resistive lamp ought to be pretty neutral, it is surprising how often either capacitance or inductance is present in the circuit (wound filament lamps often have quite noticeable inductance). Honestly I really would suggest doing the design another way. TRIACs are a really bad way of doing almost anything!. Individual SCR's are always better, and even better GTO SCR's or MOSFET's. 99.9% of the interference problems in domestic electronics are caused by TRIAC based circuits.....

aaronik19 · Joined: 25 Apr 2011 Posts: 297

Thanks Ttelmah for your feedback. So for mosfet I found the following. I am posting the link for others as well to help them:

http://forum.allaboutcircuits.com/blog/controlling-an-ac-load-with-a-mosfet.518/

What do you think?

Ttelmah · Joined: 11 Mar 2010 Posts: 19381

As a further comment to this, the two MOSFET's in series approach given slightly further down the thread, is the basis of many of the AC/DC SSR units being sold. Smile

Going even more ingenious, you can use zero crossing on this, and can do the opposite of a TRIAC, turning 'on' at the zero point, then 'off' mid cycle. This is what is needed (or a GTO thyristor that can also be turned off like this), to control capacitive loads.

asmboy · Joined: 20 Nov 2007 Posts: 2128 Location: albany ny

Ttelmah · Joined: 11 Mar 2010 Posts: 19381

And it is amazing what a good transmission aerial your house wiring makes!.... Evil or Very Mad

aaronik19 · Joined: 25 Apr 2011 Posts: 297

To continue discuss this topic..I want to ask you, is it possible to drive ac load with two mosfets in series driven by optocoupler? I searched on google and saw many forums which says that mosfets are for dc not ac.

Ttelmah · Joined: 11 Mar 2010 Posts: 19381

That's the point about the two MOSFET's in series.

Each MOSFET can only control current in one direction. However use two in series and you can switch current flowing each way.

As I said, this is the basis of some of the AC/DC SSR's.

If you look for a 'random turn on SSR', this has all the high voltage parts, and the isolation already built for you, and is a far safer way to go, than anything involving actual mains circuitry....

Have a look at this:

<http://www.crydom.com/en/tech/newsletters/solid%20statements%20-%20ssrs%20switching%20types.pdf>

and consider using a suitable SSR.

aaronik19 · Joined: 25 Apr 2011 Posts: 297

Thanks for your help, I came across this as well. I am posting this for more reference. It is for AC Signal only since it can handles only 0.12A at 380VAC

http://uk.rs-online.com/web/p/solid-state-relays/7085459/

temtronic · Posted: Wed Apr 06, 2016 12:57 pm

hmm they bring the center tap out labelled 'DC' so you could use it on DC one would think....

There's bound to be a few apnotes about them. Nice to interface to PIC as it's optocoupled and PIC can direct drive it.

Jay

Mike Walne · Joined: 19 Feb 2004 Posts: 1785 Location: Boston Spa UK

What I've gleaned from the data sheet:-

The switching times are in the ms region.
So you can only think in terms of slow speed control.
PWM type operation (dimming) is probably out of the question.

The DC current rating is twice the AC one.
Presumably you take the DC out via the centre tap.
That way your only passing the current through one MOSFET.

Mike

They're not showing much detail on how the MOSFET is driven.
I'd guess it's a PV diode (or few), hence the speed, or lack of it.

asmboy · Joined: 20 Nov 2007 Posts: 2128 Location: albany ny

are your lamps filament or led ?
----------------------------------------------
what POWER level :: Volts and AMPs please- when fully ON.
-----------------------------------------
-do you have PRIOR experience with designing AC dimming circuits?
----------------------------------------
because the circuit discussion so far makes me nervous that answer number
3 is no ...... Sad

temtronic · Posted: Wed Apr 06, 2016 4:12 pm

Then there's the KISS way of PWM 'light'. Use a PIC, mosfet and '12 volt' LEDS. They are EASY to control, cheap and SAFE !!! Anything with 'mains' power HAS to be treated with the utmost of respect. If getting zapped doesn't kill you, whacking your head on the table across the room sure hurts. BTDT.

Jay

gpsmikey · Joined: 16 Nov 2010 Posts: 588 Location: Kirkland, WA

aaronik19 · Joined: 25 Apr 2011 Posts: 297

Hi all, continued to make some more experiments and i built the attached circuit using IRF740 Mosfets.

http://postimg.org/image/wwa7bebl3/

The circuit worked perfectly and when i replaced 1k resistor and used a pot to dim, the filament dims accordingly and perfectly. The only issue is that the mosfets become very warm.

I would like to test this circuit using dimmable led to see the result.