Maintaining a triac in the close state using a simple AC circuit

[Mr_Smits]

Hello,

This is my first post, I'm not really familiar with how forums work, if I make mistakes explain them to me please  .

I am currently working on a project that includes a triac. In the standard configuration, the triac is controlled by a microcontroller (PIC18).
However, I would like to incorporate a switch to manually trigger the triac, bypassing the card's control.
To achieve this, I have designed the following schematic.

I would appreciate it if you could provide recommendations for improvement or identify any mistakes I may have made.

Triac : BTA16

PS : The load would be fans, which are highly inductive. However, the inductance value in the schematic is not the actual value.

[Andy Chee]

Could you redraw your schematic to include the PIC line? It’s not immediately obvious how you are planning to connect your PIC and bypass switch.

 (just draw a wire and label it “to PIC”, no need for the entire PIC circuitry)

[Mr_Smits]

I added le pic control label ( the pic isn't directely connected to the gate. it is connected to ctrl_triac .

But unfortunately, I think my design wouldn't work.
I tried to build it, but R1 burned out each time (tried 2 times, second time with 2W resistors  :'( ).
I should use a transformer to reduce the AC voltage to an acceptable level,
but I wanted to keep the circuit as simple and robust as possible because the forced triac operation can serve as a safety feature if the microcontroller is blocked or dead

[Dan123456]

Hi and welcome to the forum 

I’m not much use to you as mains stuff still scares me but there are plenty of smart people here who I am sure would be happy to help 

Have fun and be safe!

[Andy Chee]

Connect your bypass switch across collector-emitter of Q501.

That should theoretically do what you want, unless the controller is located some distance away?

[coppercone2]

in that case put a relay controlled by a switch. maybe with a small RC to debounce

[Mr_Smits]

Thanks for your response,

That's what I did in my first version, but in the case where the low-voltage part of the card is down, it wouldn't work.
That's why I tried to do it directly from the main voltage.

[coppercone2]

oh you want a bypass to activate it if the control power goes down?

I mean if its already mains IDK how much you are saving by getting a smaller switch. maybe just bypass the whole semiconductor circuit with a beefy switch?


Otherwise you need to make some kind of low voltage power supply from the mains to have 3v power

[Mr_Smits]

Certainly! Here's a corrected version:

Yes, if there is a problem with the control part of the cart, such as a dead pic or transformer, etc.,
because transformers are regularly dead after a big storm, and I need to still be able to activate the ventilation.

I usually use a power switch, but they are big and expensive.

[MrAl]

I added le pic control label ( the pic isn't directely connected to the gate. it is connected to ctrl_triac .

But unfortunately, I think my design wouldn't work.
I tried to build it, but R1 burned out each time (tried 2 times, second time with 2W resistors  :'( ).
I should use a transformer to reduce the AC voltage to an acceptable level,
but I wanted to keep the circuit as simple and robust as possible because the forced triac operation can serve as a safety feature if the microcontroller is blocked or dead

Hello,

The gate is supposed to be driven with respect to the other main terminal, the one closest to the gate in the drawing of the triac.  Your drawing seems to show it being driven with respect to the wrong main terminal.  That would put a lot of power in R1 or else blow out the triac.  You should test the triac to make sure it still turns on properly, and off properly.

The following circuit may work better as long as you have all your resistor values correctly calculated.
Notice now that resistor R1 now connects between the gate (through R2) and the other main terminal when the switch is on.  The gate is marked with a "G".  To turn on the triac, you energize the gate G with respect to the main terminal on the bottom of that triac shown in the new drawing.

You also have to be careful how you connect the microcontroller to the circuit as shown.  When the switch is turned on there may be a conflict because the uC pin that connects to the switch may get overpowered.  Maybe a diode in series with the uC pin would help.

Also be aware that not all triacs turn on in both directions with just a positive gate signal.  Some require a negative gate signal for the opposite polarity of the two main terminals.  This means you should make sure your triac can turn on in both directions with just a positive gate drive.

You also have to make sure you have enough current flowing into the gate to turn it on.  The gate current requirement could be different for the two different polarities of the traic main terminals also.  It could be higher for one polarity than the other.  It has to be high enough for both polarities.
Also, once it turns on, the voltage across the 100uf cap may decrease which means the gate may not get enough current for the power line cycles that follow.  You'll have to check that too and modify if needed.  A zener may help there to regulate the voltage used to turn the triac on with the switch.  That should keep enough energy in the 100uf cap for all cycles.

A 230vac power line has a peak voltage of around 325 volts, you'll want to take some precautions as that is potentially very lethal.

[Mr_Smits]

In this diagram, you are cutting the neutral, but are there standards indicating that the phase should be cut, if I'm not mistaken?
To ensure there is no voltage at the load between the phase and PE when the triac is open 

[MrAl]

In this diagram, you are cutting the neutral, but are there standards indicating that the phase should be cut, if I'm not mistaken?
To ensure there is no voltage at the load between the phase and PE when the triac is open 

Hi,

Not sure what you are saying here,  What is PE?

The new drawing shows a fairly standard way to use a triac, but if you still have questions try to be a little  more clear.

When the triac opens, the load gets no CURRENT.  That's the main thing.

I assume the neutral is connected to ground.

[Mr_Smits]

PE is for protective earth.
The phase carries live voltage, and interrupting it ensures complete de-energization.
This practice aligns with safety standards for preventing residual voltage between the phase and protective earth.

I may be wrong 

[themadhippy]

If storms are killing the electronics  then surely theres a chance the triac could also get hit,in such cases id adopt the kiss principle and

just bypass the whole semiconductor circuit with a beefy switch?

How much current are we talking?

[Mr_Smits]

the Triac is 16A but i'm limiting the installation to 12A

[themadhippy]

the Triac is 16A but i'm limiting the installation to 12A

suitable switches aint that big nor expensive

The phase carries live voltage, and interrupting it ensures complete de-energization.

Indeed,however ,in uk wiring regs at least,a semiconductor is not classed as suitable for safe isolation as there tends to be a bit of a bleed through so that circuit you think is dead isn't

[Mr_Smits]

You're probably right, I'm going to try a design with a power switch.
It will likely be simpler and more secure.

And I thank you, I didn't know that a semiconductor wasn't considered as a form of protection.
I will go to bed less ignorant 

[coppercone2]

you can protect it with other semi conductors but for a over ride, the switch is king.

also chances are, if you fried low voltage electronics ,there is no reason to assume the triac will always survive. this lets you bypass a open triac, hopefully by replacing a fuse and then turning a switch. you did put a fuse on the input. or cut a shorted triac lead, replace the fuse, and then use the switch until you can repair it.

it just means your surge was not big enough yet.

you could add a big MOV to protect the triac a bit too. the surge should go through the MOV and blow the fuse.

and probobly your 3.3 v supply might break less if you put a MOV on it. there is more stuff but the MOV is simple cheap and easy without much research

I have seen plenty of bad triacs that are blown up....... oddly enough usually the triac breaks but the digital part is OK. I think you got some kind of problem on your digital power supply... (i see broken triac in things for heating systems)



Am I the only one that is suspicious hearing that someone feels a line connected TRIAC is more reliable then a control board?! that is BIZZARE.

[Mr_Smits]

I already have a MOV on the power supply for low voltage but haven't added one for the triac yet. (I will add one, though.)
   
It's my first time designing with a Triac, so I don't really know if it's easy to fry. 
   
I thought triac were very robust components, even if we are talking about currents greater than 10A.

I thought having fewer components between the triac and control would make it more reliable because there are fewer possible points of failure.
hence considering a direct connection to the line.

[Terry Bites]

It may be simpler to use Q501 in a common collector set up. Then you can bypass to 0V.
You have protection against indcutive load and mains transients which may cause false triggering.
Always worth a double check.
There's a handy app note from ST on confguiring the photo triac circuit for indcutive loads.
Triacs are very robust as long as you dont over driver the gate

[MrAl]

PE is for protective earth.
The phase carries live voltage, and interrupting it ensures complete de-energization.
This practice aligns with safety standards for preventing residual voltage between the phase and protective earth.

I may be wrong 

Hi,

Ok no problem.  Just reverse a few things.  See new drawing.

I've added another resistor too R4.  Also a diode D2.

Note I also added some notes...

The first is just to label the triac terminals.  A1 and A2 are used on some data sheets, MT1 and MT2 are used on other data sheets and in the literature.  Note some websites get MT1 and MT2 mixed up so are not correctly shown on those websites.  For a positive gate voltage like 1.5v, the gate is positive with respect to MT1 not MT2.

The second is to show the different quadrants.
Note I also updated the triac part number which if that is correct, then this may not be the right triac for your application.  That's because the data sheet seems to indicated that this particular triac does not work in quadrant IV.  That's the quadrant where the gate is positive and the MT2 terminal is negative.  If that is true, then it would never turn on when the line polarity reverses, meaning the triac would create a half wave rectification for the load, which could be very bad or just not power the load with enough energy.

The data sheet also seems to vary a bit on the required gate current for that triac.  One says 10ma and the other says 50ma.  You'll have to look into that.  That means the gate has to get at least that amount of current in order to turn on for each half cycle.