Use a triac to trigger another triac

[pk178]

Hi everybody !

I don't want to use pulse transformer in trigger triac (or thyristor) circuit, I want to use optopcouple (MOC30xx), but optocouple just permits small current at output, so with a triac need big trigger current (Ig), we can't use optocouple, I have an idea, I'll use MOC to trigger a small power triac (sensitive triac type), then I use this triac to trigger big power triac, this is that circuit :




How do you think about that circuit ? Is it possible ?

 

[goldsmith]

Hey Dude!
Hi
Your theory is ok but your circuit has a problem! your opto coupler can turn on your first triac but when your triac turn on the dangerous voltage will supply your gate ( the gate of second triac and it can destroy your second triac.)
Best wishes
Goldsmith

---------- Post added at 20:07 ---------- Previous post was at 20:06 ----------

you can use opto couplers in parallel with together for example some moc 3021

 

[pk178]

Thanks for reply !

Yes, you're true, above circuit have that problem, I'll solve it by add a resistor between MT2 gates of two triac. This resistor will restrict current to G gate of second triac.

Parallel MOC can be supply bigger current, but with a triac that needs a high current (ex : 500mA) to trigger it, I'll parallel many MOC.

 

[KerimF]

Hi,

I am curious to know which triac needs a triggering current as high as 500mA since I used to work with triacs rated up to 40A only. Their typical gate current is about 50mA (worst case 100mA).

Kerim

 

[goldsmith]

Dear Friends!
High MOC3021 can give you the current about 1 amperes as well as alone. i thing it is twice than the current that you need!
Best lucks!
Goldsmith

---------- Post added at 21:52 ---------- Previous post was at 21:48 ----------

Dear Kerim
Hi
the high current triacs will need more gate driving currents! the gate current can be about 5 amperes!!!!
All the best
Goldsmith

 

[KerimF]

Hi Goldsmith,

So there must be triacs rated for very high currents (more than 1000A)... costly ones I guess

About the MOC302x series you are right, their maximum instantaneous current is about 1A. But these opto series is of non-zero crossing type so the triggering phase could be of any value between 0-180 deg. For this type, the value of the output series resistor is determined, for the worst case, with:

Rs = I_max / V_peak
where V_peak is of the mains AC voltage.

But when the trigger phase (a) is short in order to get the maximum load current, the load voltage v(a) is relatively low proportional to SIN(a). Therefore the output current (to trigger the big triac) is low as well:

I_out = v(a) / Rs

In this case, even the optotriac is on from (a), the big triac will turn on only when the instantaneous v(x) is high enough to satisfy:

v(x) = I_trg * Rs (neglecting the semiconductor drops)
I_trg = 500mA for example.

This limits the range of the big triac control between v(x) and v(180-x).
So at best, the big triac can be turned on from phase x to 180 deg.
And the lowest load current that can be made to flow is between 180-x to 180 in each half cycle.
Triggering after 180-x, the big triac will stay in its off state.

Please let me know if I missed something in this analysis.

Kerim

 

[pk178]

Hi Kerim,

Yes, I thinks this is a disadvantage of MOC.

As your formula, when I use 22Ohm resistor, and I want to get a current 50mA across optocouple, the voltage trigger must be bigger than 22*50 = 1100 mV, with a 220VAC-50Hz source, we can calculate minimum value of x :

The voltage at t(s) will be :

220*sqrt(2)*sin(314*t)

To trigger triac, voltage must be bigger than 1,1V, so we have this equation :

220*sqrt(2)*sin(314*t) = 1,1

=> t = 6,45*10^-4 (s) = 0.645ms.

So minimum value of x will be :

x = 180/10*0.645 = 11.61 degree.

So the open angle range will be x to (180-x) ~ 11.61 to 168.39

But note that when you use another way to trigger triac, you'll never get full load (0-180 degree), because triac (thyristor also) have a parameter named Ihold, current pass triac must be bigger than this value.

By the way, your formula : Rs = I_max / V_peak, it's not correct

 

[KerimF]

Hi PK,

Let us see your MOC example in a real circuit.

Rs=220 Ohm
220*50mA = 11000 mV = 11V
But there are also two voltage drops:
Vg of the big triac which is about 0.8 to 1V
V_opto of the opto triac when on which is about 2V

So for 50mA trigger current, the mains voltage should reach about 11+1+2=14V
14=220*sqrt(2)*sin(a)
a=2.56 deg
The load power in this case will be close to 99.9% (almost 100%).

Now let us see what will happen if we let a=90 deg to get 50% output power.
The output opto voltage = mains V_peak = 220*sqrt(2) = 311V
The output opto current = (311-3) / 220 = 1.4 A
This value exceeds the rated current of the opto triac. Therefore a higher value for Rs should be selected:
Rs = V_peak / 1A
To have a good margin, I usually assume that the mains RMS voltage may become as high as 250V so I choose Rs in this case as:
Rs = 250*1.414 / 1 = 353 ==> 390R

OK?

Kerim

 

[pk178]

Yes, with Rs = 220Ohm, open angle range will be so small. So, if we trigger it by another triac as I mentioned in post #1, I think we can increase open angle range. With sensitive triac, Ig may be smaller.

By the way, could you post your schematic circuit when you trigger triac 40A, is that BTA40/41 triac ?

 

[goldsmith]

Hi my friends!
If you supply the led of moc as well as , its out put triac will be short circuit(approximately) . (from 0 to 360 degree )(approx)

 

[FvM]

As already discussed, the assumption in the initial post, that a MOC30xx can't supply the gate current required by a 25 or 40 A triac isn't correct. It can supply 100 mArms and much higher peak currents. The problem is however about minimum series resistor and acceptable turn-on voltage. As long as no specification is given for the acceptable output voltage loss caused by the turn-on delay, we can't decide if MOC30xx is a suitable triac driver.

Your angle calculations in post #7 are however incorrect, I think. If you assume a turn-on voltage as high as 10 V, you get an respective phase angle of arcsin(10/314) = 1.8 degree, respectively a trigger delay of 1.8/360*20 ms = 0.1 ms, which isn't that much. While the missing part of the full wave is already small in terms of area, it's effectively neglectable in terms of power supplied to a resistive load.

In so far, I wonder if you are trying to solve a nonexisting problem.

Professional triac/thyristor power controllers, e.g. in stage lighting have been using trigger transformers since ever. They can completely avoid the (minor) disadavantages of popular opto triac drivers.

 

[KerimF]

By the way, could you post your schematic circuit when you trigger triac 40A, is that BTA40/41 triac ?

Yes, the available 40A triacs here are BTA40/41.

Since a triac has only one gate (unlike the case when two thyristors anti-parallel are used) and my end controller is usually out of touch, I connect its A1 terminal (equivalent to cathode for thyristor) directly to a hot line (or neutral). Then I let the DC triggering current (continuous or pluses) be negative, that is its direction from A1 to G. The easiest way is to connect A1 to the 5V supply. Then the gate is driven by an npn transistor (emitter to ground) with a limiting resistor between the collector and gate. Also this direct method helps detecting the zero crossings also directly without adding an isolating circuit. But if the controller should receive some external data, I use an opto isolator for the data transfer.

So only when I have to drive 2 thyristors in parallel (equivalent to one triac), I try to use an isolating circuit and using a suitable MOC is one of the possible solutions.

Kerim