Delayed turn off triac??

[RyanG]

I'm trying to set something up to turn a triac off after about 250ms. Would something like this work? Don't make fun of me too much if it's a dumb idea.

[Benta]

You can't turn a triac off. Once on, it will stay on until the current through it drops below the "holding current" (check datasheet for value).

[Zero999]

That is true. a TRIAC requires the current to be interrupted for it to turn off. In this case, the current is interrupted every 8¹/₃ms, as it's run off the mains.

I think what the original poster is trying to do is make a solid state timer relay, which turns on for 250ms. It seems a bit overcomplicated. The 555 is run from a transformerless power supply so having an opto-coupler seems pointless, unless it's only to make use of the zero crossing feature of the MOC3041.

[RyanG]

I am trying to do a solid state timer relay and I did want to use the zero crossing. So, would this work for what I have designed it to be?

[Zero999]

Q1, R2 & R3 can be eliminated, if U2's input is connected from U1's pin 3 and +V.

R9 & R8 run quite hot, dissipating over 1W of power each. Consider using a capacitive dropper and reducing the power consumption, by changing to a CMOS 555 and more sensitive opto coupler.

Is it important if it takes a long time to reset when the power is disconnected? C1 and C2 will remain charged for awhile after the power is removed.

[RyanG]

I don't really know why I put that transistor in there. I think that I might have to change C1 and C2. The shortest reset time I need is probably about 1 second. I was planning on 2W resistors for R8 and R9. How do you calculate the dissipation in AC circuits? I'm assuming it's the same but in this case I take the total and divide by 2.

[Zero999]

I don't really know why I put that transistor in there. I think that I might have to change C1 and C2. The shortest reset time I need is probably about 1 second. I was planning on 2W resistors for R8 and R9. How do you calculate the dissipation in AC circuits? I'm assuming it's the same but in this case I take the total and divide by 2.

It's the same P = V²/R.

[RyanG]

I remember why I put the transistor in. It's PNP because I need the triac on until the 555 sends out the high signal, so wouldn't I need that there.

[Zero999]

I remember why I put the transistor in. It's PNP because I need the triac on until the 555 sends out the high signal, so wouldn't I need that there.

It still is unnecessary. If you want the emitter in the opto-coupler to turn on when the 555's output is low, then it can be connected between the 555's output and +V.

[RyanG]

So, I should connect pins 3+4?

[Zero999]

So, I should connect pins 3+4?

You mean like this? If so, yes.

[RyanG]

Yup, thank you for the help.

[RyanG]

After looking at this some more I don't think I really require the zero crossing of the opto. So, if I remove that and drive the triac from the output of the 555 will I have enough current available from the transformerless power supply? The triacs I am using now are 50ma. I don't know the limitations of the power supply here. I only know that it works the way I drew it up. And I would have to connect pin 1 of the triac to the ground of the transformerless power supply, correct?

[ebclr]

"You can't turn a triac off"

This is not true

Turn-off (commutation)

In order to turn the thyristor off, the load current must be reduced below its holding current IH for sufficient time to allow all the mobile charge carriers to vacate the junction. This is achieved by "forced commutation" in DC circuits or at the end of the conducting half cycle  in AC circuits. (Forced commutation is when the load circuit causes the load current to reduce to zero to allow the thyristor to turn off.) At this point, the thyristor will have returned to its fully blocking state.

[BrianHG]

If your turn off accuracy only needs to be around +/-50ms, you don't need the 555, you just need to tune the value of C3 and power the opto directly.  (This is if I understand what you are trying to do...)

If this were part of your NO/NC triac circuit, I would say just add a series diode with parallel resistor and cap to gnd between 2 gates of your 74HC14 and you will get a fairly clean adjustable delay.

Even with this circuit, you can use a 74HC14 with cap and resistor in place of the 555.

[RyanG]

I don't need it to be exact preferably longer than 200ms but shorter than 500ms would be good. It isn't part of my other project but is related. The other one is to turn the power to a motor on and off with a braking circuit. This one is for starting the motors without having the problem of burning our start windings into charcoal. I just want this to automatically disconnect the start winding after a certain amount of time so that it will not burn up if the motor fails to start correctly.

[BrianHG]

Ok, so you want it to turn on during power-up, and only stay on for 250ms, then turn off and stay off until power is cycled once again, correct?

[RyanG]

That is correct. Like I said, the time that it is on can range and longer isn't worse, but the absolute max I would like is 750ms.

[BrianHG]

Give me a day, there's got to be an ultra simple way to do this without any low voltage supply, 555, and optocoupler.

In the mean time, this is not my circuit, I just googled it: (It appears to depend on the current consumption of the motor's main winding, IE, if the motor is turning too slow, the triac gate is turned on and powers the starter coil.  This also means when running, if you physically stop/break the motor speed too slow, the start coil will once again be energized until the motor is spinning fast enough.  Also, I assume the caps in the circuit are AC, not DC as drawn!  I also suspect you would want to add a series resistor, something like 100ohm, on the gate of the triac.)

[RyanG]

The original design of the motors used a centrifugal switch, which has it's own problems. I have replaced most of those with solid state switches, which use current sensing, but those don't work correctly if there is a mechanical problem somewhere. For example, if the bearing inside the cap of the motor slides down on the shaft the rotor sticks to the stator, then the start winding continuously runs. If that happens the thermal breaker should stop the winding from burning but it rarely does. So, I thought of using a time delay instead of current sensing because that would protect the winding no matter what and is easy to reset by resetting the power to the motor.

[BrianHG]

Ok, this is an untested idea which will need comments from other users & refinement of the cap and resistor values I listed.

The bridge rectifier is an off the shelf 4 pin dip, or smd, or 4 diodes similar to 1N4007.

The 1m resistor discharges the 10uf cap to 0v.  At power-up, the 0v cap drives the gate until the first 1k resistor charges it to the supply voltage.  This 1k may work better at 100 ohm, but charging up the 10uf cap faster means a shorter time on the start coil.
For the start coil to get a full jolt, the AC main needs to be off long enough for the 10uf cap to discharge enough once again drive the triac gate.  This means if you cycle the AC off, then back on really fast, the triac wont trigger.  But in this case, I assume the motor wont stop spinning and wont need the starter coil to be fired in the first place.

Third party comments???
You can also make my AC bridge drive a small low power triac feeding the main power one to amplify the signal if needed.

[RyanG]

I am still looking at your circuit there to see how it works. I do have continuous run motors that it would work fine on, but I also have intermittent run motors that I am not sure about because of the reset time. I don't know the exact time between stopping and starting those motors, but for now I would guess around 1 second.