TRIAC remains ON with inductive load

Good day to all.

I need to control the power to a 12V 1000W heating resistor using phase control (firing angle). Also, the power must be ON for 200ms and rest (OFF) for another 1000ms or so. This timing is controlled by PLC and the angle by a manual potentiometer. The 12V are obtained from a big 220VAC/12VAC transformer.
So I need to combine phase control with an "enable" signal from PLC. I'm using the attached circuit.
All goes well when the firing angle is large (Low power). But after a certain point when I ask for more power, Triac remains ON even if the 24V signal from PLC is OFF. If the requested power was not too much, you can turn the POT back and regain control. If it was too much, the Triac may get damaged, MT1 and MT2 get shorted, and the heating resistor catches fires and burns.

I think that the inductive nature of the Triac's load is giving me problems.

Any ideas? Sadly, I don't have access to an oscilloscope.
Maybe another circuit arrangement.

Thank you so much for any help/suggestions. This has been driving me crazy for weeks now.

Regards,
Bruno.

Is that circuit really correct, or I'm missing something ?
I think that U2 should be in series with TR1 primary.

[EDIT]

I liked the way you specified TR1:
Several KGs transformer.
;-)

andre_teprom said:

Is that circuit really correct, or I'm missing something ?
I think that U2 should be in series with TR1 primary.

Click to expand...

At this point my brain is saturated, but I think Triac U2 is in series with TR1. I have put the ground below the MAINS. Now I see that I chose a wrong way to graph it.

I liked the way you specified TR1:
Several KGs transformer.
;-)

Click to expand...

Lol !!
:wink:

This is basically a lamp dimmer circuit with the opto-triac working like a series switch in the gate current. It's an interesting idea but I'm not sure it will work because the zero-crossing circuit will see a skewed version of the mains phase. It would make more sense to move all the control to the LED side of the opto and use the triac only as a switch.

Brian.

What is the kind of circuit that drives the opto, open-collector, push-pull, etc ...? By the way, you are right regarding to the path of the power, for some obscure reason I was mistaken in the analysis done earlier.

Just to check: Could you confirm if the Varistor is not getting hot ?

Easiest solution is just to buy a commercial solid state relay.
But it will need to have two specific features.
Be the instantaneous turn on type (not just zero crossing) and have inverse SCRs (not triac control).

These work very well phase controlling partly inductive loads, and it will also provide the optical isolation.

Something like a Crydom D2410-10 perhaps ?
**broken link removed**

Triacs ar at risk of self-triggering with insufficiently snubbered inductive load. I agree that SCR based solid state relays are a good idea, but regular triacs can work with a stronger snubber as well.

Triacs can probably be made to work, but its not always an easy problem to solve, even with an oscilloscope !

Back to back SCRs will positively commutate, but generating suitable isolated gate drive for them is not always so simple, especially with an inductive load.

I doubt that the moc3041 work fine in such conditions. Replace by a microrelay.

It could be that C3 is resonating with the primary inductance of the transformer. Try a value of .001. I have had this using a triac to switch a relay.
Frank

I doubt that the moc3041 work fine in such conditions. Replace by a microrelay.

Click to expand...

Yes it's a dubious design, but the ZCD version is less likely to cause self triggering than the triac itself. So it's probably not the problem. But it's good practice to place the MOC behind a snubber, as in the manufacturer's circuit examples.

It could be that C3 is resonating with the primary inductance of the transformer. Try a value of .001. I have had this using a triac to switch a relay.

Click to expand...

Unlikely for a "several kg transformer". More likely the capacitor is too small. I would go for a strong snubber with 0.1 to 0.22 µF X2 capacitor and a respectively lower series resistor.

I’ve used this circuit to control an AC motor, which is surely a high inductive load, and worked fine with an open-collector output drive.

broken link removed

Wow, so many responses! Thank you all!:bsdetector:

I will try to cover all the things discussed.

andre_teprom said:

What is the kind of circuit that drives the opto, open-collector, push-pull, etc ...? By the way, you are right regarding to the path of the power, for some obscure reason I was mistaken in the analysis done earlier.

Just to check: Could you confirm if the Varistor is not getting hot ?

Click to expand...

The Opto MOC3041 is driven by a Schneider PLC output. Datasheet only say "Positive Logic Discrete Transistor Output". Don't know the actual circuitry. Anyways, I can see the input led switching normally.

I will try to check if the Varistor gets hot.

Warpspeed said:

Easiest solution is just to buy a commercial solid state relay.
But it will need to have two specific features.
Be the instantaneous turn on type (not just zero crossing) and have inverse SCRs (not triac control).

These work very well phase controlling partly inductive loads, and it will also provide the optical isolation.

Something like a Crydom D2410-10 perhaps ?
**broken link removed**

Click to expand...

Going for certain particular industrial Solid State Relay is difficult due to import barriers in my country that make almost impossible for the resellers to get them. Will look for something like that anyway. How do you control phase angle firing in this case?

_Eduardo_ said:

I doubt that the moc3041 work fine in such conditions. Replace by a microrelay.

Click to expand...

I don't think that a micro relay would work in the long run due to the frequency of the commutation: 1000ms OFF ~ 200ms ON.

FvM said:

Yes it's a dubious design, but the ZCD version is less likely to cause self triggering than the triac itself. So it's probably not the problem. But it's good practice to place the MOC behind a snubber, as in the manufacturer's circuit examples.

Unlikely for a "several kg transformer". More likely the capacitor is too small. I would go for a strong snubber with 0.1 to 0.22 µF X2 capacitor and a respectively lower series resistor.

Click to expand...

I will install a stronger Snubber filter as suggested and let you all know.
I will also try installing a MOC3021 (NZC) just to see what happens.

andre_teprom said:

I’ve used this circuit to control an AC motor, which is surely a high inductive load, and worked fine with an open-collector output drive.

broken link removed

Click to expand...

Regarding the circuit you posted, how do you control de firing angle there? R11?


How about Triac's temperature? Could it be a factor in the self firing?

Best regards to all.
Bruno

Br1 said:

Going for certain particular industrial Solid State Relay is difficult due to import barriers in my country that make almost impossible for the resellers to get them. Will look for something like that anyway. How do you control phase angle firing in this case?

Click to expand...

Something suitable is often available from e-bay.

The input is typically quoted as being a dc voltage, usually 4v to about 30v.
A 5v logic signal would trigger the device "on" when the input goes to +5v.
Or a 24v logic signal would turn it on when +24v is applied, or anything inbetween.
So basically you just hook it straight up to your PLC.

There are two different types available.
The most common switches both on and off only at the mains zero crossings which you do not want.
For phase control you need the instantaneous "on" type, sometimes called "random fire". This will turn on at any part of the mains cycle as soon as the PLC tells it to turn on. Turn off will always be at the next zero crossing.

Make absolutely sure the solid state relay you order has back to back SCRs, and not a triac. If it does not definitely say SCRs, its probably a triac.

Its probably not the cheapest solution, but its guaranteed to save you a lot of grief and frustration in the long run.

All the above posts from many people are a list of typical triac problems which many of us have battled with varying amounts of success. Its not an easy thing to find a solution for, particularly if you do not have access to an oscilloscope and are not otherwise pretty well set up.

Warpspeed said:

Something suitable is often available from e-bay.

For phase control you need the instantaneous "on" type, sometimes called "random fire". This will turn on at any part of the mains cycle as soon as the PLC tells it to turn on. Turn off will always be at the next zero crossing.

Click to expand...

Thank you Tony.

My question is: How do you "sync" the PLC's output with Mains. In other words, If I want to fire at 90 degrees for example, how do I know when to start a timer for the required 5ms? (Mains at 50Hz)

Cheers
Bruno.

From your first post I assumed all of that part of it is already working, and the only problem was that the triac was not always reliably turning off due to the inductive load.