Anyone had experience of driving a T2035H-6T triac using a MOC3043 opto-isolated trigger? I'm experiencing a situation where, at high temperatures, the triac will not switch off when the opto trigger is removed. This is self sustaining as the "on" triac maintains the level of heat. Spraying the "off" opto (MOC3043) with freezer causes the triac to switch off.

 

Most likely the triac is annoyed by the heat. We can do almost nothing without a schematic.

 

Spraying the "off" opto (MOC3043) with freezer causes the triac to switch off.

The LED junction of the opto is somewhat temperature sensitive, certain series more so than others. I'm not familiar with MOC3043 in particular, but the LED voltage on level will decrease with rising temp. Is the opto completely powered off, or is there residual current below the published trigger point? As #12 noted, a schematic would make things easier to sort out.

 

Schematic attached. There are 2 drivers in this application and the one with the triac closest to its opto (on the PCB) is exhibiting the problem. We think it's thermal conduction through the tracks. We have tried a bigger heat sink on the triac which appears to solve the problem BUT is a botch (on the PCB and enclosure). I would like to know the actual failure mechanism to see if there is a more correct or elegant solution. We could take the triac off the PCB and dangle it in the enclosure but not a nice solution. Any thoughts gratefully appreciated.

 

How hot is hot? Any idea? And the load current and voltage?

 

What happens if you remove or increase the resistors R12,R14 to 1K when they latch,
also how are you driving the opto leds with a transistor or chip pin?

 

The LED junction of the opto is somewhat temperature sensitive, certain series more so than others. I'm not familiar with MOC3043 in particular, but the LED voltage on level will decrease with rising temp. Is the opto completely powered off, or is there residual current below the published trigger point? As #12 noted, a schematic would make things easier to sort out.

The opto is driven from a PIC port output and measures 4.5V when "off" with the Enable signal at 5V. I'm wondering now if that port o/p needs a pull up to 5V.

 

How hot is hot? Any idea? And the load current and voltage?

With a 2kW (240V, 10A) load we are measuring about 110C on the triac heat sink and about 40C on the opto case.

 

What happens if you remove or increase the resistors R12,R14 to 1K when they latch,
also how are you driving the opto leds with a transistor or chip pin?

I assume the circuit has been designed (not by me) from the data sheets so assume all triac resistors are as they should be. Driven from PIC port o/p.

 

Electronic Components maybe OK, what about the thermal rating of the H.S.?
Max.

 

Electronic Components maybe OK, what about the thermal rating of the H.S.?
Max.

I'll check

 

Are you pulling the opto to ground to switch it on, or to V+, i would take it to V+.

 

The opto is driven from a PIC port output and measures 4.5V when "off" with the Enable signal at 5V. I'm wondering now if that port o/p needs a pull up to 5V.

This puzzles me. Are you saying the voltage delivered to the LED in the opto-coupler is 4.5V when off and 5V when on?

Your partial schematic does not show how the current flows in that circuit.

ps, 40 C is barely above the temperature of a person. This is not a problem for silicon, but heat sneaking in through the circuit board connectors seems to have been demonstrated in post #4.

 

You are kind of pushing the triac.
The leakage of the opto goes up with temperature as well.
Might be worth a try to make R14 two or 3 times larger.

 

May I remind you that R12 and R14 do not affect current balance between the triacs very much. They are basically timing resistors. The current required to trigger D7 and D8 (the large triacs) is mostly a fixed value. R12 and R14 delay the start of the big triacs until the voltage of the power line is high enough to drive that much current through R12 and R14. Delaying the onset of the big triacs when the power voltage is near its zero crossing point has less effect than one might intuit.

The small triac is relieved of duty at the moment the large triac fires because there is only the voltage across the big triac being applied to R12 (R14), the small triac, and R13 (R15). Thus, the small triac has a small duty cycle.

These ideas were gathered with my oscilloscope at an earlier date.

 

This puzzles me. Are you saying the voltage delivered to the LED in the opto-coupler is 4.5V when off and 5V when on?

Your partial schematic does not show how the current flows in that circuit.

ps, 40 C is barely above the temperature of a person. This is not a problem for silicon, but heat sneaking in through the circuit board connectors seems to have been demonstrated in post #4.

The Humidifier/Dehumidifier Enable levels are +5V. The opto is switched "On" by a port o/p from a PIC going "Low" (Vce sat). Must admit I haven't checked the port configuration - trying to get my head around one thing at a time but appearing unprofessional.

 

May I remind you that R12 and R14 do not affect current balance between the triacs very much. They are basically timing resistors. The current required to trigger D7 and D8 (the large triacs) is mostly a fixed value. R12 and R14 delay the start of the big triacs until the voltage of the power line is high enough to drive that much current through R12 and R14. Delaying the onset of the big triacs when the power voltage is near its zero crossing point has less effect than one might intuit.

The small triac is relieved of duty at the moment the large triac fires because there is only the voltage across the big triac being applied to R12 (R14), the small triac, and R13 (R15). Thus, the small triac has a small duty cycle.

These ideas were gather with my oscilloscope at an earlier date.

Thanks, beginning to get the picture, now.

 

The opto is switched "On" by a port o/p from a PIC going "Low" (Vce sat).

The idea that a 10k resistor can pull a pin up by 1/2 of a volt demonstrates that the difference in current is less than 50 ua. The LED in the opto is not going to respond to 1/2 of a volt or 50 ua. I believe this is not your problem area.

 

Now that we have pondered the triacs, we can see that, "off" means the small triac has to endure the entire peak voltage of the power line. This might be your failure mode when hot. You have a 400 volt device holding back 340 volts at the peaks. Maybe it has failed. Try replacing it.

 

Now that we have pondered the triacs, we can see that, "off" means the small triac has to endure the entire peak voltage of the power line. This might be your failure mode when hot. You have a 400 volt device holding back 340 volts at the peaks. Maybe it has failed. Try replacing it.

Thanks