I have a bit of a problem with controlling a universal motor with a Triac.

I have built the circuit shown in the attached picture (designed by someone else) and then I've written the software to control the firing of the Triac from an Arduino. It all works very well except that every now and then I get spurious firing of the Triac that causes the motor to spin up for a fraction of a second. This is obviously not helpful, especially as I'm using the motor to drive a coil winding machine.

I don't really have great skills when it comes to Triac circuit design but I do understand that the circuit includes a snubber that helps to stop the Triac firing when it shouldn't but I still get these spurious triggers now and again. I'm certain it's not a software problem so is there anything I should change or add to the circuit to make it more reliable?

Thanks in advance for helping.

 

How are you detecting the Zero mains crossing.?

 

Check out Fairchild AN-3006 AN-3003
Max.

 

Hello,

Here are the mentioned application notes.

Bertus

 

How are you detecting the Zero mains crossing.?

I'm rectifying the AC and taking an opto-isolated output to the Arduino and then timing the trigger from that output.

 

Hello,

Here are the mentioned application notes.

Bertus

Thanks, I'll take a look

 

I'm rectifying the AC and taking an opto-isolated output to the Arduino and then timing the trigger from that output.

Ok post the circuit of that Zero crossing.

 

Ok post the circuit of that Zero crossing.

Here you go....

 

Hello,

For more ideas for zero crossing circuits, have a look at the following page:
https://sound-au.com/appnotes/an005.htm

Bertus

 

Hello,

For more ideas for zero crossing circuits, have a look at the following page:
https://sound-au.com/appnotes/an005.htm

Bertus

Thanks Bertus. Do you think's it a zero cross issue? The circuit works well for most of the time. It's just now and then when it randomly fires. If the ZC was the problem, I would have thought the problem would be evident fairly regularly.

I haven't read them in detail yet but the application notes you posted seem to indicate that an inductive load should probably have a snubber on the opto-triac as well as the main triac. I was wondering if that is the problem but I'm getting a bit out of my depth.

 

I've done loads of phase-fired triac circuits without any problems, but I have never included C3, and always used much lower values for R12+R13 - usually 100 ohm. You DO need a good zero-crossing detector. Don't be tempted to use an AC-input opto-isolator - although it looks like a very neat solution, the two LEDs are different distances from the phototransistor, so the sensitivity is different on each half-cycle which makes your triggering asymmetric.

 

I have a bit of a problem with controlling a universal motor with a Triac.

I have built the circuit shown in the attached picture (designed by someone else) and then I've written the software to control the firing of the Triac from an Arduino. It all works very well except that every now and then I get spurious firing of the Triac that causes the motor to spin up for a fraction of a second. This is obviously not helpful, especially as I'm using the motor to drive a coil winding machine.

I don't really have great skills when it comes to Triac circuit design but I do understand that the circuit includes a snubber that helps to stop the Triac firing when it shouldn't but I still get these spurious triggers now and again. I'm certain it's not a software problem so is there anything I should change or add to the circuit to make it more reliable?

Thanks in advance for helping.

Your Triac looks like it's wired wrong , terminal 1 should be MT2, terminal 2 should be MT1,
I would put a 1K resistor across the G and MT1 terminals to prevent false trigger.

 

Your Triac looks like it's wired wrong , terminal 1 should be MT2, terminal 2 should be MT1,

Yes, well spotted. That one foxed me for a long while when I first built the circuit. It just wouldn't work initially. I eventually found a comment from someone else who suggested the pins 1 and 2 should be reversed and then it sprang to life!

 

When I did my first triac circuit, I assumed that MT1 and MT2 were interchangeable. After all, if they are different, why don't they have different names like Anode and Cathode, or Drain and Source?
I quickly learned the error of my ways!

 

Hello,

Have a look about the fundamentals of thyristors and triacs.

Bertus

 

Your Triac looks like it's wired wrong , terminal 1 should be MT2, terminal 2 should be MT1,
I would put a 1K resistor across the G and MT1 terminals to prevent false trigger.

Thanks for the resistor suggestion. Funny, I found an application note yesterday from ST and their example had a resistor between the gate and MT1 so I added one of about 360ohms. I've only had one or two 'mis-fires' this evening after a lot of testing so I think it has improved substantially. So you think I should probably increase that to 1K?

 

No, your 360 ohm is between MT2 and gate and it supplies the triggering current. I'd reduce it so it switches on more quickly.
By all means, add a 1k between MT1 and gate, but if it is a standard gate current triac (which needs 50mA to trigger) it is probably unnecessary. In most triacs there already seems to be a lower impedance than 1k between MT1 and gate.

 

No leave it at that.

 

I've done loads of phase-fired triac circuits without any problems, but I have never included C3, and always used much lower values for R12+R13 - usually 100 ohm. You DO need a good zero-crossing detector. Don't be tempted to use an AC-input opto-isolator - although it looks like a very neat solution, the two LEDs are different distances from the phototransistor, so the sensitivity is different on each half-cycle which makes your triggering asymmetric.

I may remove C3 then and drop the value of the R12/13.
I'm not sure I understand what you mean about the LEDs in the phototransistor? The output from both halves of the waveform are driving the same single LED aren't they? Please excuse my ignorance if I've got it wrong.

 

I may remove C3 then and drop the value of the R12/13.
I'm not sure I understand what you mean about the LEDs in the phototransistor? The output from both halves of the waveform are driving the same single LED aren't they? Please excuse my ignorance if I've got it wrong.

AC input optoisolator http://isocom.com/wp-content/uploads/2017/08/TIL194-TIL196.pdf consists of two back-to-back LEDs and a phototransistor. The two LEDs are not identical and probably different distances from the phototransistor. Therefore, the sensitivities on opposite sides of the waveform are different.