Problem with TRIAC switching

[peter_mcc]

Hi

I've got a problem with a triac motor control circuit in one of our products which used to work but has started having troubles recently. The triac no longer switches on cleanly - it goes on/off/on/off/etc. On some triacs it's only one half of the cycle. Cooling it down makes it worse - warming it up makes the problem go away.

I've tried different triacs from my stock and the same triacs from RS Components - they all have the same problem. I've played around with component values in the snubber as well as the gate drive circuit with no success - I can only make it worse :-)

I'm at the point where I think the best way forward would be for someone to sit down with me and work through possible causes. Are there any experts in Sydney, Australia who can give me a hand? I'm happy to pay for your time/travel.

thanks
Peter

[TerminalJack505]

I'm trying to figure out if this is a triggering problem or a holding problem.  Is the TRIAC not holding once triggered?  If not then the motor's current may not be sufficient to meet the TRIAC's holding current requirements.

Also, are you triggering at zero crossing?  Or are you triggering at arbitrary phases?  Remember, some TRIACs won't trigger in certain quadrants.

If this is a triggering issue then I have to wonder if the gate trigger current is sufficient.

You might look at the datasheet for you TRIAC and inspect all of the temperature-related graphs.  There's likely to be a clue there from the sounds of it.

[TerminalJack505]

I just looked at a couple of different datasheets. 

The following parameters are sensitive to temperature:

• I_gt  The required gate trigger current goes down as the temperature rises.
• V_gt  The required gate trigger voltage goes down as the temperature rises.
• I_H  The required holding current goes down as the temperature rises.

One of these might be your culprit.

Are there any experts in Sydney, Australia who can give me a hand? I'm happy to pay for your time/travel.

Dave's in Sydney.  You should drop him a line.

Edit:  I_L, the minimum latching current is also sensitive to temperature.  Like the other parameters, it goes down as the temperature rises.

[kolbep]

Maybe you should meet up with Dave at his lab, and do an EEVBlog Video as you two identify and sort out the problem.
I found it helps a lot watching how somebody else troubleshoots.
Like when Dave had the Shorted Connector on the Quadcopter. Before I watched that Vid, I never even thought that you could do Resistance measurements along a board to locate the fault. I just used to cut the tracks until the powersupply was happy again (and then have to rejoin them - messy)

[TerminalJack505]

Maybe you should meet up with Dave at his lab, and do an EEVBlog Video as you two identify and sort out the problem.
I found it helps a lot watching how somebody else troubleshoots.
Like when Dave had the Shorted Connector on the Quadcopter. Before I watched that Vid, I never even thought that you could do Resistance measurements along a board to locate the fault. I just used to cut the tracks until the powersupply was happy again (and then have to rejoin them - messy)

Excellent idea!

[pachuma]

It very much depends on the Igt and IH as mentioned by TerminalJack505. There are some nice triacs called, logic gate triac (or something similar) which have a very low Igt. It looks as if you have a border line Igt to drive the triac.

What triac are you using?

How are you driving it? Optocoupler or direct from a uC.

If you post a schematic of how you are driving it, it would help to try and see what's going on.
Do you hold the Igt all the way for the entire half cycle or do you pulse it? If you are pulsing it, perhaps you are not reaching the hold current. However, if you keep the pulse all the way through the half cycle, then everything points in the direction that your Igt is not large enough.

I suppose that you have a smugger network for the inductive load and a filter too, right?

[peter_mcc]

Thanks to everyone who replied - sorry for my lack of response. In hindsight, posting something on Thursday afternoon before going away for the Easter long weekend wasn't a good idea! The place we stayed had no mobile coverage.
I'll work through the replies today - the issue could well be the holding current.
Peter

[peter_mcc]

Hi

After buying & building up some (more) test equipment to measure the current (1R 1% 50W resistor in a box!), I had a chat with someone who worked with the circuit a while ago. He suggested removing one of the filter capacitors. I did and it fixed the problem - no matter how much spray freeze I used the TRIAC switched properly.

I don't have time to investigate exactly what is going on at the moment (about to head away on holidays) but I suspect that the filter cap is interacting with the internal choke and the inductance of the motor somehow. That would make sense because if I tested with a different motor the problem didn't occur - it is at least partly dependent on the load.

Once I get back (and catch up) I'll do some more tests and let people know what I find.

Thanks for the help so far!

Peter