Is there a circuit using Triac and capacitor in series to control the speed of ceiling fan? There is a circuit using SSR but I wish to use Triac as SSR are bit costly.


Just for clarity the picture of the speed regulator is attached.

I have tried the circuit below. It works fine with one Triac wired but when both Triacs are wired and triggered one at a time, the Triacs fails randomly. Later it was traced that MT1 terminals of both Triacs are connected to LOAD which is a 60W incandescent bulb. So when a Triac conducts the MT1 and MT2 terminals of the other Triac also gets supply while no gate current is given. I think this cause the Triac to fail. Is this correct?

Is there a way to solve this.

Please help.

enter image description here

  • \$\begingroup\$ At a quick look it appears the trigger optoisolators are between the gate and wrong TRIAC terminal. As shown the TRIACS are on when the optos are not activated. AND havng R2 common to both TRIACS seems potentially to do violence to the other TRIAC. It's about midnight and my brain will not unwind your interesting layout wity certainty.It needs the dawn or redrawing. \$\endgroup\$
    – Russell McMahon
    Commented Aug 12, 2015 at 11:43
  • \$\begingroup\$ It seems that flipping the TRIAC main terminals and separating the gate circuits so each is independant woulld be wise. \$\endgroup\$
    – Russell McMahon
    Commented Aug 12, 2015 at 11:44
  • \$\begingroup\$ @Russell McMahon: When terminals of Triac are swapped, it conducts partly even without gate current. R2 drives Triac through opto any one at a time as triggered by MCU. \$\endgroup\$
    – Jack
    Commented Aug 12, 2015 at 12:17
  • 1
    \$\begingroup\$ Ah - I thought that you had come here to learn and find answers to your problems. But on looking at your other question (which I had not seen when I wasted my time going over your circuit) I see that your intention is to ask a question, get good advice, ignore it, argue for your failing circuit that you do not understand, connect the TRIACs wrongly and then open a new identical question. Quite what you hope to achieve from this is not evident, but it is unlikely to be a working circuit. In the other question you say that the diagram is wrong and that the TRIACS are NOT connected as you .... \$\endgroup\$
    – Russell McMahon
    Commented Aug 12, 2015 at 15:17
  • 1
    \$\begingroup\$ ...show BUT here the diagram still has the wrong connections and you are arguing here that it is OK to connect them "backwards". It's probably not. LOOK at the answers to your other (essentially duplicate) question and comments, and to @nreath's answer and comments here and try some of the things that are suggested. At a minimum UNDERSTAND the suggestions so that if you choose to ignore them you'll know what it is that you are ignoring. | Starting from the suggested configurations and working up is liable to help muchly. | Do note the comment re TRIACs which do not trigger in the 4th quadrant. \$\endgroup\$
    – Russell McMahon
    Commented Aug 12, 2015 at 15:23

1 Answer 1


Spontaneous triac triggering is often caused by a sharp rise in dv/dt or di/dt, both of which are strongly associated with inductive loads (such as the motor which you are attempting to control). The snubber you have with R1 and C2 should help with this but improper selection of values can easily nullify this effect. Snubberless triacs are made which do not operate in the 4th quadrant and have high tolerance for high dv/dt and di/dt characteristics.

Here's a good example

There are also many application notes on the subject as random triggering is a common problem with triacs when attempting to control heavily inductive loads.

This is long AND contains almost all information you could want when it comes to triacs

Additionally I agree with @Russel McMahon in that the resistor R2 should not be shared between the triacs, you should add a second resistor of the same value in parallel which feeds the second triac. Attempt to isolate the triacs as much as possible from each other, with no shared components, only shared power supply and load.

  • 1
    \$\begingroup\$ +1++ I suggest "This is long but ..." -> "This is long AND ..." :-). What a superb application note. \$\endgroup\$
    – Russell McMahon
    Commented Aug 12, 2015 at 15:28
  • \$\begingroup\$ Noted! a simple but significant change :-) \$\endgroup\$
    – nreath
    Commented Aug 12, 2015 at 15:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.