Sensitive gate Triac Driving

Hello,

I am using Bt139 E series sensitive gate triac to directly turn on or off a motor using 89C2051 micro controller (circuit attached. The problem is that when triacs are triggered at medium-phase
angle and drive inductive loads, they generate heavy voltage peaks on the gate, which could damage the micro controller. In my opinion if I put a diode in series with gate resistor will protect the micro controller. Your suggestion is requested to fix this issue.
Note: I can not use optocoupler due to some reason.

Edit: if I put a diode in series with gate resistor, this will protect the micro controller from heavy voltage peaks from the gate.

Hello,

This seems a problem because of turn-off behavior during the zero crossing. It takes time for a triac to return to the off-state during a zerro crossing of the current. You can see this as the triac turns off, just after the zero crossing (so there is some current through the traic allready). This generates a voltage spike across the triac that couples to the gate via internal triac capacitance.

If the spikes occur during the zero crossing of the current, you can be sure that triac turn-off behavior is the problem. You can solve this by using a snubber netwok. Manufacturers have application notes about how to design the snubbers.

Generally spoken sensitive gate triacs are also sensitive to false triggering due to steep voltage transients. I would not recommend a series diode, A parallel transient suppressor is better (in addition to the snubber circuit).

---------- Post added at 20:03 ---------- Previous post was at 19:55 ----------

The voltage spikes may also enter your system via other means (capacitive, inductive, etc), they may even enter the measurement itself so that you may measure things that aren't there.

Thanks for making understand in detail. snubber network is not feasible due to limited space. So if I put diode with gate resistor will block transients thus protect micro controller from being damaged. Is this a right idea?


The sensitive gate Triacs are widely used because they can be directly driven by a micro controller .

xyz9915 said:

Thanks for making understand in detail. snubber network is not feasible due to limited space. So if I put diode with gate resistor will block transients thus protect micro controller from being damaged. Is this a right idea?


The sensitive gate Triacs are widely used because they can be directly driven by a micro controller .

Click to expand...

In my opinion, it is not a good idea to add a diode in series as it may reduce dV/dt rating (false triggering). As mentioned before a parallel clamping solution would be better.

You may protect the controller by adding a small resistor directly to the output of the controller. That resistor is in series with the gate current limiting resistor. At the connection of both resistors, add a schottky diode pair. This clamps the voltage to 0.5+Vsup and -0.5V. The function of the additional small resistor is to force the spike through the shottky clamp and not PN junction in the controller.

Regarding sensitive gate triacs, you are right they are widely used, however advantage may come with disadvantage. It is good to consult some application notes NXP, ST, etc, regarding use of traics. From experience, I met sentive gate triacs that triggered on mains transients (resulting in faillures). I am sure it were mains transients as we could reproduce the effect with transient injection.

---------- Post added at 15:14 ---------- Previous post was at 15:13 ----------

If you have false triggering, you may consider snubberless triacs (NXP: Hi-com).

Thanks again WimRFP for your support