2
\$\begingroup\$

I am designing a snubber to reduce voltage transients for a TRIAC load phase angle control circuit.

The snubber is composed by an RC network in parallel with the load. The load is a 230V 5A electric pump (no more information about it unfortunately) and the TRIAC will be controlled by an MCU which is driving an opto-triac to generate the required turn-on impulse. There is also a MOV in parallel to the snubber/TRIAC to protect from the peak voltage of the transient.

I have looked through different Application Notes regarding RC snubbers in power applications and I tried to follow the calculations used in AN1048/D.

Specifically, I used the TRIAC DESIGN PROCEDURE at page 12 with these data:

\$V_{RMS} = 230\,V\$(mains voltage to the load)

\$I_{RMS} = 5\,A\;\$(and I assumed a pure inductive load)

\$\phi = 50\,Hz\;\$(mains frequency to the load)

\${\frac{dV}{dt}}_{max}=5\cdot10^6\,V/s\;\$(for the chosen TRIAC)

\$\rho=0.6\;\$(Figure 6.18 at page 9, this value seems to me a good compromise between \$V_{pk}\approx 406\,V\$ and the chosen \$\frac{dV}{dt}\$).

The results I have applying the equations in the AN give me a capacitor of \$23\,nF\$ and a resistor of \$3\,k\Omega\$. My problem is that these values seem far from those I normally see in similar circuits or datasheets (eg. \$C=0.1\,\mu F\$ and \$R=33\,\Omega\$) and I believe I am not sizing the components properly.

I have tested this circuit with these values for a couple of minutes while changing the phase angle and it does seem to work, however I did not have the chance to verify the peak voltage of the transients as I miss the proper equipment to do so.

I also believe I can reduce \$R3\$ to a \$1/4\,W\$ power rating but I am not sure about the rating of \$R6\$.

RC Snubber circuit

Any suggestion is really appreciated. Thank you.

\$\endgroup\$
10
  • \$\begingroup\$ I do have an oscilloscope but I do not know how to deal with high transient voltages and which probes I would need for that. \$\endgroup\$ Mar 22, 2021 at 18:34
  • \$\begingroup\$ Oh, that's easy then. Do you have x 100 probes? If you don't find some cheap ones off of Mouser or Digikey. You can get them third-party for like $50-$100 if I recall rather than $400 from first-party. \$\endgroup\$
    – DKNguyen
    Mar 22, 2021 at 18:34
  • \$\begingroup\$ Unfortunately not, I should definitely buy one. \$\endgroup\$ Mar 22, 2021 at 18:35
  • \$\begingroup\$ mouser.ca/ProductDetail/Cal-Test/… But if you are working off mains you need to be mindful of the fact it is not a differential probe so you can't just instrument across any component you want. Differential probes make things a lot easier but now we're talking about $400 for a dirt cheap one. Highly recommended, but understandable if you can't afford it. You should definitely have a x100 probe though, or a couple so you can simulate a differential probe via subtraction when you really need to. \$\endgroup\$
    – DKNguyen
    Mar 22, 2021 at 18:36
  • \$\begingroup\$ Use the app. note from ST, also. 3k ohm makes this snubber useless, IMO. \$\endgroup\$ Mar 22, 2021 at 18:40

1 Answer 1

1
\$\begingroup\$

You really need an oscilloscope for that kind of thing since you're basically working blind without one. Snubber calculations need to take into account the motor inductances in the circuit so without proper instrumentation you can't really do anything. One method involves just do paper napking calculations then running very carefully at low power and observing the ringing frequency and using that to narrow things down. Then you fiddle with values until the problem goes away on the scope.

If you really are working blind, there's no reason to fuss over it so much since you can't verify anything anyways. I'd just pick a 1-5Ohm resistor and a cap as close to 100nF as possible and just live with it.

\$\endgroup\$
4
  • \$\begingroup\$ The thing is that I would like to know why the calculations made in accordance with the AN give values far from the common ones I normally encounter in similar circuits. \$\endgroup\$ Mar 22, 2021 at 20:28
  • \$\begingroup\$ @aviationbus I only skimmed through it but my first guess would be that in figure 6.15, you chose a dV/dt for what the triac can withstand rather than other factors that include the motor inductance. \$\endgroup\$
    – DKNguyen
    Mar 22, 2021 at 20:36
  • \$\begingroup\$ I am confused, my understanding of the AN is that I need to use the maximum rated dV/dt for the Triac and to consider the load as fully inductive if no other details are known about it. And those numbers give me a resistor which is way bigger than expected. Is it due to the damping factor associated with the chosen rho (from Figure 6.18) and the real load model maybe? \$\endgroup\$ Mar 23, 2021 at 13:26
  • \$\begingroup\$ Even the sample calc for that particular procedure on page 12 results in R = 1400Ohms. Seems like its designing for different criteria than most other snubbers. \$\endgroup\$
    – DKNguyen
    Mar 23, 2021 at 14:27

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.