0
\$\begingroup\$

I have been using this capacitive power supply to drive a circuit (an OP amp and relay), but what I can't understand is that when I tried using an SMD resistor (R2) after some testing (turning on and off the power repeatedly) the resistor got sparks and burned eventually. Although R1 didn't burn, when I replaced R2 with a THT (through hole) resistor it didn't burn (rated at 250 mW). So why did that happen?

enter image description here

Note:

  • The circuit consumes at highest rate a power of 100 mW at 27V (D2 is a 27 V Zener Diode) so a current of 4 mA approximately.

  • The voltage across R2 is approximately 3.7 V and a current of 24 mA so the power is only 88 mW.

  • All the SMD resistors are 1206 type (which handles 250 mW of power).

  • I tried to add a inrush current dumper at D1 (pin 3) input however it still got burned.

\$\endgroup\$
13
  • 3
    \$\begingroup\$ What’s the voltage rating of R2? Please calculate the peak power of R2 when C1 is at zero voltage. \$\endgroup\$
    – winny
    Commented Jul 21, 2023 at 16:51
  • 1
    \$\begingroup\$ I would rephase the question, there are plenty of SMD components which are better than their through hole counterparts for power dissipation given the same size (i.e. big pads vs pins, say TO92 vs an SMD alternative, maybe SOT89/223). \$\endgroup\$
    – Wesley Lee
    Commented Jul 21, 2023 at 17:22
  • 2
    \$\begingroup\$ What did your simulations of peak inrush energy and surge immunity show? \$\endgroup\$
    – Tim Williams
    Commented Jul 21, 2023 at 17:57
  • 1
    \$\begingroup\$ You’re not answering the questions raised here. Please do. Also, please simulate your circuit. \$\endgroup\$
    – winny
    Commented Jul 21, 2023 at 18:30
  • 1
    \$\begingroup\$ Note the question you're asking in the text is quite different from the question asked in the title. \$\endgroup\$
    – Scott Seidman
    Commented Jul 21, 2023 at 18:57

1 Answer 1

Reset to default
4
\$\begingroup\$

"I already measured it" - have you measured it in static condition?

Keep in mind that R1/U1 do not limit voltage, only current, and only in static condition when you have continuous power drain by the load. When you power up the circuit C1 can be considered a short to the ground, and U1 can pass entire peak voltage to R2.

As to why THT component did not fail, there could be many reasons. For example it could have a "pulse withstanding" rating, or, more likely, enough thermal capacity to survive the shock.

UPDATE

how to know if the components could handle such thing

The 220V AC has 311V peak rectified voltage. That is first thing to look at. For example RC1206FR-13150RL resistor has 10V maximum working voltage and 25V maximum overload voltage. It is obviously not suitable here. But pulse withstanding ERJ-P08J151V resistor has 500V maximum working and 1000V overload voltage. Much better.

You can use one of many online calculators to see dynamic characteristics of your schematic. In this case, 220uF capacitor charging from 0 to 27V by 311V source via 150Ohm resistor will peak at 2A. Luckily, this will be a very short 3ms pulse.

Also interesting to look at the energy required to charge a capacitor. In your case it is 80.19mJ. Over 3ms time this comes down to 26.73 Watts. Pulse withstanding resistors should be able to survive this. Some of them have pulse power specified in the datasheets.

Finally, the above calculations should be adjusted, since the rectified voltage is not a DC but pulsed. It will take several pulses to charge a capacitor. For example, if time constant is increased to 50ms then the power requirement comes down to quite manageable 1.6W.

Note, that continuous current draw needs attention too. Your measured 24mA (my calculated was closer to 27mA @ 4V) made you think that 250mW rating is OK. I would look for 0.5W and make sure it has good copper pour for pads.

\$\endgroup\$
11
  • \$\begingroup\$ so in general could we consider SMD resistors not good enough for such applications? \$\endgroup\$
    – Hazardous Voltage
    Commented Jul 21, 2023 at 17:16
  • 3
    \$\begingroup\$ That is not what I said at all. My point was that either schematic or, at the minimum, component requirements should be revised to respect physical reality, regardless of the type of components used. \$\endgroup\$
    – Maple
    Commented Jul 21, 2023 at 17:20
  • \$\begingroup\$ I know mate but my question is "how to know if the components could handle such thing" (even from datasheet what parameter should be taken)? \$\endgroup\$
    – Hazardous Voltage
    Commented Jul 21, 2023 at 17:22
  • 1
    \$\begingroup\$ How to know if the component could handle such a thing? What's the voltage rating? R2 is subject to 1.414 * 120 V or 240 V, when charging U1 at the worst possible switch-on moment. Normally we would put a switch-on surge limiting resistor in series with the L input, so it then protects the diode bridge D1 from excess voltage as well. \$\endgroup\$
    – Neil_UK
    Commented Jul 21, 2023 at 18:07
  • 1
    \$\begingroup\$ @HazardousVoltage It may not be rated for it, but it might last a couple of switch flips longer before bursting into flames. Mains connected components need to be officially rated to handle it. A generic resistor may not even rated to handle 200V, but 150V, so you would need more than one resistor in series to handle past 300V mains peaks safely. \$\endgroup\$
    – Justme
    Commented Jul 21, 2023 at 21:25

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