0
\$\begingroup\$

I want an alternative low component, transformerless power supply for low power DC projects, I have used Shunt Regulators in the past but never with AC. Below I have proposed a AC to DC shunt type regulator using a Constant current diode (AL5890) to limit the current to 40ma and a 5v6 Zener to regulate the voltage to 5v. the input voltage is 230VAC,

Would it be a good idea to still use a current limiting resistor in addition to the Constant current diode? How reliable would this setup be ?

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ "mah" is not a unit of current. The usual unit of current is amperes, abbreviated as "A". A milliamp (mA) is 0.001 A. \$\endgroup\$ – The Photon Oct 30 '18 at 17:10
2
\$\begingroup\$

Do not do this. It's dangerous.

What you're doing is dangerous. Always, always, always have galvanic separation between your high-voltage source and your low-voltage output.

Your bridge rectifier, if it fails, with a high probability fails in a "closed" state, so that you end up connecting your output to grid voltage.

Whoever recommended this as general solution to the problem of generating 5 VDC out of 230 VAC is a dangerous person.

Do not do this. It doesn't regulate well at all, and wastes power.

There's literally hundreds of posts on here that explain why regulating voltage by parallel Zener diodes is a bad idea.

It's not stable. It wastes most of energy. It's, due to the required size of diode and cooling, more expensive than a linear regulator IC.

Whoever recommended the right half of the circuit as 5V regulator for the DC coming out of the bridge rectifier lacks the skill to use modern (as in: since 1977) components.

Do not do this. It's overly complicated and expensive.

Any USB charger is a 5V regulator. Buy one. They are isolated, properly regulated, properly enclosed, come with an actual plug....

You can also buy a proper power supply module. These are typically far more expensive, but can more easily be integrated in devices. Then again, I've actually seen devices where when you open the box, you'll find a plugged power supply inside: Buying a certified, cheap, off-the-shelf power supply is so much cleverer than designing your own circuit.

\$\endgroup\$
  • \$\begingroup\$ Don't! remove. That's not good. The others can learn from this :) \$\endgroup\$ – Marcus Müller Oct 30 '18 at 17:27
2
\$\begingroup\$

I'll answer the one question that @MarcusMuller did not, but I agree with his statements and conclusions.

This will not be reliable. The average dissipation of the constant current device could be as much as 9 or 10W. That is a very large amount of heat to get rid of even for TO-252. This part is designed for most of the voltage to be dropped across an LED chain.

The absolute maximum voltage the LED driver device can withstand is 425V. One would normally like to see at least a 600V device used on a 230/240V circuit to account for transients. There is not enough margin for the MOV to protect the device.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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