1
\$\begingroup\$

I am working on a circuit that is powered by a regulated 5V external power supply.

However, since I will be directly powering a Micro-Controller from this 5v supply, I was wondering how I should 'condition' the voltage from the power supply.

In other words, what circuitry do I need to implement between the 5V output of my power supply and the 5V VCC line of my Micro-Controller?

Obviously I can't pass the 5v from the power supply through a 5v regulator, so I was wondering if there is any other simple alternative to ensure the voltage from the 5v PSU remains at a constant 5V.

\$\endgroup\$
  • 2
    \$\begingroup\$ A good five-volt PSU contains all the regulators you need in it, unless you're doing something very odd. \$\endgroup\$ – Hearth Jan 27 '19 at 22:03
  • \$\begingroup\$ There are numerous ways to shut the power off when a certain voltage is exceeded. We just discussed here but also look in related topics and search the intenet: electronics.stackexchange.com/questions/419082/… \$\endgroup\$ – Fredled Jan 27 '19 at 23:25
  • \$\begingroup\$ That being said, I agree with Hearth that almost every PSU produced today, are SMPS units which modulate voltage very precisely. They may not protect against surges, however. For short durations, a few milliseconds, there could be overvoltage if a power surge appears at the main input of the PSU. Or if there is a hard switch between the PSU and the ic causing a bounce. Then a varistor may help, But a varistor will onply protect for large voltage difference. \$\endgroup\$ – Fredled Jan 27 '19 at 23:33
  • \$\begingroup\$ @Fredled Understood. So realistically the PSU will handle regulation and I only need to add some decoupling caps (I would add these for the Micro-Controller anyway). In regards to a voltage spike, I doubt there would be a voltage change big enough for a varistor to stop in my application but if that did happen I doubt it would just be my project affected. Thank you! \$\endgroup\$ – SimonPCB Jan 28 '19 at 8:07
3
\$\begingroup\$

As a @Hearth says, your PSU should do most of the work. You should add some decoupling capacitors around your µController, and you can also (although it's exceedingly unlikely that you'll need it) add a filter (ferrite beads between the psu and the decoupling, or a ferrite with a couple of turns on the +ve and -ve leads bringing in the power).

If you're worried that your PSU might give you too much voltage and damage your circuit, you can add a crowbar circuit. Like the following:

schematic

simulate this circuit – Schematic created using CircuitLab

This will short out the output of your PSU if the voltage gets too high (people often add a fuse in the input so that the crowbar blows the fuse when it triggers).

For a simple microcontroller, a Schottky diode on the input, (or a reverse biased diode between +ve and -ve), is usually all you wnat. (to protect you from someone who gets the power the wrong way round)

\$\endgroup\$
  • \$\begingroup\$ Could you explain what SCR1 is? \$\endgroup\$ – Fredled Jan 27 '19 at 23:37
  • 1
    \$\begingroup\$ @Fredled It is a (s)ilicon-(c)ontrolled (r)ectifier, hence "SCR." It's a four layer, three terminal thyristor (a 'diac' would be a four-layer, two terminal thyristor example.) The lead you see going from the SCR to the resistor is called a "gate." As long as the resistor doesn't drop much voltage, the gate is held close to ground and the SCR is "OFF" and doesn't conduct from the top side (anode end) as shown to ground (cathode end.) If the resistor is made to drop enough voltage, then the SCR turns "ON" and attempts to "short" the anode to the cathode by letting current flood through it. \$\endgroup\$ – jonk Jan 28 '19 at 0:13
  • \$\begingroup\$ This is great. I will have a look at implementing a crowbar circuit to mitigate the risk of over voltage. Thank you! \$\endgroup\$ – SimonPCB Jan 28 '19 at 8:09
  • 1
    \$\begingroup\$ Unlike a mosfet it latches on. Once it's been triggered the SCR continues to conduct (even if the gate returns to 0v) until the current through it goes to zero. \$\endgroup\$ – james Jan 28 '19 at 12:44
  • 1
    \$\begingroup\$ @Fredled james already provided a response but didn't link your name so you may not have gotten the message. But in general, the SCR when triggered stays triggered so long as there is sufficient anode to cathode current. If the power supply rail continues to flood the SCR with current, then the SCR will remain ON even when the triggering pulse is no longer present and even when the power supply rail is lowered below the zener's threshold. A mosfet won't act like that. \$\endgroup\$ – jonk Jan 28 '19 at 18:14

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.