0
\$\begingroup\$

I have a circuit that uses a Schottky diode to provide reverse voltage protection.

However, the diode chosen is on a long lead time and an alternative that is immediately available is required.

I've been looking for an alternative Schottky, but due to a number of factors (e.g. a specific footprint size) I'm struggling to find a suitable device.

I was wondering whether I might be able to use a p-n junction diode, but these will likely add more noise into the system which I want to avoid.

Another idea was to try to use a P-type MOSFET instead. However, I'm not sure if these are noisier than than a Schottky. Does anyone know if there is any work comparing and constrasting the noise between these two devices? Or is there a general rule of thumb that suggests that one is generally considered quieter than the other?

Thanks

\$\endgroup\$
  • \$\begingroup\$ How much stored energy is there? (Kinetic and reactive). FETs give the lowest voltage drop but your busbar must have the lowest ESL,ESR and L/C ratios to meet the ripple specs. This includes Vbat or PSU ESR, WIRE, Caps so that antiresonance gain does not occur. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 4 '18 at 16:12
  • 1
    \$\begingroup\$ Your PS source impedance determines the noise from dynamic loads. E.g. diodes have a commutation voltage noise that can be reduced with RF shunt caps and low Ron FETs have low V drop and give the best results to continue current thru inductance by switching path for BEMF. Current Loop area must be minimized this for LdI/dt=V ripple. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 4 '18 at 17:06
0
\$\begingroup\$

Reverse voltage protection, so then I assume (you show no schematic!) that you want a diode in series with your supply.

What makes you think that noise on the supply caused by the diode is an issue? Only in few scenarios I can imagine that to be a problem and then I would simply use decoupling capacitors after the diode to short any remaining noise to ground.

A circuit with a PMOS can also work but in general you will need to pay attention to the fact that all MOSFETs have a build-in Drain-Source diode and you do not want that diode conducting the inversed supply voltage. Look for examples how to do this.

As a MOSFET can have a very low Rdson when used properly, it will not add much noise but again, noise should not be an issue and if it is, use supply decoupling.

If you do not use supply decoupling then the variations on the supply voltage caused by the circuits operating will very likely cause more noise than the diode!

If you're desperate about your supply (noise) consider this circuit:

schematic

simulate this circuit – Schematic created using CircuitLab

Here if you reverse power it, the fuse will blow.

You can also use a (self recovering) poly fuse.

You can use almost any diode as it just needs to conduct when power is reversed.

The diode is in reverse mode, no current flows in normal operation.

It is still a good idea to add a supply decoupling capacitor.

\$\endgroup\$
  • 2
    \$\begingroup\$ If the OP is THAT concerned about noise, then the noise caused by the resistance of a fuse or PTC device will be a problem, too. \$\endgroup\$ – Dave Tweed Oct 4 '18 at 14:45
0
\$\begingroup\$

If the diode is used for protection against wrong polarity, then I assume it is connected in the power line, not the signal line, so noise wouldn't be any of my concern. Schottky diodes have a lower voltage drop during conduction than Si diodes. Before considering using a Si replacement I advise you to check whether your circuit still works correctly at the slightly (?) lowered supply voltage. Even if the Si diode would be noisier than the Schottky type, I would assume there is a big fat capacitor across the power lines that will absorb most of the noise.

\$\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.