• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Filter Caps With Diode-isolated Power Supplies

Status
Not open for further replies.

JonSea

Well-Known Member
Most Helpful Member
I'm working on a circuit and I'm not sure how to handle one issue. My primary power comes from a switching PoE (power over ethernet) supply. This supply wants several capacitors across its output (12volts).

PoE may not always be available, so I need an input from an external 12v supply. Both the PoE supply and the external 12v supply will be connected to the 12v rail via Schottky diodes. The PoE supply has a diode so the external supply can't put power on its output when it's not powered. The external supply is connected via a diode to keep the PoE supply from supplying power to whatever the external source is.

How is the location of the caps required by the PoE impacted by the diode? If I put them directly at the PoE output, the balance of the circuit does not benefit from those filter caps when an external supply is used. If they are downstream of the diode, will the PoE be happy? Or do I add caps on both sides of the diode? The PoE documentation specifies two 100uF and a 10uF electrolytics.

The 12v rail has a linear regulator to provide 5v - only a few mA needed, and a 3.3 volt switcher, where a out 350mA is required. 12v is required for an electromechanical door lock.

Thanks in advance for any suggestions.
 

dknguyen

Well-Known Member
Most Helpful Member
Do you have the text that describes the capacitors required on the PoE output? I imagine that the main concern would be if the PoE requires those caps for stability. If you have just enough caps directly on the output of the PoE (anode of diode) to achieve that, put everything else on the cathode of the diode.

If they're just required because the PoE can't handle the transient demand on its own, then you dont need any direclty on the output.

You might be able to wire up a PMOS some way to do the same thing as a diode but with the added functionality that when it gets turned on if and only if the PoE is powered, so current can pass current both ways so then you don't have to worry about which side the caps are on.
 
Last edited:

JonSea

Well-Known Member
Most Helpful Member
The PoE is a Silverel AG9800M. Datasheet

The caps are required for stability it sounds like. The AG9800M actually wants 470uF - the AG9900M only needs 2 x 100uF. The pertinent text and diagram are posted below.

I guess this answers my question. Additional caps downstream of the diode may be beneficial?

SmartSelect_20180920-110931_Drive.jpgSmartSelect_20180920-110954_Drive.jpg
 

dknguyen

Well-Known Member
Most Helpful Member
I guess it would be beneficial, but...diminishing returns. I'd see if you can't wire up a PMOS like how I suggested that is driven by the output of the PoE module. Maybe it can be driven somehow by the unstable output of the PoE on startup and then turn on and connect the rest of the capacitance and end up allowing for stability. Could also be more trouble than its worth.
 

JonSea

Well-Known Member
Most Helpful Member
Thanks for pointing out what I should have read in the first place ;)
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top