Dear readers I'm designing a small 5V SMPS which I want to use in a 5V parallel 1:1 protction configuration meaning one output is off when it is slightly lower than the output of the other. Ones one power supply is removed, the one that stays in place takes it over. I want to use ORing Diodes but these give me a drop of 0,2..0.4 Volt. So I came up with and better solution, but I don't have experience with this one.
Below you see two buck converters. In SMPS[A] I have put the ORing diode in series with the output and in SMPS[B] I have put the ORing diode before the FB resistor network. SMPS[B] does not suffer from the 0.2..0.4 voltage drop. However in case the input voltage of B is zero the FB pin will have a voltage of around the FB voltage of 0.8V.
Assuming that both SMPS[A] and [B] have input power the the combinations A-A A-B and B-B should work without a problem.
Assuming that SMPS[A] has input power SMPS[B] has no power and [B] is switched on after some time. The combinations A-A B-B will work but combination A-B I have my doubts about.
If B is powerless and has 0.8V at it's feedback pin and ten starts up will that not generate trouble like latch up or some other issues I'm not aware of.
Now my question: does anyone have experience with this and can tell me if there are things I needs to consider.
Hi LarryD Are you sure?
The PMOS for both power supplies will conduct as long the source is higher than the gate voltage. This means when one of my power supplies are a little different, say 100mV, they shortcut. I want them to act like a blocking diode if one SMPS is a little higher than the other.
I might not be have been clear about this, but I want one power supply active if the other is light different from the other. If one is not available, then the other take it over.
See what happens if you connect both outputs together (I expect nothing will happen).
Then calculate Rg so the supply outputs say 100mV more.
Connect a resistor across Rg to make that value.
I expect that one supply is now inactive, until you remove the one with the highest output voltage.
Leo..
For the ones who are interested
I did some experiments with a circuit that I found on the Internet and it works quite well.
The voltage window in which this switch is working is 150mV difference in output voltage of the two power supplies. Within this window, both power supplies supply a portion of the power. Outside this window only one of the two does, the one with the highest voltage. The picture show the edge of the window in which around 12mA is supplied by the left power supply. With the right voltage at 5.16V the left current is practically zero.
