Does my OR-ing or Ideal diode Work or not?

[Piet de Pad]

Hi,
Does anyone have experience with parallel switching of power supplies like I do depicted in the schematics below. I have designed and simulated an ORing or ideal diode based on simple components. It's based on a PMOSFET and a LM393 or LM339 comparator. It works amazingly well as far as my simulations are concerned.

The switching window is 10mV, which is the difference between the two supply voltages to select one of the two power supplies. Within this window both, power supplies work in parallel. Outside the window, only one power supply works. When you remove one of the supplies, the other one takes over.The three figures show the various states the ORing can be in.

Fig 1 shows the ORing when both power supplies have equal voltages. Both power supplies current. In fig 2,3 you can see that with a difference of 10 mV between the two power supplies, one or the other is selected.

Before I build it, I would like to know if there are others who have done this before. let me know in de commends if you have or have suggestion for things I have overlooked.


Thanks

 

[crutschow]

Fig 1 shows the ORing when both power supplies have equal voltages. Both power supplies current.

That may be problematic in the real circuit.
The sim uses perfect components, but due to actual component and comparator offset voltage tolerances, the window will likely be larger than 10mV or there may be very little window.

 

[Piet de Pad]

Hi Crutshow,
I conducted further research on the effects of input offset on this circuit, and I have come to the following conclusions.

The circuit is not suitable for:

1. Voltage sources that experience a voltage drop when loaded with a certain load.
2. Voltage sources such as batteries and small solar panels that have a variable voltage output over time or due to the load or depending on sunlight, as these make the circuit unsuitable or instable.

The circuit is suitable for voltage sources:

1. Whose unloaded voltages are sufficiently different and will never be different smaller than in the range of "the comparator offset value + the largest voltage drop due to the load of one of the source". In many cases, this will be in the order of 0.5 volts or less.

In all cases where the voltage difference between the two voltage sources falls within the range of the offset + the voltage drop due to the load, this circuit will exhibit undesirable behavior in the form of oscillation between the two sources or the short-circuiting of the two sources because both MOSFETs are open. As long as the voltage difference between the sources is greater than the total of "offset + the voltage drop due to the load", this circuit will work correctly and will always select one of the two sources.

Any additional comments are welcome.
Best regards,