Apologies if the title made you want to vomit or scream "Nnoooooooo!" in the office.

My rudimentary understanding is switching power supplies will kinda fight each other causing mad fluctuations and/or smoke

I would like to supply a motherboard with up to LiPo batteries but the dominant battery to atx product (picoPower) seems to only to go to 120W. Can diodes be used to link them in parallel?

I did find DC-DC ATX supply that claimed 500W but when only one alternative exists and at a similar price I get suspicious of it just being "badge up-rated"

 

It's a bad idea...

 

Generally connecting constant voltage power supplies in parallel does not work well and diodes don't help.

One power supply will always regulate at a slightly higher voltage than the other. Depending on the design, even a few millivolts of difference can be significant. The supply that produces the higher voltage will supply all of the current until it goes into current limiting and its voltage begins to drop. Then the next highest voltage supply will start to deliver current, and so on, if there are multiple supplies. Sometimes this is OK, though it always means that the "first" supply(s) is running at full output and in consequence will be more failure prone (mostly due to temperature rise). Where it can be a serious problem is with power supplies that don't have accurate current limiting that protects them from being overloaded. Most power supplies don't have accurate current limiting, so the possibility of overloading one to the point that it is damaged is quite real. It is possible to add external circuitry to enhance sharing, but that requires that the supplies be designed to support the add-on.

A good quality battery input power supply might actually have fairly accurate current limiting, but unless the manufacturer says so I would assume not. Another issue with battery input is that the supply may not have "galvanic isolation" between the input and the output. This can cause problems when both the inputs and outputs are paralleled.

 

Thank you both for responding!

epb- Your thorough explanation definitely convinced me of what I suspected was a foolish idea.
Although my username reflects historical results not intention, I do find it important to know _why_

 

Generally connecting constant voltage power supplies in parallel does not work well and diodes don't help.

One power supply will always regulate at a slightly higher voltage than the other. Depending on the design, even a few millivolts of difference can be significant. The supply that produces the higher voltage will supply all of the current until it goes into current limiting and its voltage begins to drop. Then the next highest voltage supply will start to deliver current, and so on, if there are multiple supplies. Sometimes this is OK, though it always means that the "first" supply(s) is running at full output and in consequence will be more failure prone (mostly due to temperature rise). Where it can be a serious problem is with power supplies that don't have accurate current limiting that protects them from being overloaded. Most power supplies don't have accurate current limiting, so the possibility of overloading one to the point that it is damaged is quite real. It is possible to add external circuitry to enhance sharing, but that requires that the supplies be designed to support the add-on.

A good quality battery input power supply might actually have fairly accurate current limiting, but unless the manufacturer says so I would assume not. Another issue with battery input is that the supply may not have "galvanic isolation" between the input and the output. This can cause problems when both the inputs and outputs are paralleled.

And, worse, many of the these supplies have an overcurrent trip rather than current limit. When overloaded they shut down until the power is cycled. The consequence for parallel supplies is that they would all shut down in turn.