Overall voltage swing for DC-DC converters in series

[slcf2003]

Hi, I'm trying to understand a schematic that I found online. The author managed to get a +26.5V to -26.5V swing using three DC-DC converters. Two RO-0524S (https://www.digikey.ca/en/products/detail/recom-power/RO-0524S/2314954) are used to generate +-24V and a similar ROL-0505S is for a 5V, and connecting them in series gives the desired voltages. The input voltage of these three converters are all 5V.

Please find the attached schematic. Also I think the 27.5V in the schematic is a typo.


My questions are:
1. Why are the outputs of U42, the middle converter, +2.5V and -2.5V. not +5 and 0? Is it because the circuit is symmetric?
2. Is it true that this voltage swing is achievable because the outputs of these three converters are isolated from the input 5V and GND?
3. If question 2 is true, does it mean that this is not achievable by replacing with switch-mode boost converters?

Thank you.

[David Hess]

1. Why are the outputs of U42, the middle converter, +2.5V and -2.5V. not +5 and 0? Is it because the circuit is symmetric?

I do not know why and it is not because the circuit is symmetric.  Maybe there is a rail splitter off to the right somewhere?

2. Is it true that this voltage swing is achievable because the outputs of these three converters are isolated from the input 5V and GND?

The outputs are obviously galvanically isolated from the inputs.

3. If question 2 is true, does it mean that this is not achievable by replacing with switch-mode boost converters?

The common non-isolated boost converter would not work in this application.  The converters need to have isolated outputs.

[slcf2003]

Hi David, thank you for your help.

Maybe there is a rail splitter off to the right somewhere?

The voltages are then fed to opamps as power rails, and I don't see a rail splitter there. Does the voltage swing of +26.5V to -26.5V or 53V difference make sense? Maybe all the author wanted was to generate such a isolated voltage difference for another part of the circuit.