4 ms·

It would not be economical to do this unless the board is small. The PCB for the transformer is especially tuned for its performance: what dielectric is used, w

by Renaud 5y ago

It would not be economical to do this unless the board is small. The PCB for the transformer is especially tuned for its performance: what dielectric is used, what copper weight, what distance between layers, how many layers, all these are complex parameters that need to be tightly controlled. On the other hand, for the rest of the board, the PCB can easily get away with 4 layers on standard FR4 which is cheap to fabricate. Using these low profile transformers is getting common, especially if you can't afford the height of a standard transformer on just one side of the PCB. You should look into the magical stuff that powers small DC to DC power supplies: transformers entirely made of PCB, tight integration of the components inside the PCB itself sometimes. It's quite amazing really (and this is 5 years old, so imagine the progress made since): https://www.youtube.com/watch?v=K6om51cr8Mk https://www.youtube.com/watch?v=K6om51cr8Mk

yarcob 5y ago
Thanks for the video! Curiously, many of the boards in the video with planar transformers have the inductor integrated in the board rather than on a daughter board, so despite the difficulties it seems like it should be doable. Maybe for revision 3 of the POE hat.
- tomn 5y ago
  In the smaller ones, the planar transformer takes up most of the board area, so there's no benefit to assembling two different PCBs. The larger DC-DC converters are not as cost sensitive as a raspberry pi product; they tend to be used in low volume industrial/scientific/medical devices and might cost £70 each. When you consider that they are also mostly inductor by area, and factor in the additional complexity of manufacturing a multiple-pcb assembly, it's also not worth it. IIRC there are also some larger PCBs in that video; for those it's possible that they already had enough layers for routing signals that it was easy to integrate the transformer core directly. Another factor I haven't seen mentioned is the manufacturing complexity. It looks like the transformer could be placed by a normal pick and place machine, while assembling a transformer around a PCB would be more of a specialised or manual process. I suspect that might be the real reason, as the transformer PCB doesn't look like it has that many layers.

by Renaud 5y ago

yarcob 5y ago

Thanks for the video! Curiously, many of the boards in the video with planar transformers have the inductor integrated in the board rather than on a daughter board, so despite the difficulties it seems like it should be doable. Maybe for revision 3 of the POE hat.

tomn 5y ago
In the smaller ones, the planar transformer takes up most of the board area, so there's no benefit to assembling two different PCBs. The larger DC-DC converters are not as cost sensitive as a raspberry pi product; they tend to be used in low volume industrial/scientific/medical devices and might cost £70 each. When you consider that they are also mostly inductor by area, and factor in the additional complexity of manufacturing a multiple-pcb assembly, it's also not worth it. IIRC there are also some larger PCBs in that video; for those it's possible that they already had enough layers for routing signals that it was easy to integrate the transformer core directly. Another factor I haven't seen mentioned is the manufacturing complexity. It looks like the transformer could be placed by a normal pick and place machine, while assembling a transformer around a PCB would be more of a specialised or manual process. I suspect that might be the real reason, as the transformer PCB doesn't look like it has that many layers.