USB type C pad distances are such that the voltage rating between adjacent contacts wtihin the plug and receptacle fall well below the IPC (or insert relevant industrial standards body here) recommended distances as voltages get pushed up over time. The original PD specification allowed for up to 20V to be delivered to a powered device, this falls under the typical threshold of ~30V peak for tight spacing present in the USB Type-C connector design.

The USB Type-C receptacle/plug can have contact-contact spacings of only 0.25mm (0.5mm pitch with 0.25mm nominal contact widths). The board side however, typically drops down to ~0.15mm on surface mount pads and annular rings on PTH pins end up approaching ~0.15 to 0.18mm as minimum soldermask webs are the only thing keeping the spacing present - if present at all.

As the power delivery standard evolved - the original 5A/20V (100W) max power capability no longer seemed adequate. Laptops with adaptors that can deliver 130, 180 and even 240W are now on the market, using barrel plugs in order to supply the required currents while keeping the voltages below 30VDC. A typical adaptor will supply around 20V and provide currents up to 12~13A. USB Type-C has found itself in a similar situation - trying to satisfy the need for increased consumer power demands while being the one-stop shop for power adaptors and high-speed data interconnects. The USB-Type C cables, pins, wires - can only support up to 5A continuous; this 5A capability comes with a lot of complexity and analysis in order to ensure the implementation within a product is safe and provides long term reliability.

The latest PD specification now allows for 28V (140W), 36V (180W), 48V (240W) - this new 48V requirement lands fairly high voltage levels next to sensitive high-speed signal pairs with minimal spacing. The IPC-2221 standard has significant spacing requirement increases once you have voltages above 30V - from 0.1mm to 0.6mm for uncoated external conductors.

The complexity of the USB-Type C specification and Power Delivery standards is something to consider. Did you know the Type-C specification now has a "liquid corrosion mitigation mode"? Now a type-C port can optionally implement a "corrosion mitigation" state... and the specification goes over how to detect liquids, test procedures, sample liquids... This 400+ page Type-C specification document and coupled 1100+ page USB PD document makes one think that the ideal cell-phone or laptop should connect power and data wirelessly. We haven't seen much in the way of a ~100-200W wireless power solution, but the complexity of a USB solution over a barrel plug is a sight.

Sometimes we do things because we can and not because we should - designs start out clean and collected (like code) and little add-ons and bits get added over time. A few hundred updates later, backwards compatibility in both software and hardware - and you've got yourself designed into something that wasn't the original spirit of the idea. How do we balance product complexity and design here? Is there an issue with having a dedicated power plug and a seperate high-speed data path? What is the problem being solved by having one connector to rule them all - instead of maybe two connectors?

Maybe like software - hardware was destined to 'stand on the shoulders of giants'. Much like how we can simply import requests - a simple module call and we can do things the appear simple on the surface but digs deep into thousands of lines of code in the background. We don't have to understand the software at that point; there's a simple API or method for interacting with the module and its fully tested right. Do we want to lose the ability to fully understand the hardware?