I think I’m gonna try to design a barrel jack card but I’ve never designed a pcb before so idk how that will go. I found a pretty nice mcu with a pd phy and a built in comparator which could be used for overvoltage protection (maybe idk). I know almost nothing about hardware stuffs tho I mainly just like reading datasheets lol
Also idk if you can advertise only 20v with pd or if you need to have all the pd voltages
You’ve described the connector you want to use, but not the purpose of the card. Is the intent for this to be an alternative power supply (charging) card? If so, that’s not going to be easy, as the card will have to act as if it was a USB “programmable power source” (charging controller), not a PD device. This will likely require quite a few parts and some way to dissipate heat.
Definitely curious where this is going. Imo I would have probably first gone with tearing down one of the off the shelve adapters and seeing how they work before trying to roll my own.
That should work. If it doesn’t you can still advertise all the allowed voltage and just not deliver if anything other than 20v is ordered.
Just keep in mind: Only connect the 20v to vbus AFTER the negotiation is done.
Unless you are just telling it “I got 20v, take it or leave it” and then pass through the 20v or not (that is technically allowed and is done by some lenovo and hp old charger to usb-c plugs). That would still need some extra parts to do the negotiation but no heat dissipation.
The easier bit would be to stick in a “powerbank controller” which can be had off the shelve but that would add an unnessecary conversion step and heat. Also you probably won’t find a 1002 one that has any chance to fit in an expansion card.
On the other hand doing something like this may be somewhat unwise as a first electronics project.
The framework 12 requires a programmable power source capable charger? I’m gonna cry if the charger i ordered doesn’t work with my fw12
That was the idea
My life in a nutshell
Edit: The idea was to have jlcpcb assemble it for me (after somehow designing my first ever PCB lol)
Definitely let me know if you can make this work, I am especially curious about the software side.
Just be careful when testing (if you are using the framework as your test victim, if the 12 is anything like the 13 the charge controller is never told to care about the input voltage so you can use a 5v source during testing and still offer it 20v). Also please use a current limited power supply and don’t blast it with an unprotected battery or way overpowered power supply.
Btw where are you getting your 20v from? If you are using a buck converter anyway it may be a lot easier to just use a keyed 4pin connector (like a phoenix one or something) and just wire the vbus and cc pins from an usb-c power supply, so you basically got usb-c over a different connector, still weird but would take a lot custom electronics and no code.
Doing pcba with your first pcb design is also adventurous, at least in my case my first pdb designs were definitely not machine asemblable. On a positive not it does force you to do the pdb design right.
I don’t know whether that is true or not, but it would be wise to plan for it. At a bare minimum your board will have to indicate the voltage and maximum current that it supplies, even if those are not adjustable.
I might be wrong but that kinda sounds like a diy pd psu with extra steps
Dell LA65NS2-01
I don’t think at a minimum that there would be many more components other than a c male connector, the mcu (it has an internal clock), the barrel jack female connector and some mosfet. So far the progress is i got distracted with trying to run rust on the mcu lol
I was under the impression your issue was with the connector and not the protocol so a custom pd cable with a non usb-c connector at the end would do the trick no? (you don’t need a special psu just special cable and a passive expansion card to plug it into.)
Why that one? If you are going through the trouble of doing cutom stuff you might as well get a propper 100w psu.
The thing with pcba is that your design needs to be a lot cleaner than what you could get away with self asembling, that isn’t nessecarily a bad thing but it certainly doesn’t make thing easier.
I suspect the goal here is to be able to use off-the-shelf 19-20V laptop power bricks that don’t speak USB at all. I’m not sure why that would be useful, since modern USB-C power bricks are usually smaller and more efficient than the older types were, but it wasn’t my idea ![]()
Well, the mcu will need a 5V or 3.3V regulator, and then there will need to be some interface chips to protect the mcu from ESD etc on the USB-C connector and the regulator will need to dissipate some heat unless you use a switching regulator which will involve an inductor, along with some tight PCB layout requirements and … and …