Pull the USB lines entirely and just leave CC1/2 and VBUS and GND.
2 Likes
Most magnetic chargers are power only.
1 Like
yea i dont see any point in having data transferred over a magnetic attachment. if im moving a large file and that magnet slightly loses connection, which it often does when i move the laptop around while charging, id be pissed. id rather directly wire into usb c for max speeds and connection stability for anything else but charging
4 Likes
My dream is a eGPU dock that connects with a single, magnetic cable. There are multiple options of magnetic USB C adapters that at least advertise 40Gb/s transfer speeds. I would want to test to make sure it lives up to it and is reliable before I go through a whole design process. Not going to be able to build one of these for a few months anyways. In theory, as long as you have the all the pins and they all have consistent contact, you’re good to go
Docking station.
All day every day I have power, gigabit, 2 external monitors, and all miscelaneous usb devices all going through a single thunderbolt port, through a magnetic 90-degree adapter.
And if the adapter were flush, or even just a bit closer to flush even if it still bumped out a little, that would mean I don’t have to remove the adapter when leaving the desk. Just leave it in there all the time and have the same magentic cable-end on the charger in my laptop bag.
3 Likes
Non-profit is great and adding the person’s time at a reasonable rate is well within its bounds. E.g. non-profits pay their employees’ salaries.
3 Likes
Seconding this. Transfer of data over magnetic connection is scary from a signal integrity standpoint, and it’s downright terrifying if that’s one millimeter away from USB PD’s 20v5a. I only use my magnetic card for charging, but that only mitigates half the risk. I still have connected data lines after the PD negotiation, and they’re still 1mm away from each other.
One bent pogo, one little component lead clipping (I do a lot of hobby electronics work) or metal filing accidentally magnetized to either side? zap. I’d much rather have the bare minimum.
1 Like
It’s very possible, but terrifying. You could probably find someone on here or on egpu.io that would be willing to test, I’ve seen people use these connectors with TB4 docks but not eGPUs so far. I have the equipment to, but it feels entirely too risky for me.
1 Like
I might be stupid here but why don’t the manufacturers of these magnetic usb-c connectors just space the pins more out? Like Microsoft and Apple do with their connectors.
I know it’s supposed to have a similar footprint to a normal usb-c cord but like adding a couple of millimeters overal to the connector to ensure you don’t short your data and 20v lines goes a long way
1 Like
-
there’s no 20V line, there is a 5V line. Only once a PD or QC communication has been established will a higher voltage be provided.
-
I would imagine on the computer side there is a capacitor to stop DC from effecting the data stream.
-
DC and data can travel the same line without problems. A 5V dc can carry modulation as data.
Small magnetic connectors are more 'handy
Here’s an image of two I use with the same cable, USB A and Lightening available 
On the USB C the space between the connections is less than 0.5mm
USB C and USB Micro
There is a possibility of data corruption if I am not careful with the cable, but a) I ma careful and b) if I am messing around I transfer by Wi-Fi fo example with ADB to a phone.
People have been using them for years with no problems I have heard of, not there hasn’t been any
Oh sorry! I assumed they were just hooked up as supplied in the schematic
The more space was kinda stupid yeah, even my Surface adapter isn’t that far apart.
The real solution to this would just be Apple and Microsoft not to hold trademarks on a magnetic connector so Framework make a first party one that’s just handshake and power connectors like the ‘new’ Magsafe connectors have.
I somehow didn’t feel comfortable using these connectors even though they’re super cool and literally have no reason to. - BTW project is awesome!
1 Like
Just preordered my framework, wondering if there are still any concerns with the mag-charge card that @Moe_Wigs designed. If I were to print/assemble the expansion card and keep the data lines intact, would it be safe/reliable for just power delivery, or would it be a good idea to cut the data lines/add diodes or something? To clarify, I don’t intend to ever use the magnetic connection for data transfer, only for charging, but I’m useless when it comes to soldering/electronic stuff, so I would like to both have it safe and also do the smallest amount of modification possible.
I’ve been using this magnetic adapter from Amazon the last few days. It does the job but it sticks out a bit when inserted.
It seems the new style has gone this way - it appears smaller in diameter than the one posted here.
I’m not an EE or very computer savvy but it does say 24 pin which I think is the current standard, correct?
I’m wondering what the difference between that and the one similar to this thread is, as the one above is more prolific it seems.
1 Like
That’s completely fair, I have my reservations about them too and would much prefer a magnetic connector with only VCC/CC1/CC2/GND, but others use them as they wish, and I still use mine lol.
1 Like
I have the second one, it’s from an earlier card design–from what I’ve seen of the first kind, my type is skinnier and when connected to the laptop, it won’t contact the table. People have complained about the newer version (first type in your post) not allowing charging while flat on the table because it contacts the table before the laptop rubber ‘foot’ does. I believe the reason for the switch was the exact amazon listing for the first kind going down, so any others and ymmv.
If you find another listing that will work with the original card design at the top of this thread, feel free to let us know!
@inline_five that 24 pin connector should be fine. Something to note when it comes to cables/connectors is unless they are an adapter, it’s just pins on one side lining up with pins on another. The thunderbolt standard doesn’t come into play until you get to the devices. Thunderbolt 3 and 4 use 24 pins.
I’m going to try using the 24 pin magnetic connector on an eGPU enclosure. If that works out, I’m going to try and make an expansion card that makes the magnetic connector flush with the side. I have to balance that side project with school though, so it could be a while
1 Like
Very cool idea in theory and I’ve seen (terrifyingly) people use these connectors with TB3/TB4 docks, so it should work, but keep in mind your signal integrity and bandwidth are going to be stunted slightly.
@Be_Far I just finished testing with the 24pin magnetic disconnect. In benchmarking, the disconnect performed 0.283% better, but I would consider that within the margin of error. No signal degradation in my case. Also, I’ve had the issue where applying slight pressure on the connection with the thunderbolt cable causes it to lose connection. This doesn’t happen with the magnetic disconnect (unless you apply enough pressure to remove the cable entirely).
I’m going to try this with a longer cable as well (6ft maybe) and see if I get the same results. If it’s good, I will try to find a solution that makes the magnetic connector flush with the side instead of sticking out. Might have to design my own PCB, which would be a little scary. I don’t think you can buy what is effectively a tiny spacer that is a 24 pin type c to type c that is just the right length.
Gotta balance this with school though
Also, I don’t know how to reply to replies 
1 Like
Highlight the text in post you want to refer to and you will see a [Quote] option that will start a new post with that quote and the author
So they’re effectively the same performance wise. My anecdotal old-wives’-talery is wrong then, good to know!