Its a bit of a mystery. Please can you take some photos, of the contact side of the interposer, your DB PCB ( of the contacts). And then the interposer screwed down onto your DB PCB, but separate from the laptop.
Maybe there is some conformal coating or solder mask that is holding them apart slightly or insulating them?
Maybe the contacts are not lined up?
Maybe try cleaning the contacts?
I am sure you have probably already tried all the above, but just throwing some ideas out there.
I cleaned, them, ive probed them with my multimeter to make sure there is no hidden insulating layer, ive tested the interposer with the 7700s. Its hard to tell from the image, but the pads have slight indents on them from where the interposer pins made contact with them…. Im just lost on how to trouble shoot this. i can double check my footprint but i just copied the same layout as the other side of the interposer and then change the hole locations
Maybe it is just a trick to the eye from looking at the photo, but the pins on the interposer where you outlined look more sunken in than the opposite side?
Edit: I just looked at the board and maybe that was by design as I am seeing one side looks raised and the other receeded.
There is not much obvious except that the pads are supposed to be gold, and not covered in solder. That being said, it should still make electrical contact. The need for gold is just so the contacts do not corrode over time due to “Galvanic corrosion”.
Maybe try putting some paper or similar under the PCB, so that the PCB is raised a little above the screw nuts. Maybe the PCB is just a little too thin.
If you look at the FW provided expansion bay designs, it has a 0.1mm gold pad at each connection pad, so maybe you need something to raise the pads a bit? Thus my suggestion to add paper behind or something like that.
I did read your mention of the interposer making marks on the pads, so I don’t really think my suggestions here are going to help.
i’m not an expert and you might have already tried this, but, have you checked there’s no flux or something on the pads? also ssd2_res, ssd1_res and ssd2_req look unpopulated, should they be? and the seemingly diodes below look burnt, or might be burnt flux.
The SSD pads are just for testing as i was unsure if they needed to be uses, the resistors below are 330k ohm (they do not come with any numbers on them)
what about the square pads anove those testing pads?
i think i found the issue
My right interposer pads are misaligned (purple is current and gold is where they should be) i think this is an issue when i copied the left footprint to the right, i must have rotated it or mirrored it and that caused misalignments. I will order the new boards today or tomorrow once i have fixed the issue :*(
Good spot.
To get the offset correct, look at “fw_ssd_pcb_ref_design_0.pdf” in the FW ExpansionBay github.
It shows exact measurements regarding the offset.
new board is ordered
Wishing you best of luck with the new board and looking forward to the results!
I updated the github page. look in folder v0.42 GitHub - Ether-Nova/FW16-pcie-expansion-bay
boards should arrive this Friday
What dock are you using to test? I’ve been looking for an 8i dock (4.0x8) and closest I can find seems to be 5.0x4 which won’t work (without downscaling to 4.0x4 anyway) as FW16 is not pcie 5.
no dock, just a pcie riser
Oh ok, I’ve never looked into one of those.
Wow very interesting
only issue is that you do have to break off the marked areas as they interfere with the laptop shell and there is no way to fix that without modifying the shell.
boards arrived, i was able to get the laptop to detect the board and write to the EEPROM. only issue is that whenever i connect a pcie card to the output the 12V line gets shorted so i need to figure out where this short is happening. I also broke the connector again when trying to separate the boards :(. Its brittle and hard to disconnect and i dont really have another solution
