In the part of my house where the signal is the worst, nmcli was reporting signal strength in the 40s using the Molex antenna. I suspected I might be able to get better results with bigger antennas. I began conducting some casual experiments in that spot using these ones today: https://www.amazon.com/gp/product/B08T8BFBCD
In line with Chartreuse’s ideas, it seems distance from the PCB and screen may be the biggest factors determining how strong the received wifi signal is. With the 8dBi antennas taped to the back of my screen, I was seeing signal strength in the 50s and sometimes 60s, though I couldn’t actually load webpages reliably and I still lost signal entirely pretty often.
Taping them to the underside of the laptop, outside the acrylic, yielded similar results.
Letting them dangle freely, however, boosted signal strength into the 70s and 80s, as did taping the antennas to the sides of my keyboard at somewhere around a 75 degree angle. Even with them taped alongside the keyboard, covering the ports, I was seeing figures in the 60s and 70s.
My inclination at present is to design and 3D print some mounts for the antennas to allow flipping them out from the sides of the keyboard when necessary and then tucking them away when they aren’t in use.
Took about three redesigns, but I’ve found something that works with minimal modification to the Reform case.
Uses a pair of M2x6 screws, replacing the M2x4 ones toward the back on the port panels, and two 8x4mm magnets. I also had to file notches in the power plug and HDMI openings in order to route the wires into the chassis.
There are probably still some rough edges to the design that could be smoothed out, but I’ve satisfied my own needs with this for now. Let me know how the design works for you if you use it, and whether you make any modifications!
Edit: My mind seems to not quite be done with this problem, and I’ve thought of a new design that might be a bit better in some ways. I’ll report back once I’ve had a chance to pursue the idea.
Edit 2: The new design’s definitely worse. I’ve learned that the separation of the mechanism for extending the base and rotating the antenna is important, as the two have different friction requirements. I think what I’ve arrived at is about as good a design as I can come up with given the space and material constraints.
Edit 3: I referenced this in a post above, but for clarity: I’m using these antennas here.
Yes, this is also how I had the Laird mounted. I think my trouble might be partially down to how many other routers are around me. Just in my own house, there are three other wireless routers besides my own. I’m pretty sure my neighbor’s router is on the other side of the wall in the room I lose signal as well.
But my old laptop had no trouble getting signal in that same situation, and I figured I wanted my new laptop to be at least as sensitive to wifi signals as my old one, right?
As another datapoint: my old laptop had the same wifi card as the Reform, so I have a high degree of confidence that the difference between the two comes down to the antennas and their placement. My old laptop had the antennas in the top half of the case, parallel to and against the sides of the screen (not parallel to and against the back, which I suspect would have diminished the antenna sensitivity significantly base on my own tests with the 8dBi antennas).
Before testing different placements at Chartreuse’s prompting, I didn’t think it would make that much difference where and in what orientation the antennas were mounted, but it certainly seems to!
The antenna placement underneath all that metal is not very good for the antenna performance. But those MIMO antennas should still get decent signal. That’s why they are mostly put inside the lid.
It should not matter how many wireless devices are arpund you, since you connect to one only.
From all these range issues(phone directly at the side and only 50% signal strength) I’d rather think the antenna connectors are not connected correctly to the WIFI card
Yeah, there may be a variety of causes across all these different problems. FWIW, I swapped connectors as well to rule that out.
The number of wireless routers around you will matter, as eventually you get competition for the same channels, and (as I understand it) there can be interference from signals on adjacent channels as well.
I see 25 wireless router signals from where I’m sitting right now, spread across 9 channels.
I can guarantee mine are. 's not my first time connecting UFL antennas - and I’ve connected and disconnected them three times so far, without making a blind bit of difference.
To add to the discussion: I never found an upgraded antenna for a halfway-acceptable price, so I’m still running the bundled Molex. I did, however, discover - through a bit of trial and error - that reception was greatly improved by moving the antenna all the way to the side of the laptop, so it’s directly over the USB ports.
It means you can’t use the clever cutouts on the acrylic to hold it in place, but it gets squished between the base and the USB ports and doesn’t seem to shift around. And now I can actually get a Wi-Fi signal I can use!
Don’t get me wrong, it’s still by far the weakest device I own - but it’s usable now, and at zero cost. I’m chalking that up in the “win” column!
So maybe this is because I have my house saturated with wifi, but I found that my wireless reception was very decent. My frame of reference was a Purism Librem 13v2. It had absolutely abysmal reception.
That said, the folding external antenna are very interesting, and would be something I could see myself being interested in.
I’m still using the external antennas and am happy with them. You do have to leave the left one up in order to charge in the current design. I’ve also managed to trip over one, tearing it off in the process, but it was just the middle ring that broke (the smallest part of the assembly). Was able to print up another one and replace it in short order. Was glad for the design then, as I’m pretty sure the thin inner ring saved the bigger pieces by snapping before they could.
