Built one of these back in 2007 to hook up to my WiFi card in the barracks in Japan. Hacked the card a bit to push the power higher than allowed (firmware) and pretty much had the best signal ever for internet, at the cost of everyone else near me. Note that the AP was about 20 feet from my window, so I was just trying to drown everyone else out, not extend my range.
Got some flak from my 1st Sergeant at the time, but that was about it. At least I could raid in WoW, unlike everyone else on that godawful WISP.
It’s kind of a trip seeing people noodling around with stuff I hacked together 20+ years ago showing up on the HN front page at the same time. i enjoy kismet.
Where I live, we only have wireless internet options, so antennas are critical infra.
Interestingly, its sort of hard to find antennas optimized for specific frequencies. Most affordable options seem to be "wide band" generalist antennas, e.g. 900mhz-6ghz. Meanwhile the nearby tower might have the best signal on a specific band.
I looked into making log periodic/yagi antennas and it seems surprisingly simple and cheap. It seems possible to make something out of PVC and copper wire that can outperform a commercial antenna costing several hundred or more.
which I think makes very clear the principle of a Yagi-Uda antenna which is that different sized antennas are above or below resonance so they are behind or above the input signal in phase. The central antenna is driven, the other two are not, but when the frequency is right the phasing of the antennas is such that the signals reinforce each other in the desired direction.
What threw me the first time I built one is the "hairpin" impedance matching network which is a single-turn coil that looks to me (from the viewpoint of DC electronics) like a short circuit so it took a bit of faith to hook my radio up to it and hit the "transmit" button.
Amazingly though I was able to talk to people in Rochester, NY with my 5 watt handheld and that antenna using a repeater on top of a ski resort mountain 80 miles away.
I built one of these when at university in 2008. The AP was up the other end of a long hallway and through many rebar concrete walls. Whenever I needed to download a dataset I would stick it out in the corridor.
This and a bunch of WRT54Gs bought for five dollars each because they were bricked let me do wifi for my whole apartment building while I was a senior in uni living off campus. Got the best connection I could afford from comcast and basically I became a mini ISP. I charged five bucks a month to 20 people and it paid for my own connection basically. The trick was that my own packets had priority :)
I built one of these in ~2004 to get internet from a Flying J a little over a mile from a house we were renting. Used a pringles can for a bit but a large can of yams worked much better. Used it for a few months until our neighborhood got actual internet service (and we lost line of sight to the truck stop)
I still have a couple of cantennas that I made, oh, about 20 years or so ago. There was a coffee shop half a block from me that didn't have any WiFi (owner behind the times), so I set up a WRT54G with DD-WRT as a client bridge, hooked a cantenna to it, and pointed it at the coffee shop. With the cantenna at the other end, I was able to get a solid signal.
These days? With mobile data anywhere I'm likely to go, that's more trouble than needed. Aside from that, modern WiFi isn't really conducive to external antennas.
Wavelength (mm) ≈ Velocity of wave (Mm/sec) / Frequency (GHz)
w = v / f
We know v will be 300 (rounded up based on the speed of light mentioned above). For f, we need to plug in the Wi-Fi frequency.
You could just use 2.4, but in order to be a little more accurate, we will use two more decimal places. For channel 6 in the 2.4GHz spectrum, we need to plug in 2.437. Solving for w yields ~123mm.
w = 300 Mm / 2.437 GHz
w = 123.102175
Now that we know the (w)avelength for our radio (f)requency, we can begin calculating the dimensions of the can based on the guidelines mentioned previously.
We can change the units to mm*GHz, then the division by GHz cancels out and we end up with millimeters. I studied this in the USA, so the number we learned was 11.8 GHz*inches.
> this was one of the things that got me bitten by the "working with computers" bug.
Do you ever regret being bitten by this bug? I was bitten by it as well, at an early age.
It worked out OK for me in that I've made lots of money, but I feel that it's come at the expense of possible greater impact and glory should this neurodivergent mind have latched onto something more "real" instead.
I was originally going to be a mechanical engineer...but the early time of the digital age put its (more immediacy-driven or maybe dopamine-driven?) hooks into my brain...and, so I finished my university major in the realm of computers. I can NOT complain, as i have made a great life for my family through my career...But now, as i am in my late 40s/early 50s...i wonder if maybe instead I should have had either a non-tech career or maybe only a slightly tech-adjacent (but not exactly tech hardcode) career, and then simply focused on computer stuff only as a hobby/pursuit/interest instead?
I know that i am not alone in this feeling...as i have heard others my age - and who were orignally not going to "go into tech as a career" - cite similar feelings. Maybe its burnout, or maybe i feel that too much intelligence in tech is wasted on getting silly clicks and things (as the famous saying goes about this age's greatest engineers, etc.)...or maybe a combination of many factors (not excluding age). Anyway, yeah, others have such a feeling about *working* with computers. ;-)
I would like to hear of this saying about this age's greatest engineers. I have never heard it, but I have often thought of the tremendous amounts of brainpower wasted on the silly and ephemeral in IT.
Apologies it's not specifically engineers...it refers to "great minds", and is attributed Jeff Hammerbacher: "The best minds of my generation are thinking about how to make people click ads. That sucks." (https://www.goodreads.com/quotes/747678-the-best-minds-of-my...)
Ah. I have a similar, if longer, itch about the tech brain drain, where we have taken people who might have been scientists, or physicians, or non-computing engineers and we have them doing any number of, at the space of five years or so things which are wasteful. We have brilliant people reporting to the mediocre and attempting to appease the foolish.
Yes, but only for the stress it causes me in thinking really really hard all day and then being exhausted at night, which i think is not unique to a computers career. I don't regret it in that it's an incredibly high paying job with very low social barriers to entry.
Man that takes me back. I used to torrent my media with a giant Yagi antenna mounted to a tripod that'd let me connect to the McDonald's down the road.
Chances are, at long ranges, one will be much better than the other. 2.4ghz if you have some obstructions. 5ghz if you have lots of height on one or both sides (it’s less tolerant of line of sight and ground obstructions but more gain for the same size)
At least wireless ISPs don’t bother with spectral diversity on consumer fixed links.
Predicting constructive/destructive interference… you’ll need to do some radio-mobile plots (I think that was the software with topographic maps built in).
If you have an LTE modem or hotspot with a removable antenna, you can do this sort of thing nearly exactly the same, just need to tune the antenna for different frequencies.
Phones tend to not have swappable antennas so you’re limited to doing something like using a parabolic dish to concentrate signals, and positioning the phone at the focal point.
Maybe you could construct something like a yagi antenna using a phone as the driven element…
Got some flak from my 1st Sergeant at the time, but that was about it. At least I could raid in WoW, unlike everyone else on that godawful WISP.