In the middle of the East Coast’s slow broil in the summer of 2018, a curious phenomenon surfaced. As a tropical air mass settled in and smothered the metropolitan New York area, a certain breed of stock speculator began feeling the financial heat as the microwave signals linking together various data centers and exchanges began to slow down. These high-frequency traders rely on getting information a fraction of a second before other traders see the same thing and take advantage of minuscule price differences to make money hand over fist.
While you won’t catch us shedding many tears over the billions these speculators lost during the hot spell, we did find the fact that humidity can slow microwave propagation enough to make this a problem for them a fascinating subject, enough so that we covered it in some detail at the time. While financial markets come and go and the technology to capitalize them changes at a breakneck pace, physics stays the same, and it can make or break deals with no regard to the so-called fundamentals.
So it was with great interest that we happened upon Tom Scott’s recent video outlining how one new stock exchange is using physics to actually slow down stock trades, in an attempt to gain a competitive advantage over the other exchanges. In light of the billions lost over the summer to propagation delays amounting to a mere 10 microseconds, we couldn’t help but wonder how injecting a delay 35 times longer using a “magic shoebox” was actually good for business. It turns out to be an interesting story.
Continue reading “Putting the Brakes on High-Frequency Trading with Physics”
This is the type of crowd that’s famous for building their own test equipment. If you need a way to program a flash chip, don’t go out and buy one — you can just build one. Need a spectrum analyzer? You can build that out of copper clad board. For his Hackaday Prize entry, [oakkar7] is building an optical power meter, capable enough to do futzy fiber work, but still completely DIY.
When you get into networking and telecom connections that don’t begin with the letters ‘RJ’, you start to stumble upon SPF transceivers. These ‘small form factor pluggable’ devices are little modular transceivers capable of handling fiber, Gigabit Ethernet, and other slightly weirder bit pipes. When used with fiber, they can measure optical power in dBm and watts, and can be debugged by a UART.
[oakkar]’s optical power meter uses these SPF transceivers, tied together with a fairly simple circuit consisting of an Arduino, a few tact switches, a Nokia LCD, and an FTDI UART. The key in tying all of this together is an Arduino library for SPF and DDM (Digital Diagnostics Monitoring), giving the user access to all the configuration bits in these transceivers.
While the circuit is simple enough to be built on a piece of perfboard, [oakkar] really knocked it out of the park with the enclosure on this one. With just a little bit of laser cut acrylic and a few standoffs, [oakkar] has a device that actually looks professional, and has most of the capabilities of fancier, more expensive tools.
Most of us have Ethernet in our homes today. The real backbones of the Internet though, use no wires at all. Optical fibers carry pulses of light across the land, under the sea, and if you’re lucky, right to your door. [Sven Brauch] decided to create an optical link. He didn’t have any fiber handy, but air will carry laser pulses over short distances quite nicely. The idea of this project is to directly convert ethernet signals to light pulses. For simplicity’s sake, [Sven] limited the bandwidth to one channel, full-duplex, at 10 Megabits per second (Mbps).
The transmit side of the circuit is rather simple. An op-amp circuit acts as a constant current source, biasing the laser diode. The transmit signal from an Ethernet cable is then added in as modulation. This ensures the laser glows brightly for a 1 bit but never shuts completely off for a 0 bit.
The receive side of the circuit starts with a photodiode. The diode is biased up around 35 V, and a transimpedance amplifier (a current to voltage converter) is used to determine if the diode is seeing a 1 or a 0 from the laser. A bit more signal conditioning ensures the output will be a proper differential Ethernet signal.
[Sven] built two identical boards – each with a transmitter and receiver. He tested the circuit by pointing it at a mirror. His Linux box immediately established a link and was reported that there was a duplicate IP address on the network. This was exactly what [Sven] expected. The computer was confused by its own reflection – but the laser and photodiode circuits were working.
Finally, [Sven] connected his PC and a Raspberry Pi to the two circuits. After carefully aligning the lasers on a wooden board, the two machines established a link. Success! (But be aware that a longer distances, more sophisticated alignment mechanisms may be in order.)
Want to know more about fiber and networking? Check out this article about wiring up an older city. You can also use an optical link to control your CNC.
[N8Mcnasty] is a HVAC tech who works on some big machines. One of his charges is a Carrier 19EX Chiller, rated at 1350 tons of cooling. 1 ton of cooling = 12,000 BTU. This particular chiller contained an odd LCD screen. It used a fiber optic bundle and a halogen light for backlight illumination. The system worked fine for over a decade. Now though, the halogen bulb has begun melting the glue on the fiber bundle, causing a dim display. The display in question shows some very important operating parameters, such as oil temperature, current draw, and process temperatures. Since they couldn’t easily see the display, the machine’s operators weren’t running the machine, placing stress on the other chillers in the building’s physical plant. [N8Mcnasty] tried repairing the bundle, however the glue kept melting.
A replacement display was no longer available, meaning that the entire chiller control system would have to be upgraded to a newer system. The new control system uses different sensors than the old one. This is where things start getting expensive. Replacing the sensors would also require draining the 15-20 gallons of oil, 4500 lbs of R134a refrigerant, and bringing the whole system down for almost two weeks, a $20,000 job. Rather than go this route, [N8Mcnasty] found an alternative. LED’s have come a long way since 1996, when the chiller was built. He simply replaced the halogen bulb with an LED and appropriate resistor. [N8Mcnasty] was even able to reuse the halogen bulb bracket. A bit of heat shrink tube later, and the fix looks like it was a factory option. He’s documented his fix here on reddit.
This chandelier keeps the light source hidden and uses fiber optics to illuminate the acrylic diffusers. It’s the second attempt [TheCreator] has made at building his own. Bother projects are interesting in their own way.
The first attempt used marbles as diffusers and had a much different look to it. This time around he’s using what he calls acrylic dowels. They’re not round, but square (which is why we’re not sure dowel is the right term), and he says they work better than marbles for several reasons. The marbles weren’t very heavy so they didn’t really weigh down the glass fibers to keep then straight. They were also difficult to attach to the fibers and prone to breakage.
To attach the dowels he drilled a hole in the end and epoxied a fiber optic strand in place. To direct light into the other end of the filament he built his own frustum (a pyramid with the tip cut off) of inward facing mirror. This helps to focus what is coming from the RGB LEDs in the appropriate direction so that as much light as possible makes it into the fibers.
He didn’t really give any final thoughts so we wonder if it puts out enough light for his needs. We’re sure that if it’s purely a mood piece he’s satisfied.
[Marcell] has always been turned off by the price tag of commercially available double flash adapters. He decided to see what kind of performance he could get out of a flash adapter which he built himself.
The raw materials used should seem quite familiar. The optical fibers act as a conduit to redirect the light from the flash, but he needed a way to hold them in place. He chose to use locline. It’s a product we often see in CNC mill builds to blow debris away from the cutter head. It’s hollow, and holds its position. This is perfect because it allows for easy adjustment and provides a channel through which the fibers can be routed. The Y adapters used here run to a hard board base which connects to the mounting lug on the bottom of the camera. [Marcell] suggests using a T-piece if available because the Y fitting made it a bit more difficult to push the fibers through.
After seeing a cool fiberoptic chandelier on Ebay for over $1,000, [Apex Logic] figured he could build one himself that would not only be cheaper, but have more features. Some of the features he was after were for it to be wirelessly controlled, have the ability for full RGB control, and of course to have a custom look. He pulled it off quite nicely as you can see in the video below. He has a wireless controller with 3 sliders representing RGB that you can catch a glimpse of in the second video below.
His page with the build details and the code seems to have suffered some ill fate this morning. Here it is, for when it returns.
Continue reading “Fiber optic chandelier with wireless controls”