Parts

679 Articles

In-Depth Design Of A Flyback Converter

April 25, 2020 by 13 Comments

It is tempting to think of analogue and digital domains as entirely distinct, never to touch each other except like a cold war Checkpoint Charlie, through the medium of an ADC or DAC. In reality there are plenty of analogue effects upon digital circuitry which designers must be aware of, but there is one field in which the analogue and the digital are intricately  meshed. Switch mode power supplies use digital techniques to exploit the analogue properties of components such as inductors and capacitors, and can be astoundingly clever in the way they do this to extract the last fraction of a percent efficiency from their conversion. Thus their design can be something of a Dark Art, so it’s always interesting to have a good read explaining some of the intricacies. [James Wilson] has built a flyback step-up converter to power Nixie tubes, and his write-up follows the whole process in great depth.

This type of converter seems at first glance to be a simple step-up design with a transformer that has a primary and secondary, where in fact it relies on the collapse in magnetic field during the off period of its duty cycle to provide a spike in voltage and thus a step-up beyond that you’d expect from the transformer alone. The write-up takes us through all this starting from a theoretical perspective, and then goes further into the realm of component selection and the effects of component properties on the waveforms involved. If you have ever battled ringing in a switch mode power supply you may recognise some of this.

If this field interests you, then there is probably no better place to send you for a start than Jim Williams’ 1987 app note 25 for Linear Technology: “Switching Regulators for Poets“.

New Part Day: The MSC313E Is A Computer On A Chip

April 22, 2020 by 46 Comments

As the onward march of technology delivers ever more powerful semiconductors, it can be instructive to keep an eye on the periphery of the system-on-chip market for niche-application devices which may have an application in our sphere. Just such a chip is the Mstar MSC313E, a SoC designed for use in IP cameras that packs an ARM Cortex A7 and 64 MB of memory, 16 MB of flash, Ethernet, USB, and all the other usual interfaces you’d expect from a microprocessor. It’s available in a QFN package which makes it tantalisingly within the reach of the hardware hacker community, so naturally there is significant interest in unlocking its secrets. A cheap and accessible part with enough power to run a stripped-out GNU/Linux operating system has to be worth a second look!

QFNs are not the easiest packages to hand solder, but if you also find yourself in that position there is at least the prospect of a ready to go development board. The BreadBee is a small PCB that packs in the chip with all its interfaces including Ethernet and USB brought out for experimentation. If you don’t fancy building one, you don’t even have to: it’s soon to be crowdfunded.

One might ask what the point is of Yet Another Linux Capable Microcontroller Platform, given the plethora of Raspberry-pi and competitor boards. The answer to that is simple enough and contains within it the essence of hardware hacking: because it is there. We might never see it again save for in a few outlying projects, or perhaps it might find a niche in our world and become popular, without this early work we’ll never know. While we’re at it, this isn’t the first such SoC that’s emerged; we’ve previously seen an action cam chip give us a hand-solderable Linux single board computer.

Thanks [anonymouse] for the tip.

New Part Day: An Open Source Ethernet Switch In The Palm Of Your Hand

April 21, 2020 by 91 Comments

When you can get a WiFi-enabled microcontroller for $3, it’s little surprise that many of the projects we see these days have ditched Ethernet. But the days of wired networking are far from over, and there’s still plenty of hardware out there that can benefit from being plugged in. But putting an Ethernet network into your project requires a switch, and that means yet another piece of hardware that needs to get crammed into the build.

Seeing the need for a small and lightweight Ethernet switch, BotBlox has developed the SwitchBlox. This 45 mm square board has everything you need to build a five device wired network, and nothing you don’t. Gone are the bulky RJ45 jacks and rows of blinkenlights, they won’t do you any good on the inside of a robot’s chassis. But that’s not to say it’s a bare bones experience, either. The diminutive switch features automatic crossover, support for input voltages from 7 V all the way up to 40 V, and management functions accessible over SPI.

If you want to get up and running as quickly as possible, a fully assembled SwitchBlox is available to purchase directly from BotBlox for £149.00. But if you’re not in any particular rush and interested in saving on cost, you can spin up your own version of the Creative Commons licensed board. The C++ management firmware and Python management GUI isn’t ready for prime time just yet, but you’ll be able to build a “dumb” version of the switch with the provided KiCad design files.

The published schematic in their repo uses a Microchip KSZ8895MQXCA as the Ethernet controller, with a Pulse HX1344NL supplying the magnetics for all the ports in a single surface mount package. Interestingly, the two images that BotBlox shows on their product page include different part numbers like H1102FNL and PT61017PEL for the magnetics, and the Pulse H1164NL for the Ethernet controller.

Make Networks Wired Again

There’s no question that WiFi has dramatically changed the way we connect devices. In fact, there’s an excellent chance you’re currently reading these words from a device that doesn’t even have the capability to connect to a wired network. If you’re looking to connect a device to the Internet quickly, it’s tough to beat.

But WiFi certainly isn’t perfect. For one, you have to contend with issues that are inherent to wireless communications such as high latency and susceptibility to interference. There’s also the logistical issues involved in making that initial connection since you need to specify an Access Point and (hopefully) an encryption key. In comparison, Ethernet will give you consistent performance in more or less any environment, and configuration is usually as simple as plugging in the cable and letting DHCP sort the rest out.

Unfortunately, that whole “plugging in” part can get tricky. Given their size, putting an Ethernet switch into your project to act as an internal bus only works if you’ve got space to burn and weight is of little concern. So as appealing as it might be to build a network into your robot to connect the Raspberry Pi, motor controllers, cameras, etc, it’s rarely been practical.

This little switch could change that, and the fact it’s released under an open source license means hackers and makers will be free to integrate it into their designs. With the addition of an open source management firmware, this device has some truly fascinating potential. When combined with a single board computer or suitably powerful microcontroller, you have the makings of a fully open source home router; something that the privacy and security minded among us have been dreaming of for years.

Test Unknown Fuses Without Destroying Them

April 20, 2020 by 31 Comments

There’s a problem with fuses. On the face of it, testing would seem to be a one-shot deal — exceed the rated current and see if it blows. But once you know the answer, the device is useless. If only there were a way to test fuses without damaging them.

As it turns out there is, and [Kerry Wong] weaves quite a tale about his attempts to non-destructively test fuses. The fuses in question are nothing fancy — just the standard glass tube type, from a cheap assortment kit off Amazon. Therein lies the problem: can such cheap devices be trusted? Finding out requires diving much deeper into the technology of fuses than many people will have done, including understanding how the thermal and electrical characteristics of the fuse element behave.

[Kerry]’s test setup is simple, consisting of a constant current power supply and a voltmeter across the fuse to measure the voltage drop caused by the resistance of the fuse element. As he ramps up the current, the voltage drop increases linearly due to the increase in resistance of the alloy with increasing temperature. That only lasts up to a point, where the fuse resistance starts increasing exponentially. Pushing much past the point where the resistance has doubled would blow the fuse, so that’s the endpoint of his tests. Perhaps unsurprisingly, his no-name fuses all went significantly beyond their rated current, proving that you get what you pay for. See the video below for the tests and an analysis of the results.

It’s handy to know there’s a way to check fuses without popping them, and we’ll file this one away for future reference. Don’t forget that you should always check the fuse when troubleshooting, because you never know what the last person did to it.

Continue reading “Test Unknown Fuses Without Destroying Them” →

Magnetic Bubble Memory Farewell Tour

April 19, 2020 by 13 Comments

There’s something both satisfying and sad about seeing an aging performer who used to pack a full house now playing at a local bar or casino. That’s kind of how we felt looking at [Craig’s] modern-day bubble memory build. We totally get, however, the desire to finish off that project you thought would be cool four decades ago and [Craig] seems to be well on the way to doing just that.

If you don’t recall, bubble memory was going to totally wipe out the hard drive industry back in the late 1970s and early 1980s. A byproduct of research on twistor memory, the technology relied on tiny magnetic domains or bubbles circulating on a thin film. Bits circulated to the edge of the film where they were read using a magnetic pickup. Then a write head put them back at the other edge to continue their journey. It was very much like the old delay line memories, but with tiny magnetic domains instead of pressure waves through mercury.

We don’t know where [Craig] got his Intel 7110 but they are very pricey nowadays thanks to their rarity. In some cases, it’s cheaper to buy some equipment that used bubble memory and steal the devices from the board. You can tell that [Craig] was very careful working his way to testing the full board.

Because these were state-of-the-art in their day, the chips have extra loops and would map out the bad loops. Since the bubble memory is nonvolatile, that should be a one time setup at the factory. However, in case you lost the map, the same information appears on the chip’s label. [Craig’s] first test was to read the map and compare it to the chip’s printed label. They matched, so that’s a great sign the chip is in good working order and the circuit is able to read, at least.

We’ve talked about bubble memory before along with many other defunct forms of storage. There were a few military applications that took advantage of the non-mechanical nature of the device and that’s why the Navy’s NEETS program has a section about them.

A Broken Inductor As A Bike Chain Sensor

April 19, 2020 by 31 Comments

If you have ever broken the ferrite core of an inductor, you’ll probably sympathize with [Oliver Mattos]. He accidentally stood on a ferrite-cored component, breaking it and rendering it useless. But utility is in the eye of the beholder, and instead of throwing it away he’s repurposed it as a chain sensor for his electric bicycle.

The broken inductor was positioned on the rear frame of the machine such that the chain passed through the area where the broken half of its core would once have been. As each link passes through the magnetic field it causes the inductance to change, and from this the speed, direction, and tension of the chain can be read.

Adding a 180 nF capacitor in parallel with the inductor creates a tuned circuit, and measuring the inductance is as straightforward as firing a single pulse at it and measuring the time it takes to go negative. Chain speed can be read by sensing the change in inductance as each link passes, tension by sensing the change in inductance as the chain is closer or further away, and direction by whether the chain is slack or not. It’s an ingenious and simple solution to measuring a bicycle chain, and we like it.

A lot of bicycle measurement systems have passed our way over the years, but it’s fair to say they have been more concerned with displays than sensors.

Using Spiral Mode To Rapidly Print Enclosures

April 13, 2020 by 24 Comments

We’ve often said that one of the best applications of desktop 3D printing is the production of custom enclosures. A bespoke case adds a touch of professionalism to any project, and considering the materials needed to print one will cost less than even the cheapest generic project box, it’s a no-brainer. There’s only one problem: it can take hours to print even a simple case.

To try and speed things up, [Electrobob] has been experimenting with running off enclosures using spiral or “vase” mode on his 3D printer. Unlike the normal layer-by-layer approach, in this mode, the printer’s hotend continually rises at a steady rate during the entire print. Think of it as akin to printing out a Slinky and you should get the idea.

Spiral printed boxes may need manual retouching

As you might expect, there are some trade-offs here. For one, the walls of the box can’t be very thick since the printer is only making one pass. The nozzle on most printers is 0.4 mm, but in his experiments, [Electrobob] has found he’s able to reliably double that to a wall thickness of 0.8 mm by adjusting the extrusion rate.

You also need to approach the design a bit differently during the CAD phase. Printing holes in the side of the enclosure, which would be easy enough to do normally, doesn’t really work when running in spiral mode. For those situations, [Electrobob] recommends designing a “pocket” into the side that you can come back and cut out with a knife. It will add a little time to the post-processing stage, but the time saved during the print will more than make up for it.

So how much faster are we talking about? In the example [Electrobob] shows in his write-up, the print time went from nearly two hours to just 18 minutes. The resulting enclosure obviously looks a bit different than the traditionally printed version, and isn’t as strong, but the concept still clearly holds promise for some applications. If you’re building a sensor network that needs a bunch of enclosures, those time savings will really add up.