Everyone needs a bench power supply, and rolling your own has almost become a rite of passage for hackers. For a long time, the platform of choice for such builds seemed to be the ATX power supply from a computer. While we certainly still see those builds, a lot of the action has switched to those cheap eBay programmable DC-DC converters, with their particolored digital displays.
This hybrid bench and portable power supply is a good example of what can be accomplished with these modules, and looks like it might turn out to be a handy tool. [Luke] centered his build around the DPS3003, a constant current and constant voltage buck converter that can take up to 40-VDC input and outputs up to 32 volts at 3 amps. In bench mode, the programmable module is fed from a mains-powered 24-volt switching supply. For portable work, an 18-volt battery from a Makita drill slips into a 3D-printed adapter on the top of the case. The printed part contains a commercial terminal [Luke] scored on eBay, but we’d bet the entire thing could be 3D printed. And no problem if you change power tool brands — just print another adapter.
Like many of the more common bench supply builds we’ve seen, [Pat K]’s more portable project relies on the ubiquitous DPS5005 power supply module, obtained from the usual sources. [Pat K] doesn’t get into specifics on performance, but supplied with 18 volts from a Ryobi One+ battery, the DC-DC programmable module should be able to do up to about 16 volts. Mating the battery to the supply is easy with the 3D-printed case, which has a socket for the battery that mimics the sockets on tools from the Ryobi line. It’s simple and effective, as well as neatly executed. The files for the case are on Thingiverse; sadly, only an STL file is included, so if you want to support another brand’s batteries, you’ll have to roll your own.
Just because something is “never used” doesn’t mean it’s good. [Inkoo Vintage Computing] learned that lesson while trying to repair an Amiga 500 and finding parts online that were claimed to be “new” in that they were old stock that had never been used. The problem was that in the last 30 years the capacitors had dried out, rendering these parts essentially worthless. The solution, though, was to adapt a modern PSU for use on the old equipment.
The first hurdle to getting this machine running again was finding the connector for the power supply. The parts seemed to have vanished, with some people making their own from scratch. But after considering the problem for a minute longer they realized that another Commodore machine used the same parts, and were able to source a proper cable.
Many more parts had to be sourced to get the power supply operational, but these were not as hard to come across. After some dedicated work with the soldering iron, the power supply was put to use running the old Amiga. Asture readers will know that [Inkoo Vintage Computing] aren’t strangers to the Amiga. They recently were featured with a nondestructive memory module hack that suffered from the same parts sourcing issues that this modification had, but also came out wonderfully in the end.
The problem is the combination of hardware typically used to run these LED strings. They’re quite bright and draw significant amounts of power, each pixel drawing up to 60 mA at full-white. In a string of just 10 pixels, the strip is already drawing 600 mA. For this reason, it’s common for people to choose quite hefty power supplies that can readily deliver several amps to run these installations.
It’s here that the problem starts. Typically, wires used to hook up the LED strips are quite thin and the flex strips themselves have a significant resistance, too. This means it’s possible to short circuit an LED strip without actually tripping the overcurrent protection on something like an ATX power supply, which may be fused at well over 10 amps. With the resistance of the wires and strip acting as a current limiter, the strip can overheat to the point of catching fire while the power supply happily continues to pump in the juice. In a home workshop under careful supervision, this may be a manageable risk. In an unattended installation, things could be far worse.
Thankfully, the solution is simple. By installing an appropriately rated fuse for the number of LEDs in the circuit, the installation becomes safer, as the fuse will burn out under a short circuit condition even if the power supply is happy to supply the current. With the example of 10 LEDs drawing 600 mA, a 1 amp fuse would do just fine to protect the circuit in the event of an accidental short.
It’s a great explanation of a common yet dangerous problem, and [Thomas] backs it up by using a thermal camera to illustrate just how hot things can get in mere seconds. Armed with this knowledge, you can now safely play with LEDs instead of fire. But now that you’re feeling confident, why not check out these eyeball-searing 3 watt addressable LEDs?
Lately, [Ken Shirriff] has been on some of the most incredible hardware adventures. In his most recent undertaking we find [Ken] elbow-deep in the core memory of a 50-year-old machine, the IBM 1401. The computer wasn’t shut down before mains power was cut, and it has refused to boot ever since. The culprit is in the core memory support circuitry, and thanks to [Ken’s] wonderful storytelling we can travel along with him to repair an IBM 1401.
From a hardware standpoint core memory makes us giddy. It’s a grid of wires with ferrite toroids at every intersection. Bits can be set or cleared based on how electricity is applied to the intersecting wires. [Al Williams] walked through some of the core memory history last year and we enjoyed hearing [Pamela Liou] recount the story of how textile workers consulted on the fabrication of core memory for the Apollo missions during her OHWS Talk in October. But giddiness aside, core memory has pretty much gone the way of the dodo having been displaced by technologies that take up exponentially less space.
We chuckle at [Ken’s] mention of the core memory capacity for the IBM 1401. It has 4000 characters of memory built-in (with another 12,000 in an expansion box) and he goes on to detail that these are 6-bit characters on a machine that operates in decimal and not binary (hence 4k instead of the base-2 friendly 4096).
You may remember his work a few years back to repair core memory on the same model. The Museum has two 1401’s, which turned out to be a huge help in trouble-shooting this. After tracing out the control lines, the repair team began swapping cards between the working and non-working machines. They were able to bring it back online — establishing one of the green inductors was bad — only to be struck with a second fault in the power supply.
Get this, [Ken] comments that “the whole computer is pre-silicon”. When working through the PSU, some suspect transistors were replaced with germanium power transistors. Those may have been a red-herring, as a penciled-in fuse on the original schematics turned out to be the linchpin of the PSU repair. Buried deep in the assembly, replacing the designed-to-fail part let the ancient beast awake once more.
Machines of this quality were heavily documented, and the schematics make this type of trouble-shooting a lot more manageable. But it’s still as much an art as it is skill. Make sure to give [Ken’s] article a read, and look around at the other repair jobs he’s documented — keeping these machines in service is becoming wizard-level work and we love being able to follow along.
A lot of the items on [Medzik]’s BOM for this build are straight from the scrap bin. The aforementioned ATX supply case is one, as is the power transformer donated by a friend. Modules such as the 30V/2A regulator, the digital volt/ammeter, and a thermostat module to control the fan at higher power settings were all sourced via the usual suspects. The PSU boasts two outputs — an adjustable 0-22 volt supply, and a fixed 12-volt output. An unusual design feature is a secondary input which uses the 22-VAC supply from a Weller soldering station to give the PSU a little more oomph. This boosts the maximum output to 30 volts; one wonders why [Medzik] didn’t just source a bigger transformer, but you work with what you have sometimes. There are some nice touches, too, like custom-printed vinyl overlays for the case.
If you have an electronics bench, it follows that you will need some form of bench power supply. While many make do with fixed-voltage supplies it’s safe to say that the most useful bench power supplies have variable voltage and a variable current limiter. These are available in a range of sizes and qualities, and can be had from the usual online suppliers starting with a surprisingly small outlay.
There is however a problem with inexpensive bench power supplies. They are invariably switch-mode designs, and their output will often be noisy. Expensive linear supplies provide a much more noise-free output, but do so at the expense of excessive heat loss when regulating a high voltage drop.
One solution is a mixed-mode design, in which a switch-mode supply does the hard work of reducing the voltage most of the way, and a linear regulator drops the last couple of volts to provide a noise-free output. [Andrei] shows us his design for just such a mixed-mode supply, and it’s one you can have a go at building yourself.
His primary supply is an off-the-shelf switcher that turns mains AC into 24 V DC. This then feeds an LTC1624 buck converter that brings the voltage down to about 1.2 V above the final output voltage, this is in turn fed to a parallel pair of LT3081 linear regulators that deliver the final noise-free output. There is an INA260 for voltage and current measurement, and an Arduino with LCD display as a user interface. His prototype has been nicely constructed using a four-layer PCB, though he suggests it could be made on stripboard with the appropriate SMD adaptors. The cardboard chassis he’s used looks slightly alarming though.