Author: Jenny List

3295 Articles

Bricked Intel Tablet Lives Again

May 17, 2016 by 51 Comments

We’ve probably all taken a look at the rash of cheap Intel-Atom-based tablet computers and wondered whether therein lies an inexpensive route to a portable PC. Such limited hardware laden down with a full-fat Windows installation fails to shine, but maybe if we could get a higher-performance OS on there it could be a useful piece of kit.

[donothingloop] has an Intel tablet, a TrekStore Wintron 7, bought for the princely sum of $60. Windows 10 didn’t excite him, so he decided to put Ubuntu on it, or more specifically to put Ubuntu on an SD card to try it on the Wintron before overwriting the Windows installation. His problem with that was a bug in the Baytrail Atom chipset which limits the speed of SD card access and made Ubuntu very slow, and in trying to fix the speed issue he managed to disable a setting in the BIOS which had the effect of bricking the machine. A show-stopper when the BIOS is in a tiny SPI Flash chip and can’t be wiped or restored.

What followed was an epic of desoldering the BIOS chip and reflashing it, though that description makes the process sound deceptively easy. The specification says it is a 1.8V device, so after attempts to flash it using an ESP8266 and then a home-made level-shifter failed, he was stumped. With nothing but a cheap tablet to lose, he tried the chip in a 3.3V programmer, and to his amazement despite the significant overvoltage, it survived. Resoldering the chip to the motherboard presented him with a working tablet that would live to fight another day.

We’d have said that this work might reside in the “Don’t try this at home” category, but since Hackaday readers are exactly the kind of people who do try this kind of thing at home it’s interesting and reassuring to see that it can be done, and to see how someone else did it. A tablet that can be bricked through a mere BIOS setting though is something a manufacturer should be ashamed of.

We like unbricking stories here at Hackaday, something about winning against the odds appeals to us. In the past we’ve covered Blu-ray drives crippled by dodgy DRM and routers rescued with a Raspberry Pi, but the crown has to be taken by the phone rescued with a resistor made using paperclips and pencil lead.

Working With Mains Voltage: The Electrifying Conclusion!

May 16, 2016 by 28 Comments

This is the second in a two-part series looking at safety when experimenting with mains-voltage electronic equipment, including the voltages you might find derived from a mains supply but not extending to multi-kilovolt EHT except in passing. In the first part we looked at the safety aspects of your bench, protecting yourself from the mains supply, ensuring your tools and instruments are adequate for the voltages in hand, and finally with your mental approach to a piece of high-voltage equipment.

The mental part is the hard part, because that involves knowing a lot about the inner life of the mains-voltage design. So in this second article on mains voltages, we’ll look into where the higher voltages live inside consumer electronics.

Continue reading “Working With Mains Voltage: The Electrifying Conclusion!” →

Making A Fixed Voltage Power Supply Adjustable

May 16, 2016 by 33 Comments

Switch-mode power supplies are ubiquitous. Standard off-the-shelf modules in a consistent range of form factors available from multiple manufacturers. Globalized manufacturing and trade has turned them from expensive devices into commodity parts, and they long ago replaced iron-cored transformers as the go-to choice when a high-current low-voltage mains supply is required.

[Lindsay Wilson] faced a power supply problem for a motor he was working with, it required 7.4V and no off-the-shelf power supplies were to be found with that voltage. His solution was to take a 12V supply and modify it to deliver a variable voltage so he could dial in his requirement. A Chinese-made 12v 33A switch-mode supply was purchased, and he set to work.

In the event he was able to design a replacement feedback divider incorporating a rotary potentiometer, and achieve a voltage range of 5 to 15V. A small LED voltmeter mounted next to it in the PSU case gave him a very neat result.

Modifying a switch-mode supply to deliver a different voltage is a well-worn path we’ve covered at least once before. What makes Lindsay’s article worth a read is his reverse-engineering and examination in detail of the PSU circuit. If you’d like to learn more about all the different facets of design that go into a switch-mode PSU, it’s a detailed yet readable primer. We’d suggest reading our recent series on mains and high voltage safety before cracking open a switch-mode PSU yourself, but even if you’re never going to do it there’s something to be gained from knowing in detail how they work.

We’ve featured [Lindsay]’s work here at Hackaday a few times over the years. Check out his ultrasonic transducer power supply, which might be of use were you were building the ultrasonic soldering iron we featured not long ago, his laser stripping of ribbon cables, and his tale of decapping a USB isolator chip.

Electronic Driver Replaces Master Clock

May 15, 2016 by 13 Comments

In these days of cheap microprocessors and easy access to accurate timing through NTP or from the likes of MSF, WWVB, or DCF77, it’s no problem to ensure that any number of clocks keep the same time. In a simpler age though they didn’t have these tools at their disposal, so when a large organisation wished to ensure that all its parts ran on the same time they used an electromechanical solution. A master clock of as high a quality as the clockmakers of the day could build was fitted with a microswitch. The switch would send pulses to slave clocks which had a solenoid where a traditional clock has a pendulum. Thus every clock in the system lost or gained time at the same rate.

[Edo Lelic] has a rather nice Iskra slave clock, but unfortunately not the master that once drove it. Undeterred by this setback, he’s created an electronic driver board that generates the required 100mS pulses. His weapon of choice was a PIC microcontroller and an H-bridge driver to deliver their required voltage and polarity. The clock was designed to accept 100V pulses, but since it has an internal series resistor he determined that the solenoid was happy with a mere 24V. Source code is available, downloadable at the bottom of the linked article.

These clocks are an unseen piece of technology that is disappearing without our noticing. If you find one – or even better if you find a master clock – you’ll find it to be a very high quality timepiece indeed. A master clock would be well worth snapping up. At least now you won’t have to look too far for a driver for it.

We haven’t seen too many projects like this here at Hackaday. Save for a rather nice digital master clock build, it’s uncharted territory. Almost justification for a Retrotechtacular piece, perhaps.

Thanks [Muris Pučić] for the tip.

Classic Amplifier Reborn With Modern Transistors

May 15, 2016 by 26 Comments

Someone brought a dead Marantz amplifier to [Lansing]’s attention, a rather nice model from the 1980s with one channel entirely dead and the other very quiet. His account of its repair is straightforward, but provides some insights should you find yourself with a similar item on your bench.

Opening up the box, he was presented with 35 years of accumulated dust. It’s the annoying side of cracking open classic kit, we all have our dusty horror stories. His first task was routine: to replace all the unit’s capacitors. The mains voltage in France has gone up by 10 volts from 220V to 230V as part of EU harmonization in the years since the amp was built, so he used capacitors with an appropriately higher rating to compensate. We might have waited until the rest of the amp was proven fixed before splashing the cash on caps, but maybe we’re more thrifty.

The quiet channel fix turned out to be from a muting circuit designed to keep the amp quiet during the turn-on phase and suppress that annoying “thump”. A dead transistor replaced, and all was well. The dead channel though had a whole slew of dead transistors in it, which turned the problem from one of repair to one of transistor equivalence. Quite a few of the 1980s parts were no longer available, so modern replacements had to be found.

It is tempting to think of particularly all small-signal transistors as functionally equivalent. You will get away with this in logic and switching circuits in which the device is either On or Off and never in between, but in an audio amplifier like the Marantz things are not so simple. A lot of effort will have been made by the designers to calculate resistances for the current passing through them to deliver the right DC bias points without sending the circuit into wild oscillation. An important part of that calculation comes from the current gain of the transistors involved. [Lansing] had to carefully select his transistors for equivalence, though it some cases he had to do a bit of creative lead-bending to fit a different pin-out.

So, all dead transistors replaced with appropriate equivalents, and the amp was reborn. Success, and very much worth the effort!

We’ve covered a lot of amplifiers here in the past. Some were dead, like this little amp with blown capacitors or this smokin’ subwoofer. Others are more esoteric, like this ion wind 1KV tube creation.

Taming A Variac With A Thermistor

May 15, 2016 by 18 Comments

The folks at TOG, Dublin Hackerspace, have a large variac. A variac is a useful device for testing some fault conditions with AC mains powered equipment, it allows an operator to dial in any AC output voltage between zero, and in the case of TOG’s variac, 250V.

Their problem was with such a magnificent device capable of handling nearly 3KW, it presented an inductive load with a huge inrush current at power-on that would always take out the circuit breakers. Breakers come with different surge current handling capabilities, evidently their building is fitted with the domestic rather than the industrial variants.

Their solution was a simple one, they fitted an NTC surge limiter in series with the variac input. This is a thermistor whose resistance falls with temperature. Thus on start-up it presented an extra 12 ohm load which was enough to keep the breaker happy, but soon dropped to a resistance which left the variac with enough juice.

This is a simple fix to a problem that has faced more than one hackerspace whose imperfect lodgings are wired to domestic-grade spec. In a way it ties in neatly with our recent feature on mains safety; making the transformer no longer a pain to use means that it is more likely to be used when it is needed.

Via: TOG, Dublin Hackerspace.

Chibiterm Is A Tiny Low-Cost VGA Terminal

May 13, 2016 by 22 Comments

A common sight in the days before cheap PCs conquered the world was the dumb terminal. A keyboard and a monitor with a serial port on the back that was usually hooked up to a minicomputer or even a mainframe, these were simple devices. Anything that came into the serial port was rendered on the screen, anything typed on the keyboard was sent out through the serial port. They didn’t need to contain a microprocessor. If you are old enough, you may remember electronics magazines of the 1970s and early 1980s publishing terminal designs based entirely on 74 series logic.

The serial terminal might seem like a redundant historical footnote when viewed from 2016, but they can still find a use among those working with systems such as small embedded microcontrollers that only possess a serial port. To address this application, Hackaday.io user [K.C.Lee] has created a low-cost terminal module for a VGA monitor and a PS/2 keyboard based around an inexpensive STM32F030F4 processor.

Continue reading “Chibiterm Is A Tiny Low-Cost VGA Terminal” →