If you’ve been following along (either on the newsletter or the YouTube channel) you’ll know we’ve been experimenting with high-voltage circuits to drive these LED filaments I salvaged from some broken light bulbs.

The circuit is pretty simple: it uses a MOSFET, an inductor, a diode, and a smoothing capacitor. We’re switching the inductor on and off at a high frequency using a PWM signal and we can vary the pulse width to change the voltage level.

These filaments start to light up at around 90 volts, and we can go up to about 95–96 volts for a fairly bright output. My meter only starts registering current above 20 milliamps, so it doesn’t pick up the lower values.

The circuit works quite well, but has no safety features - there’s no over voltage or over current limiting on either the output, or the input. So it’s tends to blow itself up quite readily.

Here’s the remains of a capacitor that blew up. It was rated for 100 volts and it really didn’t like gong higher than that…

I’m not sure if the failure was due to the diode or the MOSFET, because I’ve had issues with both. Here’s a failed MOSFET—it’s now acting like a resistor divider, which is not ideal.

So, I decided to invest in a proper DC-DC converter. I picked up this one, which can output anywhere from 45 to 390 volts. If you use both positive and negative rails, in theory, you can get up to 800 volts.

I also picked up a nice meter that can measure up to 200 volts and also does current.

Unfortunately, this one’s dead. It was working at first, though it made a strange clicking noise—and then just failed. Now the input is shorted.

This particular DC-DC converter board can get quite hot. I’ve seen other people mention that it can get too hot to touch. There is a connector for a fan, which I might have to use - if I can get a board that actually works!

Looking at this board, it’s… rough. My soldering isn’t perfect, but this definitely isn’t great.

After cleaning it up with a bit of 99% pure alcohol - nothing seems to be particularly wrong - yes the soldering is not great - but it all looks to be connected ok.

I have managed to find a schematic online:

Looking at the components, we have:

• 78L09, a 9V regulator - power supply for the switching IC and other components

• UCS3843B, switching supply control IC

• LM393, op amp

• A few diodes and a zener reference voltage.

On the output side, we’ve got two diodes.

I tested both of these, the one for the positive rail seems to be fine. The one for the negative rail is pretty much a dead short.

On the input side, the MOSFET - is also dead. Drain and source are essentially shorted - looks very much like some of the MOSFETs I’ve killed myself…

I’ve ordered a couple of replacement boards. Hopefully they will arrive soon, and we can finally do the experiment we set out to do: lighting those LED filaments properly.

You can watch the video of me troubleshooting this board here:

High voltage!

And the previous video is here:

Broken Lightbulbs

Related Posts

Whats wrong with my lightbulbs - In the saga of my beleaguered light bulbs, I've found myself knee-deep in electrical components and a fair bit of frustration. Of the four LED bulbs I invested in, only one survives unscathed, while another's flickering demise is imminent. My investigation revealed overheating bases, resilient capacitors, and an interesting LED filament setup. Ultimately, it seems one LED strip in the bulbs went open circuit, setting off a cascading failure. It's been a journey through minor electrical engineering heartbreak, with some messy disassembly and a sprinkle of high voltage excitement!

The PCBs are in production - what have I messed up? - After some stress and trepidation, I finally took the plunge and sent my PCB design off for manufacturing. My design centers around building a large seven-segment clock with LED filaments. Jumping hurdles such as voltages, pin usage, and limiting the load on my power supply, I've settled on the ESP32 as the system's heart and come up with a final circuit design. While doing this, I've quickly realized I could improve my layout and fixed a small mistake. Also, I've prepared for either types of LED filaments - the high-voltage ones or the larger, 3v ones. However, I did bungle up a couple of things on the enable line of the shift registers and board layout. But hey, this is a learning curve, right? Can't wait to get the boards and see what other exciting errors surface!

Dave come see - it's beautiful man - Hey folks, I revisited my flame lamp project from 2020 and gave it a serious upgrade - we're talking a leap from 300 to a whopping 1024 LEDs thanks to these nifty flexible WS2812 panels. I've kitted it out with a versatile PCB from my Crystal Balls project, tackled power injection challenges and kept things cool with a current-limiting power supply. I had to wrestle with 3D printing and getting the panels to behave inside the lamp's tube, but the end result? A mesmerizing blaze of colors with gorgeous fire effect, all orchestrated by the WLED software. It's like my very own Phoenix rising from the 2020 ashes. Curious to see this electronic inferno come to life? Check out my videos for the full, fiery scoop!

Hue Light Hack - In a fun-filled deep dive into a Philip's Hue Color light bulb, I disassembled, analyzed, and ultimately restored it to working order. Over the course of the hack, I dismantled the bulb’s interior, from the aluminum PCB to the tiny ICs on the logic board, gave a detailed breakdown of their functions and even reverse engineered the signals controlling the colors. Most of the colors worked out pretty well, except for the dodgy yellow. Also, I tried running the hacked bulb setup with my old Moonlamp project. It was a great challenge and an exciting journey!

I broke it and then i fixed it - Started off on a techie adventure with an RP2040 DVI board from Adafruit, aiming to get the pico-zxspectrum going. Things started acting haywire once the USB hub got involved, ending in a complete shutdown, intuitive diagnosis showed the power pin was a ghost town! After pin-pointing (and removing) the dead IC, I performed a delicate soldering operation to bless it with a new 3V3 regulator. Working the kinks out with the resolution led to a thrilling revival and, voila, we were back in business with DVI output and SD Card support!

Related Videos

⚡️High Voltage!⚡️ Possibly... - In this video, I tinker with high-voltage circuits to light up LED filaments salvaged from broken light bulbs. Using a makeshift setup with a MOSFET, inductor, diode, and smoothing capacitor, I explore switching at high frequencies to achieve bright light. However, limitations with low-voltage capacitors led to some explosive results. I also experiment with a DC/DC converter to reach higher voltages, but it too failed, leading to a troubleshooting session under the microscope. Despite these setbacks, I'll keep tinkering and waiting for replacement parts to continue my experiment.

Not what I was expecting! Why are all my lightbulbs failing? - So, we've been dealing with a little epidemic of light bulb failures in the house. I took apart a couple of dead bulbs to get to the bottom of what's going on. Turns out, one of the bulbs is kind of flickering and faintly working, while another bulb's base gets surprisingly warm. After some poking around, I found the PCB inside the bulb has a bridge rectifier and a LED driver IC, but all components test out fine! The investigation took a turn when I checked the LED filaments with a makeshift power supply, and surprise surprise, some filaments are completely dead. Still not sure what's causing these failures, but it was fun to tinker!

Boosting USB 5V to Light Up LED Filaments: Real-Life Testing & Circuit Simulation - Learn how to build a USB-powered LED filament circuit using an ESP32 microcontroller and DC boost converter. This project explores the use of a gate driver, inductor, diode, and capacitor to create a circuit that can power multiple LED filaments from a 5V USB supply.

1024 LEDs - WLED Flame Lamp - Today we're elbow-deep in LEDs for another exciting project — we're building a high-resolution flame lamp! Kick-starting the endeavour with a 1024-pixel LED matrix, we proceed to throw in the towel with a faulty old flame lamp boasting a mere 300 LEDs. Mislabelled PCBs and questionable diodes keep us on our toes, making this undertaking a little more Sparky than usual. With the help of competitive soldering skills, a beefy 5V power supply and a 3D printer, we channel our inner Edison to bring our LED matrix to light. After a pinch of struggle, our LED matrix rollercoasters into a tube shape, faintly resembling our coveted flame lamp. Tune in to bask in the warm (or cool) glow of our LED flame lamp.

DIY SMD Wireless LED Lights: Better than Store-Bought? - Learn how to create sleek, wireless LEDs using surface mount components! This tutorial demonstrates the soldering process for a 0503 inductor package, 1206 size LEDs, and capacitors, providing an attractive alternative to through-hole components.