MAKE Asks: Epic Fails

MAKE Asks: Epic Fails

Make: Asks is a new weekly column where we ask you, our readers, for responses to maker-related questions. We hope the column will spark interesting conversation and that we’ll get to know more about each other.

One time I accidentally shorted the transformer on one of my guitar amps, sending sparks flying. This was definitely an epic (and dangerous) fail for me.

This week’s question: What was your most epic mistake on a project you were working on? Were you able to fix it?

Post your responses in the comments section.

100 thoughts on “MAKE Asks: Epic Fails

  1. Greg Williams says:

    Here’s my latest epic fail. Built a rock tumbler I was quite proud of only to find out my motor was underpowered. I’m rebuilding now with a much larger motor, so I think I can save it.

  2. apeine says:

    Well, that should have been when I first learned how to solder. When I tried to do it, I created a pool of solder, with all resistors, transistors and IC practically together. When I connected a battery, I learned how burnt components smell like.
    20 years later, saw a video on youtube, and darn, soldering is so easy, even to get a good finish… Wish I had learned that before…

  3. oldsweng says:

    I was about 13 when I built my first transformer, in my bedroom. The primary was two wraps of 22 AWG magnet wire and the secondary was a single wrap of the same wire around an empty toilet paper roll. I then hooked a light bulb to the secondary and stuck the primary into the 115VAC. I was able to put out the fire pretty quickly and decided I needed to get a little more electrical knowledge before experimenting again. I’ve now been in the electronics industry for 40 years, but that incident still sticks in my mind.

    1. adcurtin says:

      I’m sorry, but this is hilarious. it’s just so ridiculous to imagine. Also, the thought process to know about transformers, but not to know that a cardboard tube won’t make a good transformer it pretty funny.

  4. Lear says:

    I was patterning a Mokume-gane billet (think Damascus steel… but in precious metal) and was hot-forging it on an anvil. Lost my grip with the tongs and reflexively reached out caught hot metal with my opposite (and ungloved) hammer hand.

    Two weeks later I did it again.

    I had a lovely crucifix shaped burn on the palm of my hand for about 6 months. I may be a slow learner but I always wore gloves on both hands no matter what thereafter.

  5. RockDoggy (@RockDoggy) says:

    When I built my first computer from parts (probably 20+ years ago) it came without instructions or information on how to wire it through the mechanical power switch in the case. Two leads in, two leads out, no idea which lead was which.

    In the absence of a multimeter, and being young and a tad anxious to get it running, I guessed on how to wire it. The breaker tripping in my apartment was a spectacularly loud and obvious method of finding out what a bad guesser I was. Not the smartest move, obviously.

    Luckily for me, I managed not to fry the power supply, wires, or any components, and the fire department didn’t need to be called. I believe I bought my first meter that night. The computer, once properly wired, worked perfectly for years to come. And I’ve since become a little more patient and methodical about these sorts of things, especially where electricity is concerned.

    1. adcurtin says:

      with old AT and XT power supplies, I think the mains ran right to the switch. so you probably just shorted out the mains when you closed the switch. Could’ve destroyed the switch, but not really anything else.

    2. Michael Durham says:

      When I first wanted to get into building computers was when my Gateway power supply bit the dust about five years ago…. Not realizing they used proprietary PS’s I just bought a generic ATX supply thinking I’d tapped into the grand secret underground of DIY computer wizardry and outwitted the manufacturers trying to charge too much for a replacement. This also created a loud pop inside the computer, in addition to tripping the house’s circuit breaker. I never did try a real Gateway PS to see if the MB was ruined, I just did more research about actually building my own, and I’m still using the one I built then as my main computer today.

  6. Allan says:

    Tried t oconnect a vintage Bridgeport milling machine wired for 3-phase to 2-phase,
    while inteligently shorting two. Differential switch in workshop panel flew and left a mark on the opposite wall.

  7. Allan says:

    Tried to connect a vintage Bridgeport milling machine wired for 3-phase to 2-phase,
    while intelligently shorting the other two terminals. Differential switch in workshop panel blew and left a mark on the opposite wall.

  8. Lewis Baumstark says:

    I had a monitor with a push-on/push-off switch. The switch stopped working for some reason so I wired around it to make the monitor “always on” (this was in the day when computer power supplies had an output for powering a monitor so my plan was to just use that to turn it off and on).

    Unfortunately, my first attempt at wiring ended up shorting supply power to ground, with the expected breaker trip. Fortunately I didn’t do any permanent damage to the monitor, my parent’s house, or me, and I was able to re-wire it properly. That monitor lasted several more years until I could afford better.

  9. RossinDetroit says:

    Soooo many. But one is relevant to MAKE. Take a look at the Squelette article or the online page. At the left rear is a silver button at the top of the chassis. This is a cover plug to fill a hole I accidentally made when drilling for the power switch. It was 2:00 am. Deadline loomed. I was in a hurry, drilling holes from the underside and I got my orientations reversed. I could either get more metal and spend 2 hours remaking the piece or pop in a plug and ignore it. Amazingly, nobody ever asked why that was there, and it was faithfully reproduced in all of the drawings in the article.
    Now you know.

  10. kjunkins says:

    My first work with electronics was a Sinclair computer kit. I was a horribly sloppy solderer, but the ads for the kit made everything look so easy. The backup plan was that for $20 more, you could send your kit back to Sinclair and they would send you one already assembled. So I had nothing to lose. I did all of the hard work, plugged it in, and NADA. The extra $20 kept me interested in computers, but it didn’t force me to troubleshoot and fix things on my own. Now I am a software guy who tinkers with electronics and teaching myself things I should have learned 26 years ago. It was a FAIL on multiple accounts if you take that into consideration.

  11. mental404 says:

    While building a sous vide machine a year or two back, I was having trouble with the relay not tripping properly, so I grabbed the multimeter and started testing various points in the circuit trying to figure out where the problem was. Apparently I got a little carried away with taking notes and not really paying attention to what I was doing; I went to test the last set of contacts on the PID, which didn’t quite register in my head that the were pretty much wired directly into the wall socket at that point. A short (ha!) bang-flash-smoke later, I had a dead multimeter w/ melted probe tips and a dead PID. Not exactly my most stellar moment, but a good reminder to pay attention while playing with live circuits.

    1. Mattwa says:

      Ha, that sounds familiar. I made a sous vide system myself. When installing my first PID i got the wiring diagram confused and wired the mains (240v in Australia) into the thermocouple input. Bang, smoke, and I was on ebay buying another PID 15 minutes later.

  12. barbecuesteve says:

    A couple weeks ago I was building a coil gun for a “high striker” style game using the iPhone accelerometer. The action circuit for the coil gun uses the flash from a disposable camera. In my testing, I triggered the circuit while holding a lead in each hand — sending 300v right across my chest, knocking myself to the ground, sending my eyeglasses flying, and making my left hand act palsied for about a half hour.

  13. RossinDetroit says:

    Here’s a fail that wasn’t my fault. Back in the old days just after the Earth cooled and the meteors stopped falling you could buy very high quality sophisticated audio equipment as kits to assemble yourself. It took tens if not hundreds of hours to put one together but you saved money and got bragging rights. Over the years I have fixed not one but TWO fine tube stereo amplifiers that were completed but never ran because of a single soldering error out of hundreds of joints. Finding the error was like a needle in a hay warehouse. One was soldered correctly but a tiny filament of solder splash had bridged two traces on a circuit board. I found it with a magnifying glass. I’ll never forget flicking it off and having the amp run perfectly. These units were basically rare 50 year old New Old Stock. Can you imagine working a month of evenings to make a tube stereo and just put it back in the box because it didn’t work? Both of those units ended up in the hands of happy new owners in Japan.

  14. Travis Will says:

    I was 12 or 13 I don’t remember what I was trying to make but it involved me plugging an electromagnetic I don’t know what into a wall socket without a ground. It tripped every breaker in the house and caught on fire!

  15. Andrew Huntley says:

    I actually have two that really stick with me. The first was whenever I was really young, experimenting with my father’s multimeter. I wanted to see if it worked like his circuit tester, and so I stuck the leads right into the light socket on a plugged in lamp. I managed to spot-weld the leads in place.
    The second big error was whenever I was building my first Jacob’s Ladder. I didn’t realize there was a capacitor inside the transformer, and that it didn’t discharge. It was a huge shock whenever I touched a screwdriver to the positive side, and it almost embedded said screwdriver in the wall.

    1. adcurtin says:

      I dunno, while the second one probably shouldn’t’ve happened, I don’t think it’s a fail. You used a screwdriver, and I assume you didn’t get shocked yourself. While it sucks that the screwdriver flew like that, it did do it’s purpose of making sure the cap was discharged before it hurt you.

  16. clearpaint says:

    I started out in Avionics, working on transponders and comm radios. I had learned about taking conducted measurements on high wattage output in school, but had never actually done it at this point. I took a probe and placed it on the output pin, the can was ground and of course I slipped shorting the two together. It was enough to give me a really good shock. Yeah it hurt. Luckily I didn’t damage the transponder.

    Most recent epic fail. I was working on a project with a friend and I explained to him it was important to not totally discharge the lith-ion batteries. So I thought hey he knows and went on not adding an automatic threshold shut off; wow do those things expand once you drain them completely.

  17. balloondoggle says:

    I had just spent some lawn mowing money on a yard-sale car radio – probably AM only. I took it home and “powered” it with a cut-off extension cord. All the smoke came out.

  18. RossinDetroit says:

    I once rebuilt and assembled a motorcycle engine, installed it in the frame, and the last thing to do was install the spark plug. I cross threaded the plug and wrecked the threads in the head. Had to tear the whole thing apart and mill out the plug hole for an insert. That’s a mistake I’d rather make on the bench when things are already apart.

  19. MSilvia says:

    I decided to liberally apply hot glue to a board I had just finished soldering up, in order to add some structural strength. After doing so, I seemed to have a bad connection, so I plunged my iron through the glue and re-melted the solder.

    It turns out molten solder and molten hot glue can mix into a horrible blob of difficult-to-fix goo that has alternating tiny pockets of insulation and conducting. It was a nightmare. I had to strip the board, clean the contacts on all components and holes, re-solder the board, and swear I’d never do it again.

  20. timesuptim says:

    Amplifier fail was using an old tube guitar amp as a vocals amp in the rehearsal room. Great, until it started distorting. We looked over and there was thick yellow smoke pouring out of the back of the tube amp, toxic as hell.
    Earliest experiments in 220V wiring by twisting the ends together led to fantastic fireworks and sputtering molten copper all over a bright purple vinyl beanbag (it was 1979 or so) that put me off voltages above 12 for a long long time.
    Double Ouch.
    But the real fail (same time period) is:
    Building an extension cable with piggy back plugs on both ends for added flexibility.

  21. tonyv says:

    I was trying to make an electromagnet for a cubscout badge, long, long ago. My mom bought me plastic insulated wire and a 6-volt lantern battery. The wire didn’t look like the shiny lacquer-coated bell wire in the cubscout manual, so I peeled off all the plastic insulation and then wound my coil. It looked JUST LIKE the picture, except my battery was rectangular and the one pictured was cylindrical. This was undoubtably why the magnet didn’t work, and why my battery got really hot….I had my mom get me a cylindrical battery, exasperated that she had failed to follow my directions the first time…

  22. edison101 says:

    My very first project was a winch made from an RC car. It was too weak to pull anything. I figured more voltage means more power so I decided to hook up a 10V drill battery in addition to 5 AA’s. I still keep the charred remains as a reminder to learn from your mistakes.

    1. edison101 says:

      My second attempt got me and my Destination Imagination team to state finals for the first time ever. Now were going to global finals this year. I guess I learned from my mistakes after all!

  23. baztastic says:

    Was backstage in a theatre, looking at the light fixtures all around the mirror, daydreaming wondering what the spring behind the contacts felt like (some bulbs were missing) – stuck my finger in, not realising that if I could see, the lights were on. Ouch. Luckily they were wired in parallel so I didn’t get all the current.
    Second one was when I used a low-voltage circuit tester to check if a lamp was working – pop, fizzle, smoke.
    Last one was a friend of mine – in the computer lab when he notices that on the backs of the power supply there’s a switch that says “110/240” (I’m in Ireland), so he switches it to 110 – that let all the magic smoke out. No idea why such a switch would be on the *outside* of a computer, but there you go.

  24. Chrome6 says:

    Built a Wheatstone bridge with one of the diodes reversed- boom!
    Hot pieces of diode down my shirt, small temporary burn. I always check polarity now, second nature.

  25. ethicalcannibal says:

    Mine was recent. I was carving some foam, and coating it in that black tool dip type stuff to create a prop that would look like a skull, but was safe to beat on someone’s head. I had the whole thing covered in tool dip, when it just popped out of my hand. Time froze as it descended down to bounce around on my apartments beige carpet. I don’t know about you, but explaining permanently bonded tool dip in the vague shape of a skull to your husband is very anxiety inducing. For his part, the hubby laughed, and said we were probably losing the deposit anyway. He’s very kind about what I’m tinkering with.

  26. ooph says:

    I work for a commercial diving company, and we use the same wire for underwater communications as we do for our cathodic protection sensor. tried like hell to get the comms working one morning, supervisor reminding me of the cost per minute from the delay, only to figure out 30 minutes later that the multimeter for the cathodic protection sensor is not a radio. hooked the divers speaker and mic assembly up to the other wire and nothing happened. used the multimeter to make sure i didn’t have a short or break, and plugged the divers comms back in. still nothing. after some choice words and door slamming, the newest guy on the job moved the divers comms to the other wire and we’re back in business. i had been plugging into the wrong wire not 1, but 3 times. in a row.

  27. John_NY says:

    I love these stories. Since you have Frits Lyneborg writing “Latest In Robotics”, perhaps he’s heard of the LMR page “Ignoble Idiocy” (at
    I’ve had bad luck taking shortcuts in my hobby electronics, not calculating current draw is a particularly bad habit. In one case, I upgraded a stepper motor to a (much) bigger one, and discovered why chopper drivers are nifty. In another, I discovered why I might upgrade a cheap 100 mA voltage regulator to a 1A voltage regulator. Blue smoke and singed fingers smell like learning.

  28. Karel K says:

    When tinkering and soldering at home I used to listen to a small battery operated radio. When the batteries die, what do you do? Hook it up to a lab power supply (5V).
    Then continue working, and after a while you forget about the wires that went to the radio. After completed my soldering, I wanted to test my work. Connect it to the (same) power supply (5V) and it didn’t work. I guessed cranking up the voltage a little might do the trick. Slowly increasing the voltage… BANG, some capacitor or transistor in the radio gave up, making me jump up a small distance in the air. The radio casing was damaged, and a foul smell filled the attic…

  29. Philip Millar says:

    I was mixing some two part polyurethane foam for a puppet head. Using an electric drill with a propeller style mixer required two hands, resulting in the container with the foam going into a spin. As a result I sprayed fast rising foam in a circle around the garage. Sadly I was working next to my landlord’s just sold BMW motorbike and the foam coated it neatly. I spent the rest of the day with a toothbrush and solvent cleaning the bike. Happy days.

  30. ka1axy says:

    My most epic fail occurred in 3d semester Physics class at college. We were measuring temperature with a diode and plotting the current vs temperature curve. The first step was to evacuate the air from the metal enclosure containing the diode. This enclosure was made of aluminium, about the size of a soda can, but 5mm thick walls and a bolted-on aluminium cap (which fortuitously faced down…more on this, later). The cylinder was on the end of a long metal tube with a valve, which was how we removed the air. The cylinder was duly evacuated and the valve closed (I thought). The next task was to lower the cylinder into a container of liquid nitrogen, cooling it down to get a low temperature point for the measurements. We proceeded to begin our measurements and removed the liquid nitrogen to allow the container to warm up. At some point in the process, I heard hissing. Deducing that I had not completely closed the valve, I called the TA over and asked what to do. He told me not to worry about it. Fearing that the air would ruin the experiment, I tightened down the valve.

    You can all probably guess what happened next. Some air had become trapped in the low temperature sealed metal cylinder. As we blissfully made our current measurements, the temperature of the cylinder slowly returned to room temperature, and the air inside expanded. The Ideal Gas Law, which I knew by heart, but had not sufficiently considered in the light of my recent actions, says that the pressure of a gas in a fixed volume will increase with increasing temperature. Well, it did, and it blew the bolted on cap right off the cylinder, leaving an impressive dent in the concrete floor, and thankfully not injuring anyone. The bolts (maybe 10 or 12 #6) were sheared off and the tube by which the cylinder was supported was bent impressively.

    I have never forgotten that experience. Probably the most memorable Physics lab I have ever done.

  31. ChrisW says:

    I was about 10 years old, and had recently learned how to hook up batteries in series (+ to -). A neighbor needed to jump start their car. Nobody was very sure of the procedure so I spoke up and insisted that they must be connected positive to negative. The resulting sparks and shower of melted lead and copper taught me about connecting batteries in parallel.

  32. ChrisW says:

    After repairing a 5 volt 300 amp power supply, I connected it to a dummy load and gave it a100% load test overnight. Confident in my repair, I reinstalled the power supply, but it didn’t work. I removed it and again it worked fine on the dummy load. Re-installed it and again it would not work- no output. Back out and on the dummy load it worked fine. Finally I realized that one of the 20 case screws was an eighth of an inch shorter than the others. It was supposed to go in the hole right next to the positive output bus bar. A longer screw was there and was shorting the positive output to ground. The dummy load didn’t care that the positive output was shorted to ground, but the equipment sure did.

  33. John Honniball says:

    Long ago, reel-to-reel tape recorders had valves (vacuum tubes) in them, and ran on 200V DC power supplied via a selenium bridge rectifier. Yes, selenium, not silicon diodes. I was working on a Gundig TK-25 of this type in the late 1970s. I very briefly shorted the DC output of the rectifier, but that was enough to make it fail and burn up. Burned-out selenium rectifiers smell like rotten fish, very strongly. Soon afterwards, the whole house smelled of rotten fish.

    A few months ago, we had a BBC Micro (6502-based home computer of 1982) running at the Bristol Hackspace. It started to let out smoke after a hour or so in operation. The capacitor across the mains input had failed due to age. This time, it smelled like a boy scout’s camp fire.

    Therefore, I conclude that one can diagnose many types of electronic failure by smell.

  34. Dr.Spaz says:

    In graduate school, I was given a partially functional cheap vacuum cleaner. Hoping to maximize it’s efficiency with it’s paltry motor, I took it completely apart and liberally lubricated everything with WD-40, including the interior of the electric motor. When I plugged it in and turned it on, there was a “WHUUUUUMP”, and a jet of flame shot out of the exhaust tube, across the room, and left a greasy black circle on the wall. Amazingly, the vacuum cleaner worked after that, although no better than before.

    1. Ross Hershberger says:

      YouTube is full of entertaining videos of people exploiting the combustibility of WD-40. It’s safe once it dries, but the propellant is pretty reactive.
      I fix a lot of vacuums in the course of a month. If the motor bearings aren’t shot, squirting them with a lube for mechanical locks that contains graphite often helps. Other lubes will attract dust and gum up the bearings.

  35. Daniel says:

    I have had several epic fails. One of my biggest occurred when I was very young. I tried to use a Nintendo virtual boy as a HMD for a wearable computer. It looked great for the time (early 90s) and I was very proud of it. My friend wanted to be the first to use it so I indulged him. The next thing I know my friend is telling me how hot his head is getting. I turned around after a few seconds and noticed that the whole unit was smoking. I yanked it off his head just as it burst into actual flames. A very close call. It taught me a few good lessons that have stuck with me. First is test early and test often. The second is to never give up. Failure is just another way to learn.

  36. EricG says:

    About a month ago I was working with an Arduino and an LED strip. I needed to put out more light, so I wired it up to my benchtop power supply, turned it on. The lights went crazy and the arduino got super hot. Couldn’t figure it out; everything was fine, so I went to bad.

    Next morning I got up and checked. Instead of plugging in +5V and ground, I plugged in +5V and -12V. $60 later, I was up and running again.

  37. Ross Hershberger says:

    This was an expensive one. I was testing a vintage high end loudspeaker. I could hear the tweeter playing but I wasn’t getting enough level on the microphone for a clean reading so I kept turning it up. then the sound stopped. Burned out tweeter. The meter registered low because its battery was flat, and I over-drove the speaker by mistake. Try finding a pair of NOS hand matched tweeters for a famous speaker that went out of production 30 years ago.

  38. phlamingo (@phlamingo) says:

    I can’t say this fail was epic, but the cover-up approached epicness. Epickicity? Epictude?

    Anyway, the scene is Boulder, 1985, in the CU Engineering building. The project is to build a functioning CPU from discrete TTL. After long weeks of wire-wrapping, micro-code, circuit debugging, and sleeping in the lab, we have it running. But not at the designed clock frequency of 10 MHz. Dial it down a few orders of magnitude. No, a few more. There, primary clock running an 100 Hz. With our ten-phase clock, that meant we were getting ten instructions per second.

    One of our team members had picked up a voice-synthesis chip, very primitive by today’s standards, but an amazing part for its time. We wired it up to our CPU and wrote a simple counting program to drive the chip.

    We had the only talking CPU in the class. Everyone was impressed.

    It wasn’t until we were disassembling the project that I found a quad XOR chip that was the open-collector version. Doh! We were lucky it worked at all, but if we had found that earlier, we could have wired in some pull-up resistors and run (hopefully) somewhat faster.

  39. Michael Durham says:

    The most memorable lesson I learned was figuring out how to check a breaker without consulting an electrical engineer first…. I also learned how many AMPS a Sears multimeter can safely measure without exploding and/or melting test probes! Thankfully the voltage side still worked, so I could eventually rule out the breaker as the culprit :p

  40. Bill Rowe says:

    Hardly epic but I remember looking idly at the tip of my laptop power cord and deciding to taste it like I would a 9v battery. Bad bad bad idea.

    1. D says:

      thats epic stupidity!! like plugging caps into the power outlet only worse..

  41. Steve says:

    Being tired of lying on cold concrete in my garage when ever I changed the oil in my cars, I set about building a vacuum extractor that would suck the oil out via a plastic tube fed down the engine’s dipstick tube. There are commercial versions, but they cost over $100 (and I’m a cheapskate). I modified a $60 air compressor I had laying around to draw a vacuum instead of pumping air (that worked beautifly). But my choice of a vessel to hold the vacuum and the removed oil was a 5 gallon plastic paint bucket, which, after turning on the pump, collapsed in on itself, ripping the bottom of the bucket way from the weaker sides and spilling 2 quarts of dirty engine oil everywhere.

  42. Rob says:

    I have two epic fails. No.1 Refilling an inkjet cartridge for my printer. Sounds simple. Pick up some syringes from the local chemist. I got some funny looks when I asked for the biggest they did.( I was wearing a leather jacket with denim cut-off covered in badges etc) Then some Quink ink. Sitting at the kitchen table I began syringing the ink into the cartridge. There seemed to be some pressure against the syringe so I pressed harder. The needle came away from the syringe and I sprayed ink all over the wall and ceiling of my parents kitchen. That made me popular.
    No. 2 My father was among other things a machinist, and I had seen and used his lathes under close supervision. When it came to metal work at school I new what I was doing. Step 1. Place metal in chuck. Step 2. Tighten chuck with chuck key. Step 3. Fire up lathe. Step 4. Find chuck key expelled from lathe chuck at high speed, bouncing off floor, walls, ceiling etc. I have never done that since……..

Comments are closed.

Discuss this article with the rest of the community on our Discord server!

In addition to being an online editor for MAKE Magazine, Michael Colombo works in fabrication, electronics, sound design, music production and performance (Yes. All that.) In the past he has also been a childrens' educator and entertainer, and holds a Masters degree from NYU's Interactive Telecommunications Program.

View more articles by Michael Colombo


Maker Faire Bay Area 2023 - Mare Island, CA

Escape to an island of imagination + innovation as Maker Faire Bay Area returns for its 15th iteration!

Buy Tickets today! SAVE 15% and lock-in your preferred date(s).