[Johnny] had a monitor that he was particularly fond of. The whole monitor appeared dead, and he decided to open it up and find out what could be wrong. He wound up fixing it — sort of — using a hairdryer. While we think his explanation of the problem is unlikely, we hate to armchair quarterback, and we applaud that he opened it up and got it working.
When something is dead, it is always a good idea to check the power and power supply, but that didn’t pan out in this case. In fact, the power supply board inside had what looked like reasonable voltage values throughout. The problem had to be something more subtle.
There are a pair of low dropout regulators on the interface board, and [Johnny] suspected that one of them had gone bad. We didn’t fully understand his paragraph about LDOs getting old and not being able to be re-enabled with the quiescent current if the component isn’t hot enough. On the other hand, heating the components up with a hairdryer seemed to fix them. We suspect it may just be a failing solder joint, but regardless of the explanation, it works.
Apparently, the problem occasionally occurs again, and heating the spot on the case next to the regulator fixes it. Without seeing it firsthand, it is hard to be sure what’s going on, but we are still going to bet on a finicky solder joint. If you want to argue for or against either hypothesis, we are sure there will be plenty of discussion in the comments.
There was a time when it was easy to work on consumer gear and it was relatively easy to get the schematics. These days everything is tiny, hard to replace if you can find the replacements, and service documentation is rare. But you can still get lucky sometimes. You just have to try.
We’ve noticed a trend of hair care products sneaking into repairs. Sometimes repairs can seem mysterious, like an infamous case of less being more.
Not the same issue, but I once fixed a dead monitor by putting the logic plate on a toast oven. I figured that the problem were some brittle solder joints somewhere on the board, so I cranked up the temperature and I performed a crude reflow repair.
The monitor was already doomed at that point, so it was worth a try.
After some burnt plastic smell, I mounted everything and the monitor sprung to life! Yay!
I had an HP printer that had imaging board problems that were solved by putting it in a toaster oven at 350 degrees F for about 30 seconds. I had to do it every 14 – 18 months but it worked every time. Apparently this was a solder paste formulation problem with all of that particular model.
The larger problem was explaining it all to my wife as I sat there intently watching the board in a toaster oven with a timer and temperature monitor going….
Done that for a paying customer, even. Nothing quite establishes you as a tech support wizard by regaling your customers with the tale of how you fixed their printer by baking it in the oven.
Remember the whole RROD problem? There were plenty of guides on ‘toaster oven reflows’ to reflow the fractured solder balls. We got in one unit that was a burnt melted mess asking if we could repair it: turns out that disassembling the 360 and removing the board for reflow was too complex, so the customer figured that just sticking the whole assembled unit in the microwave would work just as well. It did not.
What’s the #1 killer of displays? Probably heat. Put some ice on it.
Addendum: Drown everything in flux Rossmann-style and reflow all of the joints.
Addendum Addendum: Reflow all the joints using the hair dryer
Just install a hairdryer in your monitor, bro.
We’re all nicks down here
It’s Knicks all the way down.
Seal the vents and continously circulate liquid flux through the enclosure duing operation. Use an air-flux heat exchanger for cooling.
Submerge the whole monitor in flux and put it in the fireplace. That should do the trick.
I don’t know about heating an IC to make it work but in the 70s I used a large complicated laboratory instrument that would occasionally start acting erratically. Found a board with a chip that got too hot to touch. Hit it with can of compressed air and cooled it down until the machine worked again. I think this was a genuine hack.
My last ship had an Wärtsilä RT-flex96C and one turbo charger sometimes needed a gentle nudge with a sledgehammer…
That is a tool I call “persuasion”. If all fails, try some persuasion. It’s a very convincing argument.
“hand me that persuader, would you?”
Reality mallet.
When your reality doesn’t match the doc, hit it with the reality mallet until it’s ‘close enough’.
That can have disastrous consequences:
“As crew members frantically pounded the doors with hammers, water flooded into the cargo hold.”
-> https://www.history.com/this-day-in-history/sloppy-safety-procedures-lead-to-ferry-sinking
Precision adjustment 😁
Percussive maintenance.
Compared to the size of that engine, a sledgehammer is a precision instrument, though.
The real challenge is that when the turbo charger is replaced that there are no complains from the manufacturer about a “modified” housing.
It needs some sort of expertise.
We need a total for the amount of damage caused by people wearing VR goggles, between the broken bones and the smashed displays and the holes in the drywall it’s probably pretty significant.
That would be caused by incorrectly setting the play area boundary.
Don’t forget the value of things covered in barf, bet that’s the #1 most expensive category.
Another Hair Dryer repair. . .recently ‘fixed’ a Kenmore refrigerator with a flickering interior light. A google search revealed it as a common issue with several remedies: replacing overheated smt resistors and heating the led board with a hair dryer.
Of course I laughed off the hair dryer and went looking at the SMT resistors, but forget my Aven magnifiers. So what the heck, just for fun, I’ll heat up the board.
Well the flicker was no more!
Over time, moisture creep in the conformally coated PCB probably affected circuit operation and warming up the board dried it out. its been several weeks and no more flicker saving $60 until the flicker returns
Some DC/DC converters and LDOs will not stabilize if there is enough of a load. I can kind of see what what he is saying, but my guess, with the fact its designed to be energy star certified, that isn’t the case.
Won’t stabilize if there _isn’t_ enough of a load. (Which I suspect is what you meant to type.)
My plasma TV seems to have problems firing up the display when it’s cold, it manifests as some pixels staying dark until the panel has been on a while and they slowly turn on.
Delaminated contacts on a flexible ribbon cable or cold joints on FFC pads.
Soldering fault on SO-8 package seems pretty unlikely, and the post says the solder joints appear ok. The explanation in the post is not very complete and it is hard to know the full truth without through-out testing, but it is true that the transconductance of bipolar transistors increases with temperature.
It seems weird how often Hackaday articles second-guess the original writers with poorly justified explanations.
Heating with a hair dryer won’t just heat up that package, it’ll heat up… just about everything else on the board. It’s far more likely that something else downstream has a flaky joint and heating that device pushed it into contact. How the problem reoccurs (after it’s been off for a while) is far more consistent with mechanical than electrical.
^This
My bet is on those capacitors nearby. Failing circuits that “magically” start to work when heated up are a common occurrence – especially when there is an obvious switching regulator involved …
How hot is that hair dryer?
No chance it melted the solder. Thermal expansion or humidity.
Protip: To do a home reflow you need to know the melt temp of the solder, preheat in oven, final heat soak with a temperature controlled hot air workstation, then back to oven for slow cooldown. Has worked for me.
Your typical hair dryer can put out 1500-1800 watts, which is enough to do a lot of damage. What prevents the device from turning into a fire starter is the amount of air flowing through it, dissipating the heat. If you do something to prevent the heat from being dissipated so quickly, such as restricting the intake or restricting where the out-flowing air can go, you can build up lots of heat quickly. Restricting the air flow like that should eventually cause the hair dryer’s thermal protection to trigger.
In the video, the hair dryer is shown pointing perpendicular to the board about 1″ away (with a concentrator attached to the hair dryer), which is enough to restrict the airflow and get some really hot temperatures. But on the other hand, he only did this for about 15 seconds, which is probably too little time to melt unleaded solder.
He says the second time he heated the board through the case, which has a metal shield, so it’s virtually impossible to reach solder melting temperature in that situation.
You couldn’t heat a board to solder melt in 15 seconds with a hot air rework tool. Which are designed to melt solder.
1800 watts is all you can get out of a 15A 120V circuit. American hair dryers are mostly 1kW.
On your first point, I agree (assuming the board starts out cool).
On your last point: If you search Amazon for hair dryers, most claim 1800W-1875W. A couple claim 2200W. The lowest advertised wattage I saw on the first results page was 1400W.
Personally, when I use a hair dryer, I prefer the mid-heat, high-fan setting.
One one hand you have claims of some Amazon sellers.
On the other the laws of physics.
I suppose those hair dryers could plug into an electric range outlet.
1800W is right on the edge of tripping a 15A breaker, by itself. (15*120)
Be skeptical. Look at the size of the cords. Bet the euro version has 1800 watts.
I’ve seen a product claiming to contain (140000 mg) 5 ounces of active ingredient on Amazon. Gross shipping weight was 2ozs, including glass bottle. Must ship in a Tardis.
A 15A breaker takes about 1 hour before tripping at 15A; look up the trip curve for a breaker some time. They really do pull 1800W.
Roger:
15A for the whole circuit. When I use my welder I have to make sure nothing else on the circuit is on, but a hair dryer?
You believe the 2200 Watt 120V device claim?
What’s the gauge of that hair dryer cord?
I repeat my ‘shenanigans’ call.
Besides, I’ve used 1kw hair dryers. They provide plenty of heat. There is no reason to go hotter, just marketing. A 2200W hair dryer could set hair on fire.
Looking at YouTube videos of people who’ve actually measured the power consumption of hair dryers, it seems that the typical hair dryer will use about 1500 watts on the highest setting, even though it may be labeled as a “1875” hair dryer. One could probably assume a similar 80% de-rating for the 2200W hair dryers.
Most US household circuits are 20A, not 15 (15A is more common in mass-produced apartments, housing projects, etc). At 20A, you could theoretically pull 2400 watts.
“No chance it melted the solder. Thermal expansion”
Thermal expansion is absolutely great at restoring contact on a cold joint. Huge hysteresis, too. There was a project I worked on that had devices deployed in the field (so they were inaccessible) that a few of them had poor QA and an individual bit on an ADC went open. Plotted it versus temperature and got a beautiful hysteresis curve, with like 10-15 degrees separation between the going-closed and going-open sides.
I once fixed some new 1M SIMMs by heating up the RAM pins so the solder would actually connect to the pins.
I had a tv converter that stopped working properly, a hair dryer would bring it back to watch tv. But someone gave me a new one before I tracked down the problem. Itwas a way to cut down on tv watching, an extra step to turning it on.
Remember, hair dryers are useful tools, but not so good for hair.
Interesting flip on the traditional approach of finding a faulty component by heating it — or by letting it warm until it failed, then hitting it with circuit chiller (spraycan of freedom or something similar) to see if that cured the problem.
Just checked, and freeze spray us still available. Hopefully now in less polluting forms.
My story doesn’t involve heat, but probably something with transistor electron migration, quantum or whatever:
A friend of mine owns a Twin Famicom (Famicom+Disk System in one case) that also outputs RGB. The red channel doesn’t work. Since the machine is rare and the exhibition it’s shown at is already running, I couldn’t open it. I put my Testofon to the output and its sound went wub wub wub wub squeeeeeiiii (the latter indicating that the output transistor started working again).
The “fix” lasted two or three weeks and could be repeated. Man I love the Testofon, just when you’d think you found all its uses…
I once had a faulty CRT TV and found out that it did a beam current measurement at startup and that failed. After fooling the circuit with a resistor to pass some more current the TV started working again, but that resistor also distorted the image. So I added a microswitch and a piece of string to it, and you had to pull the string after turning it on. That was enough to add a few more usable years to the thing.
But for this monitor? Nice if you can find the culprit with a hairdryer, but I won’t call it a “repair”. If the chip is broken, just replace it.
Sometimes Hot Air works and sometimes it is just Hot Air.
I would recommend checking your Power Cable first. That is the Power Cable of your Hair Dryer First. Remember the Rule is always Hardware first and then Software. A two speed budget model works well if you have Power at the socket.
What’s armchair quarterback? That means nothing to me in europe.
Someone who comments from the comfort of an armchair (usually with no experience). Can be anything from an armchair expert to armchair quarterback.
Refers to a person watching a (US) football game from the comfort of their living room and thinking that they could do a better job (or decide a better play strategy) than the quarterback (key player) on TV who is actually doing it.
More or less means the same as “back-seat driver”.