I had a cheapy Samsung curved 27inch 1080p monitor that use to live in a tax office. It didn't work so it was given to me to see if I could fix it. I noticed I couldn't plug in an HDMI cable without the back of the monitor popping the cable out of it's socket. The only way to get the HDMI cable to plug in all the way was to angle the monitor down at the table. If I try to face it up, the bezel tents out and pushes the cable out of socket.
I put the monitor in a corner and left it for around a year. A year later I get another similar monitor with a broken OSD nipple like the one in this article. I noticed that it had no problem with cables being pushed out of socket. I looked at the back of the monitor and saw that there's a little screw just above the inputs on the back of the new monitor.
So I snag the screw from the new, broken monitor and try to screw it in to the old 'working' monitor in hopes that it would fix the shell, and it won't screw in. WTF right? Same monitor, same model.. No screw hole but there was a place for a screw, just no hole.
So I pull the bezel up enough to notice that there actually IS a threaded screw hole behind the shell, there's just no hole in the back for the screw to go in. So I break out a drill, make my own hole and the screw screws right in. Perfectly fixed.
For something like this, I feel like the work to investigate the guts of the switch mechanism was overkill. I'd rather desolder it wholly, and try to find some component to replace it with. I have little faith in non-break-itude of teeny things when handled slightly incorrectly.
That being said, I once replaced the thumbsticks on an X360 controller due to stick drift. It was very fun to destructively open up the old mechanisms and see how they worked.
I also operate my sit/stand desk with a pair of tweezers, by shorting some pins on a breakout board.
Hey now, the electronics of the sit/stand controller are safely housed in a little cardboard box (that the BoB came in). The breakout board sticking out is labeled "FPC-10P 0.5MM", and I only need 3 of the 10 pins. The others are unused by the controller.
> And here I thought I was nuts using a bench power supply clamped onto some loose jumper wires to charge my wireless keyboard/trackpad.
Also, at your point, couldn't one bodge something together with a little charging circuit or "variety pack of M/F connectors on wires" from Sparkfun? If you're willing to cut away some material, you could build your own special charging port, so that the loose wires end up outside of the device.
Agreed. The tolerances inside a 5-way stick switch like that are so tight, just manually disassembling and reassembling a brand new one is likely to make it flakey. It would be much better to find the right replacement component and desolder the dead one and replace it. The challenge would likely be identifying the part and getting a single replacement part. If it happens to be something available from Digikey or Mouser that's great, but there's always a chance that it's a slightly tweaked version of a "standard" part or even something that was made only for LG. I found some parts from E-Switch and Alps on Mouser quickly that look visually similar, but I would need access to the original PCBA to evaluate dimensions etc...
I have the same control doodad on my LG monitor. Not only does it apparently fail quite often, it sucks entirely to begin with and is an incredible regression in ergonomics and human interface. So stupid.
It beats figuring out which of the 5 typical buttons are up/down/left/right, in my opinion. It's only the build quality that seems to be the issue, not the concept
In truth it actually does have an ergonomics issue. Since the button position is orthogonal to the screen and mounted upside down, the left and right directions are evident but up and down directions are ambiguous.
It turns out you 'pull' the thing toward you for 'down' and 'push' it away from you for 'up'. Did you guess right?
> It turns out you 'pull' the thing toward you for 'down' and 'push' it away from you for 'up'. Did you guess right?
The only thing dumber than that is that DJI remotes use the same braindead scheme to control camera tilt: pushing the wheel moves the camera up and pulling on it - moves it down. Anyone who piloted anything would, of course, agree that it's insane (you push the stick to go down and pull on it to go up). I had to open mine up and swap the wires because DJI in their infinite wisdom didn't make that configurable in their app.
Oh my god, I've had this exact problem. So many "cinematic" drone videos ruined by me tilting the wrong way...
I spent ages searching for that option in the app, but never thought of physically swapping the wires in the controller. You've just given me a new project.
It's so stupid that DJI doesn't make that configurable.
Anecdotally, I'm using a 27" 4K LG monitor I got in about mid 2017, so going on eight years now, and I've never had an issue with the nub, except that when the monitor turns on, it will automatically pull up the nub menu after a few seconds, so if you go to change the input at the wrong time, you'll accidentally shut it off.
Not saying you're wrong about the ergonomics, but I personally like it.
>With the rear housing off I had access to the power button housing—it’s the white plastic thing at the bottom center of the panel. While the logic board and power components are covered by a metal box, I must emphasize that even while unplugged it’s still possible to encounter spicy levels of current lurking inside the capacitors in the power supply. Be careful if you try this at home.
This was true when CRTs monitors were around, and also if you're tearing into an old plasma screen. However, the power supply inside a new LED or OLED flat screen monitor will drop straight to zero after just a few seconds.
You don't see them much any more, but cold cathode backlight lcd panels also have a high voltage section to drive the ccfl's
And thank goodness, I like to tear things apart when they break, I am not very good at fixing them but finding the root cause helps me find closure. anyway, nearly every single time an lcd would die in that time period(beginning of lcd flat panels to 2015 or so) I would open it up to find the high voltage control chip smoked.
just to be clear, the stray shock potential of a CRT had to do with the flyback circuit for tube itself, not the power supply near the power button/power conditioning.
The only desoldering tool that I use because I gave up on desoldering wick at about age 12 when it just didn't seem to work. A decade or two later, I learned that you have to put the wick between the soldering iron and the tin... pretty obvious, right?!
I remember when I was young, fixing the power button for my Commodore monitor with a switch salvaged from a hairdryer. It just hung off the monitor, dangling with several inches of extra wire.
Are there any examples of modern monitors that actually have understandable and reliable buttons for power and configuration? Looking at my Dell monitor, I have six tiny, identical, unlabled buttons under the bezel, and a slightly bigger, slightly different-looking power button. All made of cheesy, ready-to-break plastic.
For the unlabeled control buttons, you have to press one and then a little on-screen GUI pops up with icons that barely tell you what they do. When you choose an option on the on-screen GUI, the function of each button changes. There is about 1 second of latency between when you push any of these buttons and when it does something. Absolutely awful.
Pretty much all high end monitors have a little joystick under/behind them, have decent UI latency, and come with vendor software for changing settings without physically touching the monitor. This is done over I2C so it should be pretty easy to reverse engineer. Try searching "monitor ddc control", it might work out of the box for your monitor
Surprising to find a post about the same series of LG monitor as the one I'm reading it on. The power button on mine hasn't failed... yet?
Very brave to go disassembling the button component itself, and it paid off. Personally I wouldn't spend that time, but the worst case is ordering a new button part either way, so I might try it in future.
I use https://github.com/kfix/ddcctl to change inputs on my external Acer EB275K (bmiiiprx) from my Mac. Brightness can also be adjusted.
Simplified the usage by creating shortcuts in Keyboard Maestro that runs short ddctl commands.
I have the same stick button on my LG monitor, but it’s been useless for a while now. I’ve been sticking to ddcutil for now, and I’ve lost those little flower pad things already. But the build quality of that button is quite bad.
Given the opportunity I usually install a heavy toggle switch when a cheap plastic PCB switch fails. I don't think I could pull that off with a 5-option joystick like that.
Personally, I woulden't even both for internal part requests like that, especially when devices are often stamped "no user servicable parts inside".
If you have the know how to rework PCB's like this person (and all the equipment already on hand), I would estimate that it's the same amount of effort to just fix it like this yourself, than to seek out the manufacturer, communicate and identify the internal part, and arrange payment.
I had a cheapy Samsung curved 27inch 1080p monitor that use to live in a tax office. It didn't work so it was given to me to see if I could fix it. I noticed I couldn't plug in an HDMI cable without the back of the monitor popping the cable out of it's socket. The only way to get the HDMI cable to plug in all the way was to angle the monitor down at the table. If I try to face it up, the bezel tents out and pushes the cable out of socket.
I put the monitor in a corner and left it for around a year. A year later I get another similar monitor with a broken OSD nipple like the one in this article. I noticed that it had no problem with cables being pushed out of socket. I looked at the back of the monitor and saw that there's a little screw just above the inputs on the back of the new monitor.
So I snag the screw from the new, broken monitor and try to screw it in to the old 'working' monitor in hopes that it would fix the shell, and it won't screw in. WTF right? Same monitor, same model.. No screw hole but there was a place for a screw, just no hole. So I pull the bezel up enough to notice that there actually IS a threaded screw hole behind the shell, there's just no hole in the back for the screw to go in. So I break out a drill, make my own hole and the screw screws right in. Perfectly fixed.
First time I ever fixed a monitor with a drill.
For something like this, I feel like the work to investigate the guts of the switch mechanism was overkill. I'd rather desolder it wholly, and try to find some component to replace it with. I have little faith in non-break-itude of teeny things when handled slightly incorrectly.
That being said, I once replaced the thumbsticks on an X360 controller due to stick drift. It was very fun to destructively open up the old mechanisms and see how they worked.
I also operate my sit/stand desk with a pair of tweezers, by shorting some pins on a breakout board.
>I also operate my sit/stand desk with a pair of tweezers, by shorting some pins on a breakout board.
And here I thought I was nuts using a bench power supply clamped onto some loose jumper wires to charge my wireless keyboard/trackpad.
Hey now, the electronics of the sit/stand controller are safely housed in a little cardboard box (that the BoB came in). The breakout board sticking out is labeled "FPC-10P 0.5MM", and I only need 3 of the 10 pins. The others are unused by the controller.
> And here I thought I was nuts using a bench power supply clamped onto some loose jumper wires to charge my wireless keyboard/trackpad.
Also, at your point, couldn't one bodge something together with a little charging circuit or "variety pack of M/F connectors on wires" from Sparkfun? If you're willing to cut away some material, you could build your own special charging port, so that the loose wires end up outside of the device.
Agreed. The tolerances inside a 5-way stick switch like that are so tight, just manually disassembling and reassembling a brand new one is likely to make it flakey. It would be much better to find the right replacement component and desolder the dead one and replace it. The challenge would likely be identifying the part and getting a single replacement part. If it happens to be something available from Digikey or Mouser that's great, but there's always a chance that it's a slightly tweaked version of a "standard" part or even something that was made only for LG. I found some parts from E-Switch and Alps on Mouser quickly that look visually similar, but I would need access to the original PCBA to evaluate dimensions etc...
https://www.mouser.com/ProductDetail/E-Switch/JS1400BFQ?qs=E...
https://www.mouser.com/ProductDetail/Alps-Alpine/SKRHACE010?...
I have the same control doodad on my LG monitor. Not only does it apparently fail quite often, it sucks entirely to begin with and is an incredible regression in ergonomics and human interface. So stupid.
It beats figuring out which of the 5 typical buttons are up/down/left/right, in my opinion. It's only the build quality that seems to be the issue, not the concept
In truth it actually does have an ergonomics issue. Since the button position is orthogonal to the screen and mounted upside down, the left and right directions are evident but up and down directions are ambiguous.
It turns out you 'pull' the thing toward you for 'down' and 'push' it away from you for 'up'. Did you guess right?
> It turns out you 'pull' the thing toward you for 'down' and 'push' it away from you for 'up'. Did you guess right?
The only thing dumber than that is that DJI remotes use the same braindead scheme to control camera tilt: pushing the wheel moves the camera up and pulling on it - moves it down. Anyone who piloted anything would, of course, agree that it's insane (you push the stick to go down and pull on it to go up). I had to open mine up and swap the wires because DJI in their infinite wisdom didn't make that configurable in their app.
Oh my god, I've had this exact problem. So many "cinematic" drone videos ruined by me tilting the wrong way...
I spent ages searching for that option in the app, but never thought of physically swapping the wires in the controller. You've just given me a new project.
It's so stupid that DJI doesn't make that configurable.
Anecdotally, I'm using a 27" 4K LG monitor I got in about mid 2017, so going on eight years now, and I've never had an issue with the nub, except that when the monitor turns on, it will automatically pull up the nub menu after a few seconds, so if you go to change the input at the wrong time, you'll accidentally shut it off.
Not saying you're wrong about the ergonomics, but I personally like it.
>With the rear housing off I had access to the power button housing—it’s the white plastic thing at the bottom center of the panel. While the logic board and power components are covered by a metal box, I must emphasize that even while unplugged it’s still possible to encounter spicy levels of current lurking inside the capacitors in the power supply. Be careful if you try this at home.
This was true when CRTs monitors were around, and also if you're tearing into an old plasma screen. However, the power supply inside a new LED or OLED flat screen monitor will drop straight to zero after just a few seconds.
You don't see them much any more, but cold cathode backlight lcd panels also have a high voltage section to drive the ccfl's
And thank goodness, I like to tear things apart when they break, I am not very good at fixing them but finding the root cause helps me find closure. anyway, nearly every single time an lcd would die in that time period(beginning of lcd flat panels to 2015 or so) I would open it up to find the high voltage control chip smoked.
just to be clear, the stray shock potential of a CRT had to do with the flyback circuit for tube itself, not the power supply near the power button/power conditioning.
Flybacks are deadly when live, however the tube itself was also a massive cap.
Sources for those that need it:
https://repair.wiki/w/CRT_General_Repairs
[...] and the golden tube that goes pop (I can’t remember the name) couldn’t vacuum up the solder [...]
It's called a solder sucker amazingly enough (perhaps desoldering pump is the more formal name). Very handy when they work.
The only desoldering tool that I use because I gave up on desoldering wick at about age 12 when it just didn't seem to work. A decade or two later, I learned that you have to put the wick between the soldering iron and the tin... pretty obvious, right?!
Also if you use flux, or at least flux-impregnated wick, it works way better.
I remember when I was young, fixing the power button for my Commodore monitor with a switch salvaged from a hairdryer. It just hung off the monitor, dangling with several inches of extra wire.
I hope you loosely wrapped one of the connections to the switch with a slightly too small piece of electrical tape.
Pro Hack
Are there any examples of modern monitors that actually have understandable and reliable buttons for power and configuration? Looking at my Dell monitor, I have six tiny, identical, unlabled buttons under the bezel, and a slightly bigger, slightly different-looking power button. All made of cheesy, ready-to-break plastic.
For the unlabeled control buttons, you have to press one and then a little on-screen GUI pops up with icons that barely tell you what they do. When you choose an option on the on-screen GUI, the function of each button changes. There is about 1 second of latency between when you push any of these buttons and when it does something. Absolutely awful.
Pretty much all high end monitors have a little joystick under/behind them, have decent UI latency, and come with vendor software for changing settings without physically touching the monitor. This is done over I2C so it should be pretty easy to reverse engineer. Try searching "monitor ddc control", it might work out of the box for your monitor
Your description of your dell matches mine and triggered me. I can only conclude the designers are sadists.
Surprising to find a post about the same series of LG monitor as the one I'm reading it on. The power button on mine hasn't failed... yet?
Very brave to go disassembling the button component itself, and it paid off. Personally I wouldn't spend that time, but the worst case is ordering a new button part either way, so I might try it in future.
I use https://github.com/kfix/ddcctl to change inputs on my external Acer EB275K (bmiiiprx) from my Mac. Brightness can also be adjusted. Simplified the usage by creating shortcuts in Keyboard Maestro that runs short ddctl commands.
I have the same stick button on my LG monitor, but it’s been useless for a while now. I’ve been sticking to ddcutil for now, and I’ve lost those little flower pad things already. But the build quality of that button is quite bad.
Given the opportunity I usually install a heavy toggle switch when a cheap plastic PCB switch fails. I don't think I could pull that off with a 5-option joystick like that.
Impressive that it was fixable in the end! Quite brave to keep taking it apart, although I suppose the monitor couldn't get any worse at that point.
Is a new button assembly from LG not available? The JST connector suggests it was designed to be easily replaced.
I found a component that looks like it would be able to replace the one pictured, but can't be sure without measurements: https://www.mouser.com/ProductDetail/SparkFun/COM-10063?qs=W...
Personally, I woulden't even both for internal part requests like that, especially when devices are often stamped "no user servicable parts inside".
If you have the know how to rework PCB's like this person (and all the equipment already on hand), I would estimate that it's the same amount of effort to just fix it like this yourself, than to seek out the manufacturer, communicate and identify the internal part, and arrange payment.
Anyone had to fix the power button on a Dell U3011T? Advice / tips welcome.
Love this guy's Schiit stack!
[flagged]