SHOCKING Monitor Issue – How to handle defective products
This monitor has a shocking issue, no, no literally it will SHOCK you. This is the video I sent to their PR rep.. Yeah, you are seeing that right. Current arcing from the monitor’s metal body to the display cable. I caught this issue before many of these monitors were shipped out, I was one of the first people to have one in-hand, and I’m really glad I found this before anyone got hurt.
In this video I want to walk you through how I found the issue, then diagnosed the cause, fought the supplier, how the seller reacted, and explain how companies should handle defective products.
The issues started the second I got it out of the box. I plugged it in to start testing, specifically to power first, then plugged in a display cable specifically displayport. I turned on the PC, expecting the monitor to come on but it didn’t. I looked around, found the power button and pressed it. Nada. I double checked the power was on at the plug, that the plug was connected properly but still nothing. I tried holding the power button down too, again nothing. I left it alone and messaged the PR rep saying the monitor wouldn’t turn on and carried on with my day.
The reply I got started the alarm bells in my head. He said “As odd as it sounds, this happens when it’s too cold”. Now I’d understand if I had left it outside in the arctic or kept it in my freezer, but it was sitting in my living room for a few days before I got it out and started testing. My house is pretty consistently 25°c which I wouldn’t consider ‘too cold’. Either way, I came back to find the monitor’s power LED was now lit. So, I went to plug the display cable in. Then I saw it. An electrical arc from the DisplayPort plug housing to the corner of the cable. I filmed a little clip of it which I sent to their PR rep. He asked if I was using the included DisplayPort cable in the box, I wasn’t, so I used an HDMI cable I had spare as they don’t include one in the box, this time dragging the cable across just the metal casing. It sparked the whole way across.
To say I was shocked would be an understatement. A monitor with 230V going into it was arcing to display cables. What if that was a finger searching for the inputs blindly and you happened to be grounded?
Their PR rep asked me to check another monitor to make sure it wasn’t some issue with my system or anything, so I grabbed the ultra-cheap 1440p 165Hz MBEST monitor I use for testing and dragged the HDMI cable over it’s casing. No problems. Being overly inquisitive, I cracked out my multimeter and checked the earthing. As Big Clive would tell you, products, especially ones with mains current, should have any metal casings earthed through their plug. So, I checked continuity, basically can current pass from one point to another, and surprise surprise the metal casing wasn’t connected to the earth pin in the IEC C13 (kettle lead) plug. Their PR asked me to check the other monitor for that too, which despite having an external power supply, still had the exposed metal casing grounded through the central pin.
I asked if he was happy for me to disassemble the monitor, which he agreed, so I got to work. There are a couple of screws under the inputs label, then a load of clips all the way around the backplate that you pop open, then remove the flat display cable and the two pin power for the panel, all the tape, then you’ve got the back free from the panel.
The very first thing I noticed was that the power supply was completely disconnected from the scaler board, save for the 12V and ground wires. No physical connection between the earth pin and the scaler board or the metal casing. Nothing. I removed the PSU board, popped off the backplate, shorted the capacitors with an insulated screwdriver for good measure, then started looking at the layout. It’s not a horrendous board, it has a separation zone between the ‘Hot’ and ‘Cold’ sides, even if one of the traces does run into the zone, and it’s a proper transformer with opto-isolator unit so that’s fine. What’s not is the distinct lack of a path from the 12V ground to the earth pin in the plug. The multimeter confirms that, no continuity between the low side ground and the high side earth. That’s even further confirmed when you check continuity between the ground pin on the PSU and the metal casing. No continuity. Not connected. Not earthed.
So, there’s the problem. Or, at least, there’s the reason it got so far as to shocking things. The reason the casing has power flowing to it? Well, I also checked the resistance between the 12V output of the PSU and the casing. It was a couple of kilo-ohms. To explain just how bad that is, your skin measures a couple of MEGA-Ohms, that’s 10^3 more resistance. Hell, the live and neutral pins on the plug of the PSU read a few mega-ohms of resistance so it doesn’t short circuit. Taking a look at the scaler board, I’m not surprised this is leaking current. Look at the solder quality. Uneven, some barely covered some with literal mounds. These pads are meant to be the ground pads for the board to the casing and they only have the slightest coverage on just a quarter of the hole. How is this meant to create a reliable connection? Clearly it must because it’s leaking current.
So, I photographed everything and wrote a report. I sent it over, receiving many thanks for my work and went to bed. The next day I wake up to find a message from their PR rep who’s spoken to their supplier, the people who actually made these monitors, who proclaimed “the monitor is grounded through the power plug”. No, it isn’t. I tested that. I looked. There is no way beyond quantum tunneling that it could be grounded. They sent a video of them testing a sample they had and it did have continuity between the casing and the ground pin. I shot back this picture showing definitively that it is, in fact, not. They apparently took issue with my multimeter settings, deciding that the logo for the continuity mode, specifically the logo of a diode, a component that only lets current flow in one direction (until you reach its breakdown voltage), so I reply with a picture of the same test but using the resistance mode. Again, no connection.
This one got me though. “Low voltage ground can not [be] connected to the high voltage earth.” What?? So, if the high voltage earth can’t be connected to the low voltage side, how did you manage to find continuity between the earth pin and the casing? They then said “the metal part of the I/O casing is grounded to the earth wire on the plug” WHICH IS IT BOYS? It can’t be connected, or it definitely is? Can’t be both.
Later that day, I get a barrage of messages. They say “the black wire is grounded to the scaler board, that is why we did not add an earth wire to connect [the] scaler board and [the] power supply board”. This would be fine, IF THE 12V GROUND WAS COMMONED TO THE EARTH PIN OF THE PLUG. They then sent me my own picture of me showing continuity between the PSU and the scaler – but they picked the wrong one. This isn’t the dunk they thought it was. That picture, ladies and gentleman, shows continuity between the 12V pin and the grounding tab meant to give a secondary ground to the panel. Yeah.
They then lied again, saying “black wire is connected to earth wire on plug”. BUT BOYS, YOU TOLD ME EARLIER IT CANNOT BE CONNECTED! WHICH IS IT?
Now, they left it at that and again I woke up the next day to a lot of unread messages. The first one reads, “Good news! Our OEM has investigated and it is their mistake”. “The OEM also apologises to you”. He sent over an image of what the ‘fix’ is, surprise surprise it’s an earth wire from the scaler to… the ground pin of the plug. Hallelujah.
So that’s the issue, and the fight with the supplier, but how did the UK company react? His very first message after I told him it was sparking was “Thanks for this, I’ll check to see if this is happening on any others and check with the OEM on this issue.” He acknowledged the problem and began an investigation there and then. When I followed up with the second video of it arcing across the whole I/O plate, he said he will check other samples, and that he’ll have a “stern word to our OEM”. He asked me to hold off on publishing a review as they will likely recall the units and issue a fix. To make it clear, this is good. Me publishing a review on a product with an active safety recall would be poor form at best, so this request is reasonable.
He apologised for the issue, saying it shouldn’t have happened in the first place and that they will be checking retail samples they have. On the same day I reported the issue, he said if he finds any models he tests have the same issue he will recall all shipped units and halt sales until they are repaired. They set the monitor to be out of stock the next morning, then when the OEM admitted fault they requested the monitors that customers had bought be recalled, and offered refunds, and they had all 500 units built be repaired by a professional firm in the UK.
I have one of those repaired units here, which I can confirm is now grounded as it should be. Here’s a look at what’s changed inside.As you can see, what’s been added is a grounding wire from the scaler to the earth pin of the PSU board. That’s enough to ensure no current will leak from the scaler to the casing, although technically it doesn’t fix the cause of the current leakage. With that said, I verified continuity between the earth pin and the casing so it’s at least now safe.
This is how you handle a defective product. In fact, here is the TechteamGB guide for companies.
1 – acknowledge the user’s problem, begin an investigation to check if the problem exists on other units, or if it is localised to that singular unit.
2 – If you find the fault in other units, halt sales, especially if it is a safety concern.
3 – Establish what needs to be done to rectify the issue, and execute the fix. That may involve recalling already sold products, and where customers are unhappy to have been sold a defective product you should be issuing full refunds.
That’s it. Pretty simple, right?
This specific issue should never have made it to market. It shouldn’t be up to an inquisitive nerd like me to find, then diagnose, then fight the supplier’s blatant lies. Although I will note that I had their PR rep test the monitor he had with him and it was grounded, so this sort of issue comes from the supplier trying to save a few pennies on each unit, rather than an all out failure to check at any point. Either way, I’m really glad it was me that found it in a somewhat controlled manner, and that I know enough about electronics to fight to get this resolved, rather than an end user finding out with a jolt of DC.
