1

I have a Dell LCD (LED backlight) without speakers which produces an audible "hiss" when connected the HDMI input. DisplayPort much less. I have recorded the following spectra:

monitor off (this is about the minimum the phone can record, so it's not very stable) monitor off

monitor on HDMI UHD @30 Hz monitor on hdmi

The noise starts at about 9 kHz and goes up to 20 kHz.

I tried changing brightness 10% and it didn't help.

I am ready to check the monitor myself (while turned on, I'll be careful with the PSU), but I need to know in advance how to treat whatever component may be producing it.

Which components should I check? I guess that only capacitors, transformers, coils and ferrites can produce such noise.

Can I simply abundantly hot-glue any suspicious component to keep it in place and reduce vibrations?

Edit

I opened it. It sounds like the noise comes from the PSU, but I cannot open it further to check inside it. See photos.

PSU

enter image description here enter image description here

Or LED driver board

enter image description here

Opening the PSU would destroy it.

However, I measured the voltage at no load or when the monitor was turned on, but only the electronics was powered (no LCD, no inputs, ...). The voltage was 19.8 V. It matches a text on the silkscreen of the main mainboard which refers to "19 V USB 3".

enter image description here

The PSU has only that single output, no feedback or anything. I think I could replace it with an external one providing stable 20 V (my tester in AC mode and 20 V scale gave 0.01 V "true RMS"). I'm not sure about the current rating I would need. If Dell states 120 W maximum, I need a 6 A PSU or better. I guess I can replace the AC socket with a DC barrel socket, which are rated up to 10 A. Laptop power supplies already provide 19.5 V and sometime 7.6 A, however I have no idea whether the output voltage is stable enough for the purpose. There is also a Mean Well 20 V 6A, 180 mV p-p with DC barrel plug.

In any case, I hot glued the inductors I could see, but the effect is limited.

It looks like it's a design fault, people had the issue since the beginning.

Result

I replaced it with an external Mean Well power supply and I replaced the US-style power connector with a socket matching the external power supply. The hissing completely disappeared. Now I only need to find some velcro to stick the power supply to the vertical column of the monitor stand.

FarO
  • 1,306
  • 1
  • 13
  • 33
  • 1
    Capacitor drying out. I’m betting 10 € on it. To clarify, the sound may be coming from inductor(s) but the root cause is an nearby capacitor. – winny Mar 18 '20 at 21:08
  • 1
    are you certain that it is an LED backlight monitor? ... fluorescent backlight monitors have a high voltage power supply that can produce an audible hiss – jsotola Mar 18 '20 at 21:31
  • @jsotola it's a GB-R LED backlight to offer a wide gamut: green and blue LEDs plus a phosphor for red. – FarO Mar 18 '20 at 22:50
  • @winny then I'm in trouble because I don't have an ESR tester – FarO Mar 19 '20 at 06:49
  • Probably not needed. Can you open it and take a closer listen without electrocuting yourself? – winny Mar 19 '20 at 17:13
  • @winny I think I can, but first I have to find an arrangement to have another working monitor in the meantime – FarO Mar 20 '20 at 09:47
  • @winny even if I listen and I find the source of the noise, without an ESR meter am I supposed to replace all the capacitors nearby? it may work, if there are not many – FarO Mar 20 '20 at 14:36
  • If you are a bit lucky, you will see a swollen capacitor. A bit less lucky you hear which inductor is buzzing and we can trace it to one capacitor. – winny Mar 20 '20 at 15:37
  • @winny I think it's the power supply: the noise is strongest on the side of the power socket and I see through the cooling grid some relatively big capacitors. But I understand, I have to focus on the capacitors connected to the noisy inductor. – FarO Mar 20 '20 at 15:38
  • Sounds more and more likely. ~9 kHz is also the crossover frequency of the output on most power supplies. – winny Mar 20 '20 at 15:46
  • @winny I survived. I updated the post. – FarO Mar 30 '20 at 22:15
  • Good! Power supply is probably fine. The driver board, can you see any leakage at the bottom of the electrolytic capacitors or swelling near the top? Can you turn it on while disassembled and take a closer listen without electrocuting yourself? – winny Mar 31 '20 at 09:11
  • @winny I did, but the LCD panel was not connected because some cables had to be disconnected to get there, so the PSU was running at very low power. Still, the sound was coming from it, or largely from it. No leaks seen anywhere, the LED driver board (which in fact powers everything, the PSU has only one output) looked perfect. I think I will replace the PSU with an external one. I could try to power the display with an external PSU (DPS5005) just to verify and to check it with the oscilloscope to see how clean the PSU needs to be, but I'd like to avoid opening it too many times. – FarO Mar 31 '20 at 09:29
  • For example Mean Well GST120A20-P1M (6A, 20V, 180mV p-p), since the original one, rated 120W at the wall, probably was not outputting more than 5 A, assuming 80% efficiency. I also have a MIC29752 low dropout regulator which could be useful to bring the voltage from 20V to 19.5V. I never use the screen at more than 50% brightness anyway. – FarO Mar 31 '20 at 09:41
  • @winny in fact it my be a design issue: https://www.dell.com/community/Monitors/UP2414Q-power-supply-sound/td-p/4532956 – FarO Mar 31 '20 at 09:42
  • Any snakes found? Hissing most likely switching power supply noise, only heard by some (and dogs.) – skvery Mar 31 '20 at 10:01
  • Do you have any 19V-ish power supplies laying around to test with? – winny Mar 31 '20 at 10:22
  • @winny I have a DPS5005 which is adjustable, up to 5A and 24V, so I can use it. In fact, if it's quiet, I can already wire the barrel jack and use that one while I wait for a proper PSU to arrive. – FarO Mar 31 '20 at 11:33
  • Fantastic! You are earning EE points by the day! – winny Mar 31 '20 at 11:35
  • @winny I did it, it worked. I survived, and monitor survived too. – FarO Apr 25 '20 at 20:33
  • Good! What’s your conclusion? – winny Apr 26 '20 at 08:12
  • @winny it was the power supply. – FarO Apr 27 '20 at 07:50

0 Answers0