For clarity: I have tried swapping the power supply. I've also tried power supply in the case/motherboard out, and motherboard in the case/PSU out, and both out and both in. And tried 3 different power supplies, two from EVGA and one from Gigabyte. It's not the power supply. 😜🤓👍 Update #2: Please stop suggesting I flip my radiator/AIO around, especially when referencing Gamers Nexus. I watch GN too, and I'm following the ideal positioning that Steve from GN recommends in that very video many of you are referring to. There must be a misunderstanding in the community regarding his video, even though he made two. Tubes at the bottom is the ideal way to go for front mounted radiators provided the top of your radiator is above your pump. This way the air bubble will be at the top of the rad and thus won't get sucked into your tubes at the bottom or into your pump. This is what Gamers Nexus recommends. To suggest otherwise is a misunderstanding of his video. And I front mounted my rad because it gave better temps in this build than top mounting. Cheers!
I commented below, but I’m very curious if you have tried running your audio out from the front headphone jack? Had an issue with noise in my speakers on a previous build in an R5 Fractal case and that was the only way I could find to eliminate it. (After rebuilding my entire computer with the same case, the problem went away.)
I wondered if the screws attaching the power supply to the case were not in contact with the metal - eg the coating on the case was preventing a good ground. I feel like there would be a relay in the power supply and when power is "off" the ground is connected thru fine like you have found, when the power supply relay is on then the electrical ground has to go through the normal route. I would think that even mother board mounting screws might not be grounding too well either if the coating turns out to be the issue aswell I'd scrape away the coating in as many places as possible without destroying the look of the case
There is leak current. That is why you don't get continuity when on. Also the meter can't meter resistance with voltage applied to it, and current leak will apply a voltage differential in the ground. Also the meter introduces a very small voltage and current to meter the resistance and it might not be healthy to do that with the motherboard on.
Gerald, I'm an electrical engineer, and I can tell you it really does not mean I know any more than you do about this, as I focus on firmware design. However, having recently worked on troubleshooting some similar problems and since I work with other engineers who do this sort of engineering on a regular basis, I think I can help. One of the things we produce at work is audio equipment for medical environments. As you can imagine, they get pretty picky about safety and isolation. Ground leakage and ground loops can be a big issue, and often the best way to make sure you're always getting clean audio is to isolate the grounds, and even electrically isolate the audio if you can. Digital audio is good, since it can error correct, but in your case your USB cable was carrying a ground issue, so it wasn't the signal, but the ground that was the problem. The USB isolator will probably be a good answer regardless of the source of the problem. As for the humming sound, you're correct in suspecting the LEDs and coil whine. Both of those can radiate or conduct into the ground in systems, and a good design will actually put some effort into doing things like making sure any RGB LED PWM frequencies are outside of audible range. For example, I PWM high power LEDs in my designs at 16KHz, and while technically not outside hearing range, you're probably not going to get that to conduct well. A 4KHz PWM on the other hand, could be more problematic. Coil whine is just going to be a problem with inexpensive electrical design on a board, and you'll have to do things like isolation to get rid of it. Not only can you hear it, but the sound comes from power electrically oscillating, causing parts to wiggle at that frequency. You can bet it's electrically propagating through the system somewhere, and you'll hear it in other places where electron shuffles can make it into some kind of audio transducer. These problems often travel along the ground plane, since it's designed not to have any attenuation in it. If your ground plane has small, leaking frequencies on it, where it's varying by millivolts in audible frequency ranges, as soon as an amplifier uses that ground to produce power, you have now amplified that ripple on your ground and mixed it into your audio signal. Some systems will employ methods to try and filter the ground plane, but it's not a trivial problem to tackle. Again, this is where digital and optical isolation can help if it becomes problematic. If you connect your audio amps to your system with optical audio, and the speakers have their own power supplies, this should go away (unless it got onto the audio signal before it was converted into optical). This is harder to do for microphones, but often not hard for speakers. However, USB audio usually helps a lot, unless again, it's carrying this noise over on its ground and 'infecting' the amps with the noise. Also, there's a lot of opportunities for audio signals traveling on motherboards or other cables to pick up noise, so sometimes all this isolation won't help. The inside of a computer case is very electrically noisy. Every cable is a radiator of EM noise, and there's a lot of them inside a case. All that being said, I'm not an expert in this, but these are problems I know we've had to deal with at work, and I've personally had to fight in my systems with audio. Since this isn't my area of expertise though, some of my explanations might be a little off. If so, I'm sure someone will be happy to vehemently correct me. And sorry this is a month late, I only recently found your channel and got an interest in video.
This sounds solid to me. I used to work as a sound-man for a rock & roll touring band. I chased a ground-loop off a stage once by putting a carpeted riser under the head of our cable-snake. We were getting 60 Hz hum from the lighting dimmers that conducted through the metal stage-frame and made contact with the metal case of the snake-head. That was literally a "ground-lift". Cheers. = ]
Continuity testing rarely works well when a device is on. As current gets leaked through to ground, it causes small voltage differentials to appear in the chassis ground, which will confuse your meter. Also, I’d generally recommend against continuity checking while things are on anyway, as this applies a small voltage to the probes, and this can confuse sensitive system components (to my knowledge the highly limited current released by the meter is unlikely to cause any actual damage, though, but I have by times seen it crash a system, or introduce glitches).
So basically continuity measurements only work in circuits without an applied voltage? I didn't know that! So the meter looks for exactly the voltage it sends out?
lightsout Yep. Only “reliably works” in inert circuits. And yep, it’s looking for a trickle of current of its own production, which can easily be fooled.
IANAEE, but i was wondering if this was the appropriate method too. Wouldn't the appropriate way to test for grounding issues be to set the multimeter to VDC and measure the voltage difference between ground points?
Nick in theory, yes. But while that might work while the machine is powered on, there are so many ground points where would you start? A lot of grounding issues are caused at the component level (missing or broken decoupling capacitors, broken shunt resistors, etc), and would be nearly impossible to find with this method, without a schematic, micro tipped leads, and a lotta time on your hands. Generally the best way with advanced micro- components like these is to just do what Gerald did and swap out components until you’ve whittled down the list of possible culprits. (IAAEET? Is that the acronym? Lol)
Gerald, there may still be a "relative" ground issue when the unit is powered on. It is possible that the power supply has multiple V+ rails (most do). I suspect the power supply manufacturer is doing something funky with the grounding where maybe they are isolating each rail including ground so each becomes its own independent circuit. You may be experiencing the lack of a common ground between the circuits when powered on. You could verify that by testing the ground pin on each rail against the other rails to see if they are common when powered off but separate when powered on.
The PSU would also be one my first suspects, I don't think Gerald tried to swap the PSU did he? The case being a factor is also strange, maybe the way the PSU sits and/or cable runs through the case worsens the PSU problem (if it has a PSU problem)
JoaoRelego the FractalAudio (no affiliation with Fractal Design) thread I dug up says there is a major difference between Fractal Design R6 compared to most cases. Isolators. Rubber sound dampers separating some parts of chassi from others. Which FractalAudio says are bad for EMI. However: Other threads says the total opposite, that you should isolate everything and never ground chassi, only ground through though PSU cables. Apparently there is a ground design war I wasn’t aware of before.
@@JoaoRelego I have the same case with a different build and have never had an issue like this. I am not a computer expert, but I was considering looking into a build like that in the case I currently have. Is this something I should be concerned over or are you thinking it may be the power supply?
Remember the good old times when Motherboards were green or yellow and you had red paper washers with your board that you put between the screws and Standoffs in order to Isolate the board? Yeah ... I do. Maybe an issue there. Continuity is there when off because there is a proper path. And as soon as you power on, somewhere along the lines of a screw or so it doesn't find it's the proper way home. Tolerances are tighter and the space where the screws and standoffs are, are tiny compared to "back then". So maybe worth checking. For the coil whining and hissing: What happens when you turn your RGB Memory OFF. So no lights? Will that help? I would suspect a high freuqency module to PWM the RAM and this causes issues. Same for the GPU. You could probably check that if you "manually earth" these components. I.e. putting a crocodile clip from the GPU bracket to the PSU screw or something like that so it "has it's proper path" that you don't have when you power on the case. The Motherboard of my wifes PC has some form of "metal sponge material" to it on the motherboards "shield" so it properly mates with the IO shield that you put in the Case before installing the board. These Motherboards don't seem to have an IO shield anymore because it's all integrated in the big "block" now.
I get interference with sound when using LED fixtures. I work for a company that does concerts. We have conventional lighting as well as led lighting. We have to make sure that the lighting is on a separate leg of phase in order to limit the interference with the sound or video. So generally when a venue has 3 phase, we lock in our power distribution panel, and we plug video, lighting and sound into separate phases of power.
Great video, thanks. Around 17:30 you mention the lack of a "clear cmos" button on some MOBOs. A trick of the trade is to hook your reset button up to the "clear cmos" jumper. Sure, it has some drawbacks, but it is an option.
I was an electrician for 15 years. Normally for that intermittent grounding means something was wired incorrectly (switched neutral or even switched ground). But since the problem is only with the case, which has no wires...its almost like there's current running through the case, but that would point to a power supply problem...WAIT, IT’S probably the power cord or the plug on the back of the case. There might be a short there.
Speaking of PC’s... It would be awesome to have you make a video comparing monitors and their color accuracy for rec 709 and above. Haven’t found much content for PC’s about good reference editing monitors that aren’t $3k
I see LG OLED TVs being used by a lot of professionals. You'd need to buy a Blackmagicdesign Decklink or Ultrastudio to calibrate it, but I've heard that you can upload your own LUT into high-end LG OLED TV models.
BenQ has been offering 2 good monitors for editing for a few years now. The PD2700U and SW271 monitors (other sizes are available) are under 1k dollars actually. Many photographers and TH-camrs use this and it's not only for the price it fits in, but the other aspects too. I hope this helps :)
@@neiltodankar8034 I personally use the PD 2500Q which is the 1440p version in 25" as you said specifically made for rec 709/srgb color gamuts and it's great, i didn't believe screens could look so different until i upgraded
Little tip: Connect your RESET button to the CMOS clear jumpers. It's really easy to reset that way, and if you wanna restart your pc, you can still use the power button. Probably even more convenient than using the motherboard build in buttons (cause u won't have to open the case)
I'm totally stealing this too. That's a great idea for lazy computer tinkering. Also, for mobos outside of cases, I've always just used a screwdriver as my power button. You just bridge the pins for the power button, which are usually nicely color-coded these days!
I was looking for B550 onboard audio port benchmarks, this wasn't it, but it's still very much appreciated. As was the unintended review of the iFi product. Ran into similar issues my self with certain USB DAC/Amps and very sensitive in ear monitors years ago, ground noise going over the USB cable I couldn't solve without ripping the ground contact off the cable.
0:57 Audio interference is usually not mainboard, it's often the front panel of PC enclosure. If you're unlucky it connects Audio GND and USB GND together which is a big NO-NO. It defeats the audio star-ground on the mainboards and creates a ground loop. But it wouldn't matter if you use external interface exclusively and don't have any audio connection to PC. The ground routing can then cause the loop created between USB and audio to become an antenna which will collect all sorts of RF interference, predominantly from all the switching voltage regulators like on the GPU and elsewhere and from the PWM light control etc. To check: disconnect HD Front Audio connector from mainboard, the noise on rear audio out can disappear. You want there to not be continuity between front panel audio ground on the case and any USB ports on the case, but if you connect mainboard, the continuity will be measured through mainboard. Longer ground loops are also possible. It's needed to consider all the equipment, not just the PC, but everything connected, all the connections made. I don't believe you showed all in the video, the interface was only mentioned in passing. A few minutes in... the continuity between protective Earth pin of the PSU, the chassis, and all ground points must be continuous, the mainboard per se cannot defeat it. This is weird. Why it's not measuring like that... injected ground loop current? Wait a sec, what multimeter do you have, and does it measure continuity on the resistance range or on the diode range? I think it would be curious to switch all the multimeter modes and even try current and voltage. You want there to be zero current through the multimeter between ground points and you want zero voltage between these points. Very important, do you have an extra PSU to test? Because i don't 100% trust yours at the moment. Please perform resistance measurement from middle pin (ground) on the IEC power socket to ground outputs on ATX connector and others. First you check how much is the resistance of your multimeter and multimeter leads by connecting them directly together, you expect to see something like 0.3 Ohm, and then you check the ground path resistance through the PSU, and you expect it to not be significantly higher, maybe by 0.1 Ohm at most. It's also vital that PSU enclosure screws are similarly well connected. Because most of the switching noise gets disposed of by shunting it to house earth, so earth impedance must necessarily be very low. And well it must be low all the way to the breaker panel in the house.
Yes, I think that's what's happening. The USB is acting like an antenna for the noise in the system due to the grounding issue. I have tried another PSU, didn't make a difference. The only thing that resolves the grounding problem is switching cases. The results are the same even if I measure from the ground pin on the power cable to the computer. Works when off, not when on. The case has to be doing something right? Why else would it work in a different case?
I really like that Gigabyte wants to work with you to figure out the issue. Also, I find myself humming and singing along to your intro all the time. lol
At around 13:05 I started to laugh. I just laughed at how ridiculously relatable and lifelike this video is. How the second you turn on your newly assembled PC and it runs into a bluescreen, or a hardware failure, or basically any of the endless possible issues that can happen to your build, and the itching feeling of utter annoyance but also nerdy interest and solid determination to solve the issue, and the testing, and the monitoring, and the reassembling, over and over again... Is it a bad motherboard? Did I install my drivers incorrectly? Is it a faulty memory block? A small screw short-circuiting the electrical loop? And when you thought you have finally figured out what the problem was, it turns out it is in fact multiple problems, weirdly interacting and aggravating one another. Aaaand here we go again. Painful bliss.
Hey Gerald. I had that same sound problem in my speakers when I connected an external GPU to my Macbook Pro . The way I solved it is by putting an adapter to the electrical cord with only two pins, the ones that do not have the ground pin. That solved it right away. Seemed that there was some ground coming back thru the egpu ground pin and entering the computer thru the thunderbolt cable. Great video.
Potential electrical explanation for the high pitch whine noise: I have the same speakers and the same issue and the problem comes when using simple instrument audio jacks (mono) for the monitor speakers. They only have 2 wires and short out the shielding (ground) and the return line. Now you could either use a proper audio jack or tape of the ground pin on both speakers power cable!
If nothing else, it’s a great exercise in troubleshooting! The tough part is when you have multiple issues when you think it’s just one. I probably wouldn’t have watched this but I like your channel and I recently updated my desktop system with an Aorus X570 Elite purchased for $209 US, which I thought was a great deal. I don’t do video, but I enjoy watching you get Undone!
I had a similar problem awhile back... It drove me crazy. However, since it was months ago, I don't remember exactly how I solved that problem. But I do believe that I can direct you in the right direction. I would start with the wires. Particularly any audio cables, from the speaker, the mic, and audio interface. That sound is digital/analog noise. It can come in the form of crosstalk between touching wires... a bad insulated wire. A speaker wire that is exposed slightly from the speaker negative/positive connection. If I had to swap out hardware, I would swap out the audio interface. I would also swap out audio wires... Also It could be the microphone and their magnetic fields, relative to the speakers and their magnetic fields. Try and update the drivers from the audio interface too. I strongly believe it was a wire though...
Back in my day we solved our nasty coil-whine problems with a blob of hot-glue. Of course the coils were a cm square. I modified a stethoscope with a broken dental-pick (just jammed the tool up the tube in place of the doctor-disc). It was useful for isolating mechanical/servo problems related to friction & resonance (or vibration) and it could be used on non-conductives like hard silicone. You want to beware of resonant frequencies; they exist in objects as well as circuits. Lots of shielded heat-sinks helps on the MB. When a full spectrum sound-wave (eg: pink noise) meets an object with a resonant frequency the reflections will amplify just that particular band four times, leaving every other frequency unaffected. In a feedback loop of your amplified room that will cause destructive feedback, (ringing, squealing or thrumming.)
I can't help with the technical issue but I can say, the X570 MB paired with the AMD 3950X processor just rocks. It just chews through 4K video editing. Great work.
Most random thing. I watch your videos for all the other reasons and here you go and test a MB that I swapped to (from Taichi to Master [570]) TODAY! All I can ask/suggest is that you try to disable the RBG altogether of the ram, if it's not the RGB in itself maybe disable XMP (or enable, whichever you didn't do)? Keep it up! One of the best, if not the best cinema-tech channel out here! :)
For the issue with your RAM - it might be the LED's in it: I had internet problems regularly after I renovated my office. A month ago I had enough and called a technician to have a look. He looked around for about an hour, checking all connectors and swapped all old cables for uninsulated ones, but nothing changed. Then we turned the power off to check if it's a unisulated cable in the wall somewhere and suddenly the problems were gone. When then measuring the current on my outputs we got no error for the power cables, so we resorted to go piece by piece. Finally, when we were turning everything on again, step by step, we saw a sudde drop in speed and multiple interferences causing jitter when I turned on my LED desk light which was standing right beside the router. I swapped the LED bulb and still got the same issue. Also moving the lamp to the other side of the desk didn't help. Since the lamp is gone now and replaced for a regular bulb my issues are gone aswell. Now I don't know how this correlates to your coil-whine issue, but I assume it could be similar. It could either be that LEDs in general can cause interference in your system (especially since they're so close) or that the insulation around the diodes on your RAM is faulty and I as well had bad luck with two janky lightbulbs.
Wow - some great effort in troubleshooting there. I come from many years playing in touring bands - a ground lift was in almost every musicians bag for basic but similar problems. The process was plug something in and when the amp is turned up, if there is hum or noise like the what you were getting, then pull out one at time anything that has a seperate ground til you find the culprit and the reverse polarity or lift the ground. Plug it back in and in most cases that would fix it. And the same gear could actually be fine some places and not other places. So troubleshooting that - the venue or arena power would make a difference. But kudos to you for being so vigilant - while you were going through the whole thing here, I was thinking that I probably would have just plugged in the USB ground loop isolator and called it a day. A little out of my tech interest here, but still kind of interesting.
Hey Gerald, just chipping in - albeit late, but I think it is worth it since these videos of yours will be great reference for others. I had the same problems with a different system in a Fractal Design Define R5, connecting to an external DAC and the DAC to my studio monitors. Changing my DAC to an XLR one and connecting DAC to monitors with XLR improved it, but I had to get a USB isolator to remove the ground loops completely. This was happening both over optical as well as USB. All I want to say is that it is not necessary to spend 80 bucks on the iFi defender, since designs based on the same chip are all over the place and cost betwewen 15 and 45 bucks depending on whether you want USB 2.0 or 3.0. With 2.0 it will be enough for 24/96 audio. If you go for this route one only needs to look for ADUM 3160 chip based isolators for USB 2.0, and ADUM 4160 based designs for USB 3.0. They all work. If someone wants to take it further (and spend that money saved on the defender elsewhere), 50 bucks will get you a linear USB power supply with a 15w toroidal transformer inside, which you can use to power the isolator for an even cleaner connection (probably not necessary though). I am starting to think that the Fractal cases are the culprit, but I can't be 100% sure, since these ground loop issues are a bit "ethereal", as I'm sure you already know.
I built a 3900x system when it was launched. I almost bought that case, but wound up using my old one, I'm glad you pointed out the case issue. I have the x570 Master, it's been great, unfortunately, 1.0 doesn't have Thunderbolt like the 1.1 and later does. I wish they'd make a proper add-in card for it. Enjoy.
I'm not an electrical engineer by training, but I have designed a number of low voltage switching power supplies for LED and wireless applications, and consulted with a number of actual electrical engineers during that process. My understanding of the noise problem is that the frequency of the noise you hear can help you diagnose the problem. If you're hearing 60hz low-frequency noise, that's ground loop hum coming from the AC power supply (in some applications this is also testable by switching between battery and wall power, but if the device is inherently wall powered, you wouldn't be able to do that test). Higher frequency noise would be more related to switching power regulators, which are used to convert AC to DC power, and also to change the voltage levels of DC power (e.g. 12V to 5V, 12V to 3.3V, or even boost from 5V to 12V). The frequency of that noise can be used to determine which regulator is causing noise, if you know the specs to the regulator. Immediately pinpointing a specific regulator on a complex device like a PC is probably not possible, since there are a lot of switching regulators. But one possible hint in the diagnostic process is that the frequency and magnitude of noise will change as the power draw on the regulator changes. The problems you found with the LED RAM and the GPU under load both seem to point to regulator behavior where under low load, the switching frequency is outside the range of sound, but under increased load, the frequency falls within the range of human hearing. (It's actually a little more complicated than this because regulator designers know that this noise is undesirable and can mitigate the noise using filtering capacitors. However, those capacitors are selective in the frequencies that they are effective in filtering, so one may find that increasing or decreasing the load by some amount might increase noise, then further changes will decrease noise as the frequencies shift into and out of the effect range of the capacitors being used within the design.) This noise is propagated on the voltage and ground lines, so it makes sense that even though you don't have a ground loop, powering your audio device with a separate isolated power supply would result in the noise not reaching the device.
Also, perhaps I got too caught up in the diagnosis part of this, and lost focus of a solution. Actually trying to balance out the power draw of all your components to evade regulator noise at audible frequencies is overall, probably too hard. Probably a more stable solution, mentioned by a number of other commenters, is to ensure that the audio equipment that you are plugging into the USB ports have their audio handling power paths isolated from the data paths, which unfortunately means getting away from bus-powered devices. Some bus-powered audio devices may attempt to mitigate this with device side power filtering but it's hard, but probably the most effective solution is something like an RME Fireface UFX series device, which is independently powered. The difficulty here is that even if you manage to balance out the noise from all of the internal components, other devices you plug into the USB ports might be noisy, or simply change the power draw to the point where you enter the noise range again. Keeping audio device power segregated insulates the setup from that kind of instability.
Oh, I know how this feels. I had a similar ground loop issue, I too went through the gamut and gutted the entire system, isolating everything to the bare minimum components in attempt to isolate, removing the motherboard from the case, testing ground with a multimeter on my receptacles, etc. Ultimately I RMA'd my motherboard with Gigabyte as I had a faulty sata port on it anyway, so thought I would kill two birds with one stone. The ground loop only presented itself when using my RME audio interface as a USB Dac, and connecting a pair of its outputs to RCAs on a headphone amp. Extremely annoying and pulled my hear out as I invested a ton of time without a clear resolution. Ultimately I moved away from custom PC builds as it was a Hackintosh and it just killed way too much of my time troubleshooting over the period I had it (without being 100% satisfied with the user experience)...I'm now running full OEM simply to save time.
Meanwhile my PC case has been stuffed with so many Hard-drives that I cannot put on the side panels anymore. I got an SSD taped to the roof of my case even. The Jankiness would give Gerald a heart attack.
Check for voltage on the case. To do this, carefully measure between the pins on one of your extra four pin drive power plugs and the case. See what you get. Also, replace the power switch and see what happens. You can then unplug each wire going to a case extension (leds, switches, fans, usbs, etc. add them back one at a time and see if the issue appears with one.
One other thing you can try is a power conditioner or UPS. A lot of them are made to clear up power signals before they get to your system. Furman makes a bunch but there are also plenty of UPS (uninterruptible power supplies) you can get online that condition power and help provide stable perfect voltage to all of your computer components. Coil whine and other leaky components in a computer all contribute to EMI and electrical noise which is why designing a good motherboard is super difficult for electrical engineers. Isolating everything on different lines in the PCB and on different power rails isn't easy. Typically I don't recommend using front panel IO at least for audio. Try disconnecting that cable all together as well as UPS and using an external Audio interface if you aren't already. In some cases the only way to fix the grounding/signaling issue is with one of those nifty USB devices you already got. This is why most audio pros recommend against internal audio solutions no matter how well the Motherboard is designed. You simply get less EMI and interference by moving the audio components into their own external enclosure a few feet away from the CPU, GPU and Power supply in the PC.
I would say it's the power switch panel on the front of the case. When off, the switches on the front panel make no contact.... thus not interrupting the ground. But when engaged (powered on) it's shorting out somewhere, causing a ground issue. That, or somewhere there is a spot rubbing/touching the case doing effectively the same thing. For your next vid: :Tell me where the big bad motherboard grounded you..."
That seems reasonable, but I just checked and the problem persists even without the front switches connected. I powered up using the X570 on-board switch. So it must be somewhere in the case construction.
much much respect for the thoroughness and your willingness to investigate even intermittent problems. working in the software QA field I can kind of relate how pesky that can be :P
I experienced a very similar problem a few weeks ago connecting a USB powered microphone to a laptop via a Blackmagic ATEM Mini. The USB mic works fine when connected to the laptop but if you then use the 3.5mm audio cable from the mic to the ATEM Mini and then of course USB-C from the ATEM to the laptop you get the same pulsing electrical interference. I suspected that it was an earthing issue of some form but your video has (I think) given me the final clue to the issue and probably explains how your issue is related to the case. * I think the issue is not a failure to provide a ground but that the ground is actually receiving current. In effect I think that something is actively grounding (shorting) to the case and that is what the mic and sound card are picking up. * This also explains how it is related to the case in that if the case mounting is actually touching something it should not then your ground would be live. * The USB grounding gadget would resolve this probably because the current is so small that the ground lead just provides a lower resistance path and thus neutralizes the voltage. (Voltage is actually Potential Difference after all) * I think that your multimeter test is failing when power goes on because you are no longer testing a dead circuit where the meter's current is the only current being measured. * The link to the RGB ram could just point you towards what is shorting down to ground. All just guess work but hope it helps.
Gerald - great video but I think you’ve approached the problem in the most difficult way. Computers are nightmarish when it comes to controlling noise and interference leaking from the motherboard and its hardware into peripherals and audio interfaces. You’re better off using properly designed interfaces that are able to deal with the noise - or galvanically isolate the connection between the computer and the audio interface. One of the easiest ways to protect from usb noise is to use optical out of a computer - however this may not suit your needs. The iDefender ground loop isolator is a good solution - but a well designed audio interface should do this in the first place. The Zoom F6 is subject to significant low frequency noise from its poor power filtering (see testing on AudioScienceReview). Might be worth investigating the sound devices gear which I think is better engineered.
I know this video is a few years old now but I've been having the exact same issues with gigabyte motherboards. After talking to a friend who IS an electrical engineer and trying a bunch of different things, (cheap ground loop isolators, different plug configurations, different computer component configs including cheap vs expensive PSU's) it turned out that most likely it is because of how the motherboard is attached to the case. This is consistent with how you experienced this problem in some cases more than others, and didn't experience it when the system was outside a case. The system is successfully grounded via the PSU, however the motherboard also has a special (extra?) grounding design where it grounds via the standoff screws (consistent with your definition of ground loop being caused by multiple grounds). By putting small plastic washers between the screws and the case, I believe this will solve your final persisting issue by making sure the system grounds purely via the PSU. This would avoid having to swap out different components, use that USB isolator etc. This is just my hypothesis however, based on your tests (thank you!!), my tests, and discussions with my audio engineer friend. I've been putting off trying it as it would involve rebuilding my very tedious compact watercooled ITX build but I can report back to you with my results if you're interested.
I had the exact same issue! I would notice the issue when scrubbing through media in premiere/when the system was under heavy load. Unlike you, I got this through the front headphone jack, and interestingly enough, we have the exact same case! Fractal R6. Other than that my specs are completely different, Z270 board, I7 7700K, RX 5700XT, 3200Mhz Corsair Vengence (no RGB). I fixed the issue by buying a Fiio K3 DAC, as I figured it was a grounding issue in the front IO panel. Super interesting hearing your side of this.
Gerald, I am an EE. I'm not going to explain, but give a couple of hints. 1. get rid of the Yamaha monitors. They have no shielding, electrical or otherwise. I switched to Adam. Next use an oscilloscope not a VOM. When the PS comes on it puts a potential on the ground that is higher than the small battery in the VOM. thereby the VOM will not see a current draw from a short and will not sound the short. I had the same issues as you, the real culprit is the USB design from Intel. Note that some motherboards are shipping with a special USB that has a separate filter on it. The defender is a little over priced but like a cloud lifter it has no competition so there you are....
Thanks for the detailed review, I have a similar issue with my B550 Aorus Pro AC. The issue is my keyboard both wireless Sculpt keyboard or my wired keyboard goes erratic if attached to USB 2 port but works fine if attached to USB 3 port. The mouse and every other peripheral including Logitech camera, external USB drive and few other work fine. You are right when you mention about the PCI 4 in B550, I do have a nvme attached to the top slot running full PCI4. Maybe a future BIOS should fix this.
@@karama300video for Windows I use the driver supplied at the Gigabyte site and so far no issues but I find the internet slows down over time and it needs a reboot after that. Probably a windows issue as the ethernet runs fine on Arch linux here with the latest kernel.
@Haimura thats their top of the line board, I have the aorus pro AC wifi version and so far everything is running fine now with BIOS update. The issues that you mention is suffered by other manufacturers as well using B550 chip. Make sure to update BIOS asap after purchase, use the beta bios.
A strange but anecdotally common reason for coil whine is the power source, ie. the quality of the electrical power in the building. Again, it may sound strange but it's worth checking out if the coil whine still happens in another building. Power supplies / different components may be more resistant to coil whine so switching those may help even if the core issue is the quality of the power from the wall, leaving the impression that it was just a bad component. Also, if you're using a UPS try it without the UPS.
7:09 this is an expensive contrast card. Here's the best thing about this video. This is the video I expected from Gamers Nexus. I always considered Gerald Undone to be the Gamers Nexus of video, audio tech and Gamers Nexus to be the Gerald Undone of PC tech. I haven't had any issues with either the B550 Gigabyte Aorus Pro or Gaming X but I did put it into multiple legacy cases from an era where airflow really was a thing. From my experiences and my understanding the majority of the grounding is done from the PSU cable to ground which would explain why turning on the PC changes your ground readings. The grounding not accomplished by the PSU would normally be such a low variance and go without any issues minus what would be audible buzzes or static. The Aorus Master has more high power phases and maybe surpassing that variance in some pc cases resulting in the issues you laid out. Just a guess really. Great vid and support from Gigabyte. Thanks Gerald. Have a good one.
I'm not an Electrical Engineer either, but my guess would be the the RAM (or it's RGB controller) and the graphics card are emitting radio frequency noise and the case, due to it's sub-optimal grounding, is picking up the RF noise like an antenna and leaking it into the ground circuit. The USB defender could be interrupting the ground connection to the PC and grounding directly to the mains thus blocking noise from the computer's ground. Or it could be grounding the entire case via the USB and motherboard, and thus resolving the case grounding issue.
When your first video posted I just finished installed the B550 Aorus Master in my system. I didn't have any ground loop issues so I assumed it was something else in your system but forgot to post. I did have strange USB issues like you though, so I ended up returning that board for the X570 Aorus Master which is a great motherboard.
I think that powering up might generate eddy-currents in the case. The whine may be switch-mode power supply related. The radiated magnetic field is affecting the audio circuits by leaking into those circuits and being amplified or there is insufficient filtration of the switch-mode power supply lines and the power supply frequency is audible in the audio circuits. The USB filter is filtering out the power supply noise.
I started watching this video in the TH-cam preview panel on my phone, and the captions try to translate what you are saying. It thought you said "Hi, I'm jailed Undone" and that stopped to make me laugh. You should say that, Hi, I'm Jailed Undone... as always, awesome video... you make every other youtuber look devastatingly terrible. And you make me personally look like an idiot. But I'm happy to learn.
The only reason the case may be causing any grounding issues could be the front panel header. That’s probably the only place where other (non grounded) electrical connections can make contact with the case.
Hey, checkout the Silverstone CP14-E adapter (like $20-30 on Amazon), if you have an extra USB3.1/3.1 header you can use that to convert to USB-C header. I had the same issue, my Asus Pro WS X570-ACE didn't have internal USB-C header, I used that adapter which worked great! No USB-PD so no fast charging of phones and such.
Very thorough. I found it educational and entertaining. Sorry that I can offer no good suggestions on the continuity issue, in particular. I even learned stuff about continuity testing on a motherboard. :)
Gerald, you and I have similar minds haha. Quite happy to see you use a Fractal case just like I am. I'm using the R5 though and have had no issues. Also, I think just turn off RGB.... on everything. I'm not sure if the leak issue is just with Gigabyte stuff for this time's lineup. I'm using G SKILL ram, no rgb. For GPU, I installed the software just to turn RGB off. I'm using an Asus Z370 Prime A mobo... perhaps it's their lineup of mobos that are causing the issue? Also, I always have my pc plugged to some power bar... I only use Tripp-Lite and APC.
You should had a case as a control in the experiment. This would remove the possibility that the standoffs in the (Xi-?) case had a different impedance to ground to measure. Speaking of which, did you happen to note the impedance to the case of the PSU to say the top of a MB mounting screw?... Then measure Ohms / Impedance between the standoff and the PSU mounting screw. Those measurements should be very close. The USB adapter probably has an optoisolator/optocoupler hence the cost. Once the ground loop is created you are more susceptible to EMF from the oscillators in the case, as it acts like one winding inducing all kinds of switching, oscillators ( The RAM for instance, or video card )
I can appreciate your video. I've been building PCs for many years and don't ever recall having so many issues at one time. However I do have an issue with one of my systems where I get computer noise coming through my USB audio interface. I've tried switching interfaces and a host of other things but it's still there. I'll have to give the USB ground-loop device a try. BTW - x570 boards as SOOO much better than the B550.
I once spend 3 days trouble shooting a computer that wouldn't post only to find out it was a grounding issue with the case. Outside of specific complicated mobo problems, case grounding is the most enigmatic problem causer of all.
LED's dim by switching on and off very fast, so your theory with the RGB being the cause of the noise is actually very good! I dom't have proof though.
Wow! I had that problem with my old B350 Gigabtye motherboard and Ryzen 1700X using the DAC USB (using M-Audio SuperDAC). I also had Corsair LED RAM in it. I tried different USB ports and still that high pitched sound. Tried reversal polarity on the powered M Audio BX 5's and the sound was still there. But through the 3.5mm headphone jack, no problem. I've upgraded to a X570 Gigabtye Elite and 3990X with RGB Corsair RAM and just a slight hint of that high pitched sound. Playing anything masks it completely.
Hi Gerald, very nice and informative content, enjoyable as always! There is other solution on this case, just use PCIe soundcard (Creative Sound Blasters, EVGA NU Audio, ASUS Essence STX II or even ext. Pro USB soundcard/Audio Interface: Focusrite, Steinberg, Motu, M-Audio, Apogee, etc.) on problematic motherboard (with audio interference/ground loop/) like on AORUS B550 Series, and even on the GPU coil whine problem, it does great job of solving noise issues and get better quality on audio playback and recording.
great video, i have that same sound with my blue yeti, i was looking into that ground loop. for my next system build next month for my AMD 4950x with 64gb i will be using the 570 master and memory without RGB. system will be nvme only
Gerald- I don't know if this is too late to the game, but I would pull the power switch from the equation. Leave everything in the PC and try powering up the PC with a separate switch not connected to the case, or just shorting the terminals on the motherboard. My guess is there's something up with the design of the little switch mechanism itself, or maybe there's something poking out of the switch box when in the off position that goes back in when the button latches in the on position.
I would take all of the case wiring and diagnose it for issues. Could be the USB ports or sound causing some issues. I’ve had cases that just never worked right or would crash if I used the case ports to connect peripherals or external drives. Many times it could be a poorly seated wire on the case button area that came loose was poorly seated during manufacturing or got bumped loose during installation.
Nice investigation. I've been having problems with my X570 AORUS Ultra's USB ports as well. No blue screen, but depending on what I have plugged into different USB ports, certain ports on will seem to turn off and then turn on again. I notice this most with my Bluetooth mouse/keyboard combo. I'll be watching a Gerald Undone video, say, and I'll try to pause it to go back and relisten to a very complex point you make, and I won't be able to click on the video for a few seconds. Then, I'll get a notice that a device (the mouse) has been plugged into USB and I can use it again. So annoying. After watching your video, I'm thinking I got a problematic board.
Thank you for the video it was very informative. Right now I am get a ticking noise is my motherboard when it is under a load and I am in the process of figuring that out. I used to have the ticking noise even in idle, but since I figured out that my Corsair PSU fan was not set to on it only happens when I play a game. What is really annoying is when playing AC Valhalla everything is fine for an hour then I get severe glitchy movement in the game.
Hello, coming to this party half year late, but maybe I can attempt to address the ground issue, as I work in similar field. Power return is not the same as ground. Some signals (analog, such as audio) use ground for its reference. But power equipment like fan and lights should not use ground as a return path for its power, because PWM (I assume most people here know pulse width modulation) will introduce noise to the power bus. But if say your RAM (which does not have a dedicated power return for RGB LED?) uses the RAM's ground for power return, its RGB PWM will introduce switching noises to the ground, which is used for audio. The fact that removing the RGB-version of RAM solves the audio problem supports my explanation. Also, the ground check with your digital multimeter, when the computer is turned on, the RGB RAM may have cyclicly messed up the ground hence messed up the multimeter reading. I hope this helps explaining the puzzle somewhat. One of the reason I prefer using optical audio device is that it completely decouples the audio signal from the computer's electrical ground and any issues it may cause. I also design and build audio equipment on the side, and I have found that if a cheap power supply is used, it will introduce high frequency whine to the audio as well. And I'm talking about the speaker's power supply, not the computer power supply. However, I assume your speakers have decent power supply, and thus not the cause of your particular problem.
I'm rockin' the Aorus X570 ITX and it's been solid, but of course you would run into some crazy issues on your build! On another note you can turn on motherboards that don't have a built in button by shorting the 2 power pins in the front panel io breakout.
Hey Gerald, friendly tip! If you can its better to orient your AIO Radiator with the tubes on the top, this way it guarantees there arent any air bubbles reaching the pump and making it perform noisier.
It's actually the opposite. If you're referring to the Gamers Nexus video, he recommends tubes at the bottom. This way the air goes to the top of the radiator, which should be above the pump and thus the air will go to the top of the unused portion of the radiator preventing air from being pulled into the tubes at the bottom and into the pump. I'm worried about this misunderstanding--you're the 4th or 5th person who has commented on this very video with that tip and the tip is just incorrect. I thought Steve's video was pretty clear, but perhaps there is a larger misunderstanding in the community regarding this. Here's the timestamped link: th-cam.com/video/BbGomv195sk/w-d-xo.html Hope that helps. Cheers!
Not an engineer, but here is my best guess. If you have a meter set to test continuity, your are measuring resistance, and low/no resistance is equated as continuity by the meter (and you get a beep). If however, your meter leads are place in two locations that have a "difference of potential" (aka voltage), then your resistance measurement is disrupted in the meter and not seen as continuity. At least that's how it works with my meter. That being the case, I would guess the case is having voltage bleed to it after it is powered up. switch the meter to measure voltage and place the two leads in the same location and see if there is a voltage when the pc is on, if so I would start by checking case lighting.
i believe the leds introduce weird interference, i can hear hissing & humming on my led panel. Its more interesting because there are no moving parts. I guess its the coil whine of step up converters for the leds.
I have the x570 in my editing rig. I really like it. Still haven’t filled the 3rd m.2 in there. When you connect headphones to the front panel and set it to 2nd audio output it has a 3step amplifier which comes in handy for my 80ohm beyerdynamic headphones. I set the chipset fan to silent in the bios and the temps don’t get anywhere close for it to even turn on.
Coil whine is a mark of a cheap component. You could swap the video card. Also. I wonder if there could be any benefit from putting a good line conditioner to all the components and see if that helps in anyway. Un clean power has the ability to make instruments act in unexpected ways. Just a thought. Good luck
My aorus x570 elite boards have been great, have bought three of them for different builds and have worked great. Use it with a focusrite 18i20 and works fine
I went through a few motherboards that wouldnt post before I figured out it was one of the wires from the power on/ reset /hdd light switch group of wires. I just left them unplugged except for power on. My guess is you have something similar wrong. You could buy a switch just to test. Or maybe test continuity but I'm no electrical expert either.
![[MV] A leap of faith...ความเชื่อใจครั้งสุดท้าย (From "Petrichor the series")](http://i.ytimg.com/vi/U1piZH2CNXA/mqdefault.jpg)
For clarity: I have tried swapping the power supply. I've also tried power supply in the case/motherboard out, and motherboard in the case/PSU out, and both out and both in. And tried 3 different power supplies, two from EVGA and one from Gigabyte. It's not the power supply. 😜🤓👍
Update #2: Please stop suggesting I flip my radiator/AIO around, especially when referencing Gamers Nexus. I watch GN too, and I'm following the ideal positioning that Steve from GN recommends in that very video many of you are referring to. There must be a misunderstanding in the community regarding his video, even though he made two. Tubes at the bottom is the ideal way to go for front mounted radiators provided the top of your radiator is above your pump. This way the air bubble will be at the top of the rad and thus won't get sucked into your tubes at the bottom or into your pump. This is what Gamers Nexus recommends. To suggest otherwise is a misunderstanding of his video. And I front mounted my rad because it gave better temps in this build than top mounting. Cheers!
I commented below, but I’m very curious if you have tried running your audio out from the front headphone jack? Had an issue with noise in my speakers on a previous build in an R5 Fractal case and that was the only way I could find to eliminate it. (After rebuilding my entire computer with the same case, the problem went away.)
I wondered if the screws attaching the power supply to the case were not in contact with the metal - eg the coating on the case was preventing a good ground. I feel like there would be a relay in the power supply and when power is "off" the ground is connected thru fine like you have found, when the power supply relay is on then the electrical ground has to go through the normal route. I would think that even mother board mounting screws might not be grounding too well either if the coating turns out to be the issue aswell I'd scrape away the coating in as many places as possible without destroying the look of the case
"I'm Gerald Undone and it wasn't the power supply."
It looks like the power supply might be the issue? 👀🥴
There is leak current. That is why you don't get continuity when on.
Also the meter can't meter resistance with voltage applied to it, and current leak will apply a voltage differential in the ground.
Also the meter introduces a very small voltage and current to meter the resistance and it might not be healthy to do that with the motherboard on.
Gerald, I'm an electrical engineer, and I can tell you it really does not mean I know any more than you do about this, as I focus on firmware design. However, having recently worked on troubleshooting some similar problems and since I work with other engineers who do this sort of engineering on a regular basis, I think I can help. One of the things we produce at work is audio equipment for medical environments. As you can imagine, they get pretty picky about safety and isolation. Ground leakage and ground loops can be a big issue, and often the best way to make sure you're always getting clean audio is to isolate the grounds, and even electrically isolate the audio if you can. Digital audio is good, since it can error correct, but in your case your USB cable was carrying a ground issue, so it wasn't the signal, but the ground that was the problem. The USB isolator will probably be a good answer regardless of the source of the problem.
As for the humming sound, you're correct in suspecting the LEDs and coil whine. Both of those can radiate or conduct into the ground in systems, and a good design will actually put some effort into doing things like making sure any RGB LED PWM frequencies are outside of audible range. For example, I PWM high power LEDs in my designs at 16KHz, and while technically not outside hearing range, you're probably not going to get that to conduct well. A 4KHz PWM on the other hand, could be more problematic. Coil whine is just going to be a problem with inexpensive electrical design on a board, and you'll have to do things like isolation to get rid of it. Not only can you hear it, but the sound comes from power electrically oscillating, causing parts to wiggle at that frequency. You can bet it's electrically propagating through the system somewhere, and you'll hear it in other places where electron shuffles can make it into some kind of audio transducer.
These problems often travel along the ground plane, since it's designed not to have any attenuation in it. If your ground plane has small, leaking frequencies on it, where it's varying by millivolts in audible frequency ranges, as soon as an amplifier uses that ground to produce power, you have now amplified that ripple on your ground and mixed it into your audio signal. Some systems will employ methods to try and filter the ground plane, but it's not a trivial problem to tackle. Again, this is where digital and optical isolation can help if it becomes problematic. If you connect your audio amps to your system with optical audio, and the speakers have their own power supplies, this should go away (unless it got onto the audio signal before it was converted into optical). This is harder to do for microphones, but often not hard for speakers. However, USB audio usually helps a lot, unless again, it's carrying this noise over on its ground and 'infecting' the amps with the noise. Also, there's a lot of opportunities for audio signals traveling on motherboards or other cables to pick up noise, so sometimes all this isolation won't help. The inside of a computer case is very electrically noisy. Every cable is a radiator of EM noise, and there's a lot of them inside a case.
All that being said, I'm not an expert in this, but these are problems I know we've had to deal with at work, and I've personally had to fight in my systems with audio. Since this isn't my area of expertise though, some of my explanations might be a little off. If so, I'm sure someone will be happy to vehemently correct me. And sorry this is a month late, I only recently found your channel and got an interest in video.
this comment deserves more love
Best useful comment ever!
This sounds solid to me. I used to work as a sound-man for a rock & roll touring band. I chased a ground-loop off a stage once by putting a carpeted riser under the head of our cable-snake. We were getting 60 Hz hum from the lighting dimmers that conducted through the metal stage-frame and made contact with the metal case of the snake-head. That was literally a "ground-lift". Cheers. = ]
Gerald Undone: I'm not an electrical engineer.
Me: I want to believe him...
Continuity testing rarely works well when a device is on. As current gets leaked through to ground, it causes small voltage differentials to appear in the chassis ground, which will confuse your meter.
Also, I’d generally recommend against continuity checking while things are on anyway, as this applies a small voltage to the probes, and this can confuse sensitive system components (to my knowledge the highly limited current released by the meter is unlikely to cause any actual damage, though, but I have by times seen it crash a system, or introduce glitches).
So basically continuity measurements only work in circuits without an applied voltage? I didn't know that! So the meter looks for exactly the voltage it sends out?
lightsout Yep. Only “reliably works” in inert circuits. And yep, it’s looking for a trickle of current of its own production, which can easily be fooled.
IANAEE, but i was wondering if this was the appropriate method too. Wouldn't the appropriate way to test for grounding issues be to set the multimeter to VDC and measure the voltage difference between ground points?
Nick in theory, yes. But while that might work while the machine is powered on, there are so many ground points where would you start? A lot of grounding issues are caused at the component level (missing or broken decoupling capacitors, broken shunt resistors, etc), and would be nearly impossible to find with this method, without a schematic, micro tipped leads, and a lotta time on your hands. Generally the best way with advanced micro- components like these is to just do what Gerald did and swap out components until you’ve whittled down the list of possible culprits.
(IAAEET? Is that the acronym? Lol)
@@WesPerry IANAEE - I am not an electrical engineer. Why anyone would expect you to know that is beyond me, or, WAWEYTKTIBM. ; )
Gerald, there may still be a "relative" ground issue when the unit is powered on. It is possible that the power supply has multiple V+ rails (most do). I suspect the power supply manufacturer is doing something funky with the grounding where maybe they are isolating each rail including ground so each becomes its own independent circuit. You may be experiencing the lack of a common ground between the circuits when powered on. You could verify that by testing the ground pin on each rail against the other rails to see if they are common when powered off but separate when powered on.
The PSU would also be one my first suspects, I don't think Gerald tried to swap the PSU did he? The case being a factor is also strange, maybe the way the PSU sits and/or cable runs through the case worsens the PSU problem (if it has a PSU problem)
JoaoRelego the FractalAudio (no affiliation with Fractal Design) thread I dug up says there is a major difference between Fractal Design R6 compared to most cases. Isolators. Rubber sound dampers separating some parts of chassi from others. Which FractalAudio says are bad for EMI.
However: Other threads says the total opposite, that you should isolate everything and never ground chassi, only ground through though PSU cables. Apparently there is a ground design war I wasn’t aware of before.
@@JoaoRelego I have the same case with a different build and have never had an issue like this. I am not a computer expert, but I was considering looking into a build like that in the case I currently have. Is this something I should be concerned over or are you thinking it may be the power supply?
2:51 there's your problem. Your PC is actually a Decepticon
Remember the good old times when Motherboards were green or yellow and you had red paper washers with your board that you put between the screws and Standoffs in order to Isolate the board? Yeah ... I do.
Maybe an issue there.
Continuity is there when off because there is a proper path. And as soon as you power on, somewhere along the lines of a screw or so it doesn't find it's the proper way home. Tolerances are tighter and the space where the screws and standoffs are, are tiny compared to "back then". So maybe worth checking.
For the coil whining and hissing:
What happens when you turn your RGB Memory OFF. So no lights? Will that help? I would suspect a high freuqency module to PWM the RAM and this causes issues.
Same for the GPU. You could probably check that if you "manually earth" these components. I.e. putting a crocodile clip from the GPU bracket to the PSU screw or something like that so it "has it's proper path" that you don't have when you power on the case.
The Motherboard of my wifes PC has some form of "metal sponge material" to it on the motherboards "shield" so it properly mates with the IO shield that you put in the Case before installing the board.
These Motherboards don't seem to have an IO shield anymore because it's all integrated in the big "block" now.
Hi my name is Gerald undone and I'm a Hooligan watching hoodlums on hulu in honolulu with who knows who.
I get interference with sound when using LED fixtures. I work for a company that does concerts. We have conventional lighting as well as led lighting. We have to make sure that the lighting is on a separate leg of phase in order to limit the interference with the sound or video. So generally when a venue has 3 phase, we lock in our power distribution panel, and we plug video, lighting and sound into separate phases of power.
can we all appreciate how he followed gamer's nexus advice on setting the radiator loop from the bottom
I’m trying to comprehend and maybe appreciate the fact that he really read that comment
That's wrong though, can ruin the pump. You always want the hose to the radiator higher than the pump so you don't pump air.
Great video, thanks. Around 17:30 you mention the lack of a "clear cmos" button on some MOBOs. A trick of the trade is to hook your reset button up to the "clear cmos" jumper. Sure, it has some drawbacks, but it is an option.
A flathead screwdriver works fine. You can also take out the cmos battery.
I was an electrician for 15 years. Normally for that intermittent grounding means something was wired incorrectly (switched neutral or even switched ground). But since the problem is only with the case, which has no wires...its almost like there's current running through the case, but that would point to a power supply problem...WAIT, IT’S probably the power cord or the plug on the back of the case. There might be a short there.
Speaking of PC’s... It would be awesome to have you make a video comparing monitors and their color accuracy for rec 709 and above. Haven’t found much content for PC’s about good reference editing monitors that aren’t $3k
I see LG OLED TVs being used by a lot of professionals. You'd need to buy a Blackmagicdesign Decklink or Ultrastudio to calibrate it, but I've heard that you can upload your own LUT into high-end LG OLED TV models.
yes,very useful
BenQ has been offering 2 good monitors for editing for a few years now. The PD2700U and SW271 monitors (other sizes are available) are under 1k dollars actually.
Many photographers and TH-camrs use this and it's not only for the price it fits in, but the other aspects too.
I hope this helps :)
@@neiltodankar8034 I personally use the PD 2500Q which is the 1440p version in 25" as you said specifically made for rec 709/srgb color gamuts and it's great, i didn't believe screens could look so different until i upgraded
The B550 Vision D has working Thunderbolt 3. I'm surprised you didn't mention that. That's gold within an AMD build.
Little tip:
Connect your RESET button to the CMOS clear jumpers.
It's really easy to reset that way, and if you wanna restart your pc, you can still use the power button.
Probably even more convenient than using the motherboard build in buttons (cause u won't have to open the case)
Ooooh, I like that one! Cheers! 🤓👍💜
Why did I never think about this?
I'm totally stealing this too. That's a great idea for lazy computer tinkering. Also, for mobos outside of cases, I've always just used a screwdriver as my power button. You just bridge the pins for the power button, which are usually nicely color-coded these days!
I have no idea what half the stuff you said meant, but I believe you, lol
lmaaaaaaooooo
I was looking for B550 onboard audio port benchmarks, this wasn't it, but it's still very much appreciated. As was the unintended review of the iFi product. Ran into similar issues my self with certain USB DAC/Amps and very sensitive in ear monitors years ago, ground noise going over the USB cable I couldn't solve without ripping the ground contact off the cable.
0:57 Audio interference is usually not mainboard, it's often the front panel of PC enclosure. If you're unlucky it connects Audio GND and USB GND together which is a big NO-NO. It defeats the audio star-ground on the mainboards and creates a ground loop. But it wouldn't matter if you use external interface exclusively and don't have any audio connection to PC. The ground routing can then cause the loop created between USB and audio to become an antenna which will collect all sorts of RF interference, predominantly from all the switching voltage regulators like on the GPU and elsewhere and from the PWM light control etc. To check: disconnect HD Front Audio connector from mainboard, the noise on rear audio out can disappear. You want there to not be continuity between front panel audio ground on the case and any USB ports on the case, but if you connect mainboard, the continuity will be measured through mainboard.
Longer ground loops are also possible. It's needed to consider all the equipment, not just the PC, but everything connected, all the connections made. I don't believe you showed all in the video, the interface was only mentioned in passing.
A few minutes in... the continuity between protective Earth pin of the PSU, the chassis, and all ground points must be continuous, the mainboard per se cannot defeat it. This is weird. Why it's not measuring like that... injected ground loop current? Wait a sec, what multimeter do you have, and does it measure continuity on the resistance range or on the diode range? I think it would be curious to switch all the multimeter modes and even try current and voltage. You want there to be zero current through the multimeter between ground points and you want zero voltage between these points.
Very important, do you have an extra PSU to test? Because i don't 100% trust yours at the moment. Please perform resistance measurement from middle pin (ground) on the IEC power socket to ground outputs on ATX connector and others. First you check how much is the resistance of your multimeter and multimeter leads by connecting them directly together, you expect to see something like 0.3 Ohm, and then you check the ground path resistance through the PSU, and you expect it to not be significantly higher, maybe by 0.1 Ohm at most. It's also vital that PSU enclosure screws are similarly well connected. Because most of the switching noise gets disposed of by shunting it to house earth, so earth impedance must necessarily be very low. And well it must be low all the way to the breaker panel in the house.
Yes, I think that's what's happening. The USB is acting like an antenna for the noise in the system due to the grounding issue.
I have tried another PSU, didn't make a difference. The only thing that resolves the grounding problem is switching cases.
The results are the same even if I measure from the ground pin on the power cable to the computer. Works when off, not when on. The case has to be doing something right? Why else would it work in a different case?
Gerald Undone The grounding issue could have also fried the USB ports that didn’t work, especially if you used powered hubs.
I really like that Gigabyte wants to work with you to figure out the issue. Also, I find myself humming and singing along to your intro all the time. lol
Hi, I'm Gerald Undone. I'm somewhere between rare and well done.
At around 13:05 I started to laugh.
I just laughed at how ridiculously relatable and lifelike this video is.
How the second you turn on your newly assembled PC and it runs into a bluescreen, or a hardware failure, or basically any of the endless possible issues that can happen to your build, and the itching feeling of utter annoyance but also nerdy interest and solid determination to solve the issue, and the testing, and the monitoring, and the reassembling, over and over again...
Is it a bad motherboard? Did I install my drivers incorrectly? Is it a faulty memory block? A small screw short-circuiting the electrical loop? And when you thought you have finally figured out what the problem was, it turns out it is in fact multiple problems, weirdly interacting and aggravating one another. Aaaand here we go again.
Painful bliss.
How fortunate we are to have these kinds of problems.
It can be super annoying to do such stuff when it's your only PC and you gotta render and edit videos... Not to mention the stress, DAMN!
I fancy myself a technical guy, but every time I watch Gerald's videos I just go and stare at the middle distance and question everything I've done.
Hey Gerald. I had that same sound problem in my speakers when I connected an external GPU to my Macbook Pro . The way I solved it is by putting an adapter to the electrical cord with only two pins, the ones that do not have the ground pin. That solved it right away. Seemed that there was some ground coming back thru the egpu ground pin and entering the computer thru the thunderbolt cable. Great video.
Potential electrical explanation for the high pitch whine noise:
I have the same speakers and the same issue and the problem comes when using simple instrument audio jacks (mono) for the monitor speakers. They only have 2 wires and short out the shielding (ground) and the return line. Now you could either use a proper audio jack or tape of the ground pin on both speakers power cable!
Ah, an Arctic Liquid Freezer II , I see you're a man of culture as well.
If nothing else, it’s a great exercise in troubleshooting! The tough part is when you have multiple issues when you think it’s just one.
I probably wouldn’t have watched this but I like your channel and I recently updated my desktop system with an Aorus X570 Elite purchased for $209 US, which I thought was a great deal. I don’t do video, but I enjoy watching you get Undone!
I had a similar problem awhile back... It drove me crazy. However, since it was months ago, I don't remember exactly how I solved that problem. But I do believe that I can direct you in the right direction. I would start with the wires. Particularly any audio cables, from the speaker, the mic, and audio interface. That sound is digital/analog noise. It can come in the form of crosstalk between touching wires... a bad insulated wire. A speaker wire that is exposed slightly from the speaker negative/positive connection. If I had to swap out hardware, I would swap out the audio interface. I would also swap out audio wires... Also It could be the microphone and their magnetic fields, relative to the speakers and their magnetic fields. Try and update the drivers from the audio interface too. I strongly believe it was a wire though...
You have discover, a new semiconductor! Yours case is non conductive when power is applyaed.
You can turn the motherboard on outside the case using a screwdriver on the power jumper on the motherboard.
Back in my day we solved our nasty coil-whine problems with a blob of hot-glue. Of course the coils were a cm square. I modified a stethoscope with a broken dental-pick (just jammed the tool up the tube in place of the doctor-disc). It was useful for isolating mechanical/servo problems related to friction & resonance (or vibration) and it could be used on non-conductives like hard silicone. You want to beware of resonant frequencies; they exist in objects as well as circuits. Lots of shielded heat-sinks helps on the MB.
When a full spectrum sound-wave (eg: pink noise) meets an object with a resonant frequency the reflections will amplify just that particular band four times, leaving every other frequency unaffected. In a feedback loop of your amplified room that will cause destructive feedback, (ringing, squealing or thrumming.)
I can't help with the technical issue but I can say, the X570 MB paired with the AMD 3950X processor just rocks. It just chews through 4K video editing. Great work.
Most random thing. I watch your videos for all the other reasons and here you go and test a MB that I swapped to (from Taichi to Master [570]) TODAY! All I can ask/suggest is that you try to disable the RBG altogether of the ram, if it's not the RGB in itself maybe disable XMP (or enable, whichever you didn't do)?
Keep it up! One of the best, if not the best cinema-tech channel out here! :)
For the issue with your RAM - it might be the LED's in it:
I had internet problems regularly after I renovated my office. A month ago I had enough and called a technician to have a look. He looked around for about an hour, checking all connectors and swapped all old cables for uninsulated ones, but nothing changed. Then we turned the power off to check if it's a unisulated cable in the wall somewhere and suddenly the problems were gone. When then measuring the current on my outputs we got no error for the power cables, so we resorted to go piece by piece. Finally, when we were turning everything on again, step by step, we saw a sudde drop in speed and multiple interferences causing jitter when I turned on my LED desk light which was standing right beside the router. I swapped the LED bulb and still got the same issue. Also moving the lamp to the other side of the desk didn't help. Since the lamp is gone now and replaced for a regular bulb my issues are gone aswell.
Now I don't know how this correlates to your coil-whine issue, but I assume it could be similar. It could either be that LEDs in general can cause interference in your system (especially since they're so close) or that the insulation around the diodes on your RAM is faulty and I as well had bad luck with two janky lightbulbs.
Wow - some great effort in troubleshooting there. I come from many years playing in touring bands - a ground lift was in almost every musicians bag for basic but similar problems. The process was plug something in and when the amp is turned up, if there is hum or noise like the what you were getting, then pull out one at time anything that has a seperate ground til you find the culprit and the reverse polarity or lift the ground. Plug it back in and in most cases that would fix it. And the same gear could actually be fine some places and not other places. So troubleshooting that - the venue or arena power would make a difference. But kudos to you for being so vigilant - while you were going through the whole thing here, I was thinking that I probably would have just plugged in the USB ground loop isolator and called it a day. A little out of my tech interest here, but still kind of interesting.
Hey Gerald, just chipping in - albeit late, but I think it is worth it since these videos of yours will be great reference for others. I had the same problems with a different system in a Fractal Design Define R5, connecting to an external DAC and the DAC to my studio monitors. Changing my DAC to an XLR one and connecting DAC to monitors with XLR improved it, but I had to get a USB isolator to remove the ground loops completely. This was happening both over optical as well as USB. All I want to say is that it is not necessary to spend 80 bucks on the iFi defender, since designs based on the same chip are all over the place and cost betwewen 15 and 45 bucks depending on whether you want USB 2.0 or 3.0. With 2.0 it will be enough for 24/96 audio. If you go for this route one only needs to look for ADUM 3160 chip based isolators for USB 2.0, and ADUM 4160 based designs for USB 3.0. They all work. If someone wants to take it further (and spend that money saved on the defender elsewhere), 50 bucks will get you a linear USB power supply with a 15w toroidal transformer inside, which you can use to power the isolator for an even cleaner connection (probably not necessary though). I am starting to think that the Fractal cases are the culprit, but I can't be 100% sure, since these ground loop issues are a bit "ethereal", as I'm sure you already know.
iFi sponsoring the content lol. Thanks a lot for all the info, helps a lot with deciding what mobo to get
I built a 3900x system when it was launched.
I almost bought that case, but wound up using my old one, I'm glad you pointed out the case issue.
I have the x570 Master, it's been great, unfortunately, 1.0 doesn't have Thunderbolt like the 1.1 and later does. I wish they'd make a proper add-in card for it.
Enjoy.
I'm not an electrical engineer by training, but I have designed a number of low voltage switching power supplies for LED and wireless applications, and consulted with a number of actual electrical engineers during that process. My understanding of the noise problem is that the frequency of the noise you hear can help you diagnose the problem. If you're hearing 60hz low-frequency noise, that's ground loop hum coming from the AC power supply (in some applications this is also testable by switching between battery and wall power, but if the device is inherently wall powered, you wouldn't be able to do that test). Higher frequency noise would be more related to switching power regulators, which are used to convert AC to DC power, and also to change the voltage levels of DC power (e.g. 12V to 5V, 12V to 3.3V, or even boost from 5V to 12V). The frequency of that noise can be used to determine which regulator is causing noise, if you know the specs to the regulator.
Immediately pinpointing a specific regulator on a complex device like a PC is probably not possible, since there are a lot of switching regulators. But one possible hint in the diagnostic process is that the frequency and magnitude of noise will change as the power draw on the regulator changes. The problems you found with the LED RAM and the GPU under load both seem to point to regulator behavior where under low load, the switching frequency is outside the range of sound, but under increased load, the frequency falls within the range of human hearing. (It's actually a little more complicated than this because regulator designers know that this noise is undesirable and can mitigate the noise using filtering capacitors. However, those capacitors are selective in the frequencies that they are effective in filtering, so one may find that increasing or decreasing the load by some amount might increase noise, then further changes will decrease noise as the frequencies shift into and out of the effect range of the capacitors being used within the design.)
This noise is propagated on the voltage and ground lines, so it makes sense that even though you don't have a ground loop, powering your audio device with a separate isolated power supply would result in the noise not reaching the device.
Also, perhaps I got too caught up in the diagnosis part of this, and lost focus of a solution. Actually trying to balance out the power draw of all your components to evade regulator noise at audible frequencies is overall, probably too hard. Probably a more stable solution, mentioned by a number of other commenters, is to ensure that the audio equipment that you are plugging into the USB ports have their audio handling power paths isolated from the data paths, which unfortunately means getting away from bus-powered devices.
Some bus-powered audio devices may attempt to mitigate this with device side power filtering but it's hard, but probably the most effective solution is something like an RME Fireface UFX series device, which is independently powered.
The difficulty here is that even if you manage to balance out the noise from all of the internal components, other devices you plug into the USB ports might be noisy, or simply change the power draw to the point where you enter the noise range again. Keeping audio device power segregated insulates the setup from that kind of instability.
Oh, I know how this feels. I had a similar ground loop issue, I too went through the gamut and gutted the entire system, isolating everything to the bare minimum components in attempt to isolate, removing the motherboard from the case, testing ground with a multimeter on my receptacles, etc. Ultimately I RMA'd my motherboard with Gigabyte as I had a faulty sata port on it anyway, so thought I would kill two birds with one stone. The ground loop only presented itself when using my RME audio interface as a USB Dac, and connecting a pair of its outputs to RCAs on a headphone amp. Extremely annoying and pulled my hear out as I invested a ton of time without a clear resolution. Ultimately I moved away from custom PC builds as it was a Hackintosh and it just killed way too much of my time troubleshooting over the period I had it (without being 100% satisfied with the user experience)...I'm now running full OEM simply to save time.
Meanwhile my PC case has been stuffed with so many Hard-drives that I cannot put on the side panels anymore. I got an SSD taped to the roof of my case even. The Jankiness would give Gerald a heart attack.
Check for voltage on the case. To do this, carefully measure between the pins on one of your extra four pin drive power plugs and the case. See what you get. Also, replace the power switch and see what happens. You can then unplug each wire going to a case extension (leds, switches, fans, usbs, etc. add them back one at a time and see if the issue appears with one.
One other thing you can try is a power conditioner or UPS. A lot of them are made to clear up power signals before they get to your system. Furman makes a bunch but there are also plenty of UPS (uninterruptible power supplies) you can get online that condition power and help provide stable perfect voltage to all of your computer components. Coil whine and other leaky components in a computer all contribute to EMI and electrical noise which is why designing a good motherboard is super difficult for electrical engineers. Isolating everything on different lines in the PCB and on different power rails isn't easy. Typically I don't recommend using front panel IO at least for audio. Try disconnecting that cable all together as well as UPS and using an external Audio interface if you aren't already. In some cases the only way to fix the grounding/signaling issue is with one of those nifty USB devices you already got. This is why most audio pros recommend against internal audio solutions no matter how well the Motherboard is designed. You simply get less EMI and interference by moving the audio components into their own external enclosure a few feet away from the CPU, GPU and Power supply in the PC.
I would say it's the power switch panel on the front of the case. When off, the switches on the front panel make no contact.... thus not interrupting the ground. But when engaged (powered on) it's shorting out somewhere, causing a ground issue. That, or somewhere there is a spot rubbing/touching the case doing effectively the same thing. For your next vid: :Tell me where the big bad motherboard grounded you..."
That seems reasonable, but I just checked and the problem persists even without the front switches connected. I powered up using the X570 on-board switch. So it must be somewhere in the case construction.
Power switch in pc doesn't work like a light switch. It only connects two connectors when fully pressed.
@@geraldundone Faulty fan harness shorting to case?
much much respect for the thoroughness and your willingness to investigate even intermittent problems. working in the software QA field I can kind of relate how pesky that can be :P
I experienced a very similar problem a few weeks ago connecting a USB powered microphone to a laptop via a Blackmagic ATEM Mini. The USB mic works fine when connected to the laptop but if you then use the 3.5mm audio cable from the mic to the ATEM Mini and then of course USB-C from the ATEM to the laptop you get the same pulsing electrical interference. I suspected that it was an earthing issue of some form but your video has (I think) given me the final clue to the issue and probably explains how your issue is related to the case.
* I think the issue is not a failure to provide a ground but that the ground is actually receiving current. In effect I think that something is actively grounding (shorting) to the case and that is what the mic and sound card are picking up.
* This also explains how it is related to the case in that if the case mounting is actually touching something it should not then your ground would be live.
* The USB grounding gadget would resolve this probably because the current is so small that the ground lead just provides a lower resistance path and thus neutralizes the voltage. (Voltage is actually Potential Difference after all)
* I think that your multimeter test is failing when power goes on because you are no longer testing a dead circuit where the meter's current is the only current being measured.
* The link to the RGB ram could just point you towards what is shorting down to ground.
All just guess work but hope it helps.
Gerald - great video but I think you’ve approached the problem in the most difficult way. Computers are nightmarish when it comes to controlling noise and interference leaking from the motherboard and its hardware into peripherals and audio interfaces. You’re better off using properly designed interfaces that are able to deal with the noise - or galvanically isolate the connection between the computer and the audio interface. One of the easiest ways to protect from usb noise is to use optical out of a computer - however this may not suit your needs. The iDefender ground loop isolator is a good solution - but a well designed audio interface should do this in the first place. The Zoom F6 is subject to significant low frequency noise from its poor power filtering (see testing on AudioScienceReview). Might be worth investigating the sound devices gear which I think is better engineered.
I know this video is a few years old now but I've been having the exact same issues with gigabyte motherboards. After talking to a friend who IS an electrical engineer and trying a bunch of different things, (cheap ground loop isolators, different plug configurations, different computer component configs including cheap vs expensive PSU's) it turned out that most likely it is because of how the motherboard is attached to the case. This is consistent with how you experienced this problem in some cases more than others, and didn't experience it when the system was outside a case. The system is successfully grounded via the PSU, however the motherboard also has a special (extra?) grounding design where it grounds via the standoff screws (consistent with your definition of ground loop being caused by multiple grounds).
By putting small plastic washers between the screws and the case, I believe this will solve your final persisting issue by making sure the system grounds purely via the PSU. This would avoid having to swap out different components, use that USB isolator etc.
This is just my hypothesis however, based on your tests (thank you!!), my tests, and discussions with my audio engineer friend. I've been putting off trying it as it would involve rebuilding my very tedious compact watercooled ITX build but I can report back to you with my results if you're interested.
I had the exact same issue! I would notice the issue when scrubbing through media in premiere/when the system was under heavy load. Unlike you, I got this through the front headphone jack, and interestingly enough, we have the exact same case! Fractal R6.
Other than that my specs are completely different, Z270 board, I7 7700K, RX 5700XT, 3200Mhz Corsair Vengence (no RGB). I fixed the issue by buying a Fiio K3 DAC, as I figured it was a grounding issue in the front IO panel.
Super interesting hearing your side of this.
It has to be said.
PC Master Race. Woof!
Gerald, I am an EE. I'm not going to explain, but give a couple of hints. 1. get rid of the Yamaha monitors. They have no shielding, electrical or otherwise. I switched to Adam. Next use an oscilloscope not a VOM. When the PS comes on it puts a potential on the ground that is higher than the small battery in the VOM. thereby the VOM will not see a current draw from a short and will not sound the short. I had the same issues as you, the real culprit is the USB design from Intel. Note that some motherboards are shipping with a special USB that has a separate filter on it. The defender is a little over priced but like a cloud lifter it has no competition so there you are....
Thanks for the detailed review, I have a similar issue with my B550 Aorus Pro AC. The issue is my keyboard both wireless Sculpt keyboard or my wired keyboard goes erratic if attached to USB 2 port but works fine if attached to USB 3 port. The mouse and every other peripheral including Logitech camera, external USB drive and few other work fine. You are right when you mention about the PCI 4 in B550, I do have a nvme attached to the top slot running full PCI4. Maybe a future BIOS should fix this.
This is the big problem with these boards. Do you have the ethernet issue?
@@karama300video for Windows I use the driver supplied at the Gigabyte site and so far no issues but I find the internet slows down over time and it needs a reboot after that. Probably a windows issue as the ethernet runs fine on Arch linux here with the latest kernel.
@Haimura the issue is from AMD and not Gigabyte and newer bios with latest aegesa have fixed issues to a great extent.
@Haimura thats their top of the line board, I have the aorus pro AC wifi version and so far everything is running fine now with BIOS update. The issues that you mention is suffered by other manufacturers as well using B550 chip. Make sure to update BIOS asap after purchase, use the beta bios.
@Haimura just keep a track of the latest BIOS and also make sure to match RAM correctly and you will be fine.
Lol I love this channel! 70% of the content goes right over my head, but I feel smarter having watched them. 😜
Motherboards!?? If Gerald talk about them it‘s interessting. Scary
A strange but anecdotally common reason for coil whine is the power source, ie. the quality of the electrical power in the building. Again, it may sound strange but it's worth checking out if the coil whine still happens in another building. Power supplies / different components may be more resistant to coil whine so switching those may help even if the core issue is the quality of the power from the wall, leaving the impression that it was just a bad component. Also, if you're using a UPS try it without the UPS.
7:09 this is an expensive contrast card.
Here's the best thing about this video. This is the video I expected from Gamers Nexus. I always considered Gerald Undone to be the Gamers Nexus of video, audio tech and Gamers Nexus to be the Gerald Undone of PC tech.
I haven't had any issues with either the B550 Gigabyte Aorus Pro or Gaming X but I did put it into multiple legacy cases from an era where airflow really was a thing. From my experiences and my understanding the majority of the grounding is done from the PSU cable to ground which would explain why turning on the PC changes your ground readings. The grounding not accomplished by the PSU would normally be such a low variance and go without any issues minus what would be audible buzzes or static. The Aorus Master has more high power phases and maybe surpassing that variance in some pc cases resulting in the issues you laid out. Just a guess really. Great vid and support from Gigabyte.
Thanks Gerald. Have a good one.
Thanks, Jay! That's a huge compliment. I love Gamers Nexus. Appreciate your thoughts too. Cheers!
I'm not an Electrical Engineer either, but my guess would be the the RAM (or it's RGB controller) and the graphics card are emitting radio frequency noise and the case, due to it's sub-optimal grounding, is picking up the RF noise like an antenna and leaking it into the ground circuit.
The USB defender could be interrupting the ground connection to the PC and grounding directly to the mains thus blocking noise from the computer's ground. Or it could be grounding the entire case via the USB and motherboard, and thus resolving the case grounding issue.
When your first video posted I just finished installed the B550 Aorus Master in my system. I didn't have any ground loop issues so I assumed it was something else in your system but forgot to post. I did have strange USB issues like you though, so I ended up returning that board for the X570 Aorus Master which is a great motherboard.
Next intro: I'm Gerald undone, and the R5 is fire
*on fire ;)
I think that powering up might generate eddy-currents in the case. The whine may be switch-mode power supply related. The radiated magnetic field is affecting the audio circuits by leaking into those circuits and being amplified or there is insufficient filtration of the switch-mode power supply lines and the power supply frequency is audible in the audio circuits. The USB filter is filtering out the power supply noise.
I started watching this video in the TH-cam preview panel on my phone, and the captions try to translate what you are saying. It thought you said "Hi, I'm jailed Undone" and that stopped to make me laugh. You should say that, Hi, I'm Jailed Undone... as always, awesome video... you make every other youtuber look devastatingly terrible. And you make me personally look like an idiot. But I'm happy to learn.
The only reason the case may be causing any grounding issues could be the front panel header. That’s probably the only place where other (non grounded) electrical connections can make contact with the case.
Hello good people, I am Dimitry with Hardware Canucks and today we have...
Oh, wrong channel.
Hey, checkout the Silverstone CP14-E adapter (like $20-30 on Amazon), if you have an extra USB3.1/3.1 header you can use that to convert to USB-C header. I had the same issue, my Asus Pro WS X570-ACE didn't have internal USB-C header, I used that adapter which worked great! No USB-PD so no fast charging of phones and such.
I had a ground loop on between my desktop pc and my speakers that I resolved due to this vid.. thanks :)
Very thorough. I found it educational and entertaining.
Sorry that I can offer no good suggestions on the continuity issue, in particular. I even learned stuff about continuity testing on a motherboard. :)
Computer Conspiracies - “All I want is the perfect computer~” 🤣
Gerald, you and I have similar minds haha. Quite happy to see you use a Fractal case just like I am. I'm using the R5 though and have had no issues.
Also, I think just turn off RGB.... on everything. I'm not sure if the leak issue is just with Gigabyte stuff for this time's lineup. I'm using G SKILL ram, no rgb. For GPU, I installed the software just to turn RGB off. I'm using an Asus Z370 Prime A mobo... perhaps it's their lineup of mobos that are causing the issue?
Also, I always have my pc plugged to some power bar... I only use Tripp-Lite and APC.
You should had a case as a control in the experiment. This would remove the possibility that the standoffs in the (Xi-?) case had a different impedance to ground to measure. Speaking of which, did you happen to note the impedance to the case of the PSU to say the top of a MB mounting screw?... Then measure Ohms / Impedance between the standoff and the PSU mounting screw. Those measurements should be very close. The USB adapter probably has an optoisolator/optocoupler hence the cost. Once the ground loop is created you are more susceptible to EMF from the oscillators in the case, as it acts like one winding inducing all kinds of switching, oscillators ( The RAM for instance, or video card )
next intro: Hey guys, this is Austin.
I can appreciate your video. I've been building PCs for many years and don't ever recall having so many issues at one time. However I do have an issue with one of my systems where I get computer noise coming through my USB audio interface. I've tried switching interfaces and a host of other things but it's still there. I'll have to give the USB ground-loop device a try.
BTW - x570 boards as SOOO much better than the B550.
I once spend 3 days trouble shooting a computer that wouldn't post only to find out it was a grounding issue with the case. Outside of specific complicated mobo problems, case grounding is the most enigmatic problem causer of all.
Holy crap I have the same ground loop issue, and I'm so excited to watch the rest of this now O.O
LED's dim by switching on and off very fast, so your theory with the RGB being the cause of the noise is actually very good! I dom't have proof though.
Gerald Undone: I'm not an electrical engineer
Me as an actual EE: I think he is hiding something...
Wow! I had that problem with my old B350 Gigabtye motherboard and Ryzen 1700X using the DAC USB (using M-Audio SuperDAC). I also had Corsair LED RAM in it. I tried different USB ports and still that high pitched sound. Tried reversal polarity on the powered M Audio BX 5's and the sound was still there.
But through the 3.5mm headphone jack, no problem.
I've upgraded to a X570 Gigabtye Elite and 3990X with RGB Corsair RAM and just a slight hint of that high pitched sound. Playing anything masks it completely.
Hi Gerald, very nice and informative content, enjoyable as always! There is other solution on this case, just use PCIe soundcard (Creative Sound Blasters, EVGA NU Audio, ASUS Essence STX II or even ext. Pro USB soundcard/Audio Interface: Focusrite, Steinberg, Motu, M-Audio, Apogee, etc.) on problematic motherboard (with audio interference/ground loop/) like on AORUS B550 Series, and even on the GPU coil whine problem, it does great job of solving noise issues and get better quality on audio playback and recording.
Finished watching and I was partially correct about the case.
It’s what happens while waiting for Kasey’s latest Canon R5/R6 video to post. ;)
pro tip, you can boot a mobo outside a case without the case power button by just touching a screw driver to the two pins that are for power button
That makes perfect sense. Thanks for the tip. I don't know why I never thought to do that. Like I said, I was already jumping the CLR CMOS pins. 🤓
great video, i have that same sound with my blue yeti, i was looking into that ground loop. for my next system build next month for my AMD 4950x with 64gb i will be using the 570 master and memory without RGB. system will be nvme only
Gerald you should get into more tech reviews. With your experience in colors and cameras, you will be a top tier mobile reviewer
Big Motherboard! Would love to have this Computer so I don’t have to edit my RED 8K Footage on my phone 📱
Gerald- I don't know if this is too late to the game, but I would pull the power switch from the equation. Leave everything in the PC and try powering up the PC with a separate switch not connected to the case, or just shorting the terminals on the motherboard. My guess is there's something up with the design of the little switch mechanism itself, or maybe there's something poking out of the switch box when in the off position that goes back in when the button latches in the on position.
I would take all of the case wiring and diagnose it for issues. Could be the USB ports or sound causing some issues. I’ve had cases that just never worked right or would crash if I used the case ports to connect peripherals or external drives. Many times it could be a poorly seated wire on the case button area that came loose was poorly seated during manufacturing or got bumped loose during installation.
Nice investigation. I've been having problems with my X570 AORUS Ultra's USB ports as well. No blue screen, but depending on what I have plugged into different USB ports, certain ports on will seem to turn off and then turn on again. I notice this most with my Bluetooth mouse/keyboard combo. I'll be watching a Gerald Undone video, say, and I'll try to pause it to go back and relisten to a very complex point you make, and I won't be able to click on the video for a few seconds. Then, I'll get a notice that a device (the mouse) has been plugged into USB and I can use it again. So annoying. After watching your video, I'm thinking I got a problematic board.
Next Intro: I’m Gerald Undone and I toss coins to the witcher!
Thank you for the video it was very informative. Right now I am get a ticking noise is my motherboard when it is under a load and I am in the process of figuring that out. I used to have the ticking noise even in idle, but since I figured out that my Corsair PSU fan was not set to on it only happens when I play a game.
What is really annoying is when playing AC Valhalla everything is fine for an hour then I get severe glitchy movement in the game.
Hello, coming to this party half year late, but maybe I can attempt to address the ground issue, as I work in similar field. Power return is not the same as ground. Some signals (analog, such as audio) use ground for its reference. But power equipment like fan and lights should not use ground as a return path for its power, because PWM (I assume most people here know pulse width modulation) will introduce noise to the power bus. But if say your RAM (which does not have a dedicated power return for RGB LED?) uses the RAM's ground for power return, its RGB PWM will introduce switching noises to the ground, which is used for audio. The fact that removing the RGB-version of RAM solves the audio problem supports my explanation. Also, the ground check with your digital multimeter, when the computer is turned on, the RGB RAM may have cyclicly messed up the ground hence messed up the multimeter reading. I hope this helps explaining the puzzle somewhat. One of the reason I prefer using optical audio device is that it completely decouples the audio signal from the computer's electrical ground and any issues it may cause. I also design and build audio equipment on the side, and I have found that if a cheap power supply is used, it will introduce high frequency whine to the audio as well. And I'm talking about the speaker's power supply, not the computer power supply. However, I assume your speakers have decent power supply, and thus not the cause of your particular problem.
I'm rockin' the Aorus X570 ITX and it's been solid, but of course you would run into some crazy issues on your build!
On another note you can turn on motherboards that don't have a built in button by shorting the 2 power pins in the front panel io breakout.
Nice tip! Thanks. :)
Hey Gerald, friendly tip! If you can its better to orient your AIO Radiator with the tubes on the top, this way it guarantees there arent any air bubbles reaching the pump and making it perform noisier.
It's actually the opposite. If you're referring to the Gamers Nexus video, he recommends tubes at the bottom. This way the air goes to the top of the radiator, which should be above the pump and thus the air will go to the top of the unused portion of the radiator preventing air from being pulled into the tubes at the bottom and into the pump. I'm worried about this misunderstanding--you're the 4th or 5th person who has commented on this very video with that tip and the tip is just incorrect. I thought Steve's video was pretty clear, but perhaps there is a larger misunderstanding in the community regarding this. Here's the timestamped link: th-cam.com/video/BbGomv195sk/w-d-xo.html
Hope that helps. Cheers!
@@geraldundone good to know! that was the exact video i was referring to lol
USB headers was a problem in one of my cases. Try unplugging them and the power switches. If those ain't the problem... the power supply is.
Hey Google, what are the issue with my motherboard.
Google: Ask Gerald Undone
Not an engineer, but here is my best guess. If you have a meter set to test continuity, your are measuring resistance, and low/no resistance is equated as continuity by the meter (and you get a beep). If however, your meter leads are place in two locations that have a "difference of potential" (aka voltage), then your resistance measurement is disrupted in the meter and not seen as continuity. At least that's how it works with my meter. That being the case, I would guess the case is having voltage bleed to it after it is powered up. switch the meter to measure voltage and place the two leads in the same location and see if there is a voltage when the pc is on, if so I would start by checking case lighting.
i believe the leds introduce weird interference, i can hear hissing & humming on my led panel. Its more interesting because there are no moving parts. I guess its the coil whine of step up converters for the leds.
intro got me subscribedbed
I have the x570 in my editing rig. I really like it. Still haven’t filled the 3rd m.2 in there. When you connect headphones to the front panel and set it to 2nd audio output it has a 3step amplifier which comes in handy for my 80ohm beyerdynamic headphones.
I set the chipset fan to silent in the bios and the temps don’t get anywhere close for it to even turn on.
Coil whine is a mark of a cheap component. You could swap the video card. Also. I wonder if there could be any benefit from putting a good line conditioner to all the components and see if that helps in anyway. Un clean power has the ability to make instruments act in unexpected ways. Just a thought. Good luck
A week ago I purches this mother board. And this week my tech guy finish to build my pc.
nice to see that you picked up a propper AIO ;)
My aorus x570 elite boards have been great, have bought three of them for different builds and have worked great.
Use it with a focusrite 18i20 and works fine
But saying that I did have a problem with a usb lav mic having very similar noise on my old 1150 gigabyte motherboard, strange!
I went through a few motherboards that wouldnt post before I figured out it was one of the wires from the power on/ reset /hdd light switch group of wires. I just left them unplugged except for power on. My guess is you have something similar wrong. You could buy a switch just to test. Or maybe test continuity but I'm no electrical expert either.