Is there any possibility to set uneven speed for fans? At some point of speed they start to slightly resonate which cause unnecessary increase of noise. If it goes beyond this speed it is back again more quiet.
I would rather want to avoid stuff like Low Noise Adapters so software/bios control will be better (if possible)
Hello, PetanoCZ. Thank you very much for contacting the Intel Communities Team. I will be more than happy to assist you.
Indeed, from the Intel® Visual BIOS, you can manage the fan speed. To access to it, please press F2 while the computer is booting and then, go to Advanced > Cooling > in the Fan Control Mode drop-down you can select between Balanced, Cool and Quiet. Cool will be the mode where the fan will spin faster.
Thanks for response to my query.
I am aware about BIOS fan settings and playing with it. I did more tests and it seems resonance is appearing roughly between 1300 - 1500 rpms resulting in whir noise. Before it is basically silent and above is it just whiz noise from moving air.
I believe setting one fan spin roughly 200 rpms slower should fix issue. Or am I only one experiencing it?
Thank you for your reply, PetanoCZ.
In this case, please make sure you have the 0037 BIOS version installed. If not, click https://downloadcenter.intel.com/downloads/eula/27747/BIOS-Update-HNKBLi70-86A-?httpDown=https://dow... here to download the .bio file of this version and, follow this guide to complete the installation: https://www.intel.com/content/www/us/en/support/articles/000005850/mini-pcs.html https://www.intel.com/content/www/us/en/support/articles/000005850/mini-pcs.html
Share with us the results.
Thanks for info. I used Intel Driver Assistant during installation and it went also through BIOS update. I have latest BIOS version 0037.
Actual difference in fan speed is ~40 rpm which might be more reading error. Do you know if fans can be controlled independently or they share same fan controller?
EDIT: It is not serious issue. Some people may not even notice. I am probably oversensitive because I spent a lot of effort to make my SFF build cool&quiet so I may be oversensitive. I am just curious.
Screen shot (Prime 95 + Unigine Heaven)
Thank you very much for your reply, PetanoCZ.
In this case, the only way to manage the fans will be in the BIOS. The BIOS will actually control the two together.
I'm also hearing this kind of high-pitched "chirping" noise, tried to capture it in a short video: https://www.youtube.com/watch?v=3WadeUIUzz4&feature=youtu.be hades canyon nuc8i7HVK - YouTube.
It's very noticeable ~14-18 seconds. Was that that the effect you were also hearing? I'm not sure it's related to the fan-rotation speed -
I can also hear it when the fans themselves are near-silent.
note: I have bios 0037 installed.
There are a couple of possible explanations for the resonance that you are experiencing.
Worst Case Possible
One or both of the fans in the HVK, have bad bearings or a failing/misaligned fan-blade (one or more).
Tests (While the unit is running and while you hear the resonance)
If the resonance doesn't change at all during any of these tests, sending it back for repairs is the only solution at this point.
If the resonance changes at all during those tests, it is unlikely you have a bad bearing.
I recommend you consider something loose in your NUC, or the case screws need tightening, or you're placing the NUC on an uneven surface creating a vibration.
Something is creating that resonance. Would you please describe what the NUC usually rests on?
Thanks for your suggestions, giving it a try:
# 1 So, when hearing the sound when it's standing on the desk
... and picking it up - sound is still there.
... and holding it up-side down / sideways - sound still there
... squeezing top/bottom together more tightly - sound still there.
Nuc is standing on a small wooden "rack" which puts it 10cm above wooden table; tried putting a rubber iphone hull below it to dampen more, but no change. Crazily enough it feels like I can somewhat influence how it sounds by sth trivial as clicking refresh in a browser. I'll experiment a bit more, see if I can capture that on video. Thinking of also trying to boot the nuc without nvme/memory - to be 100% sure which of the 3 is causing it.
I've thought about it further since first posting, and another two steps would be a compressed air cleaning (not suggesting that you haven't - just following a protocol), and if it doesn't produce a change, a complete disassembly deep cleaning (where you can also isolate which fan(s) are creating the resonance).
I do the same them as you, with blocks underneath the NUC8i7HVK, just to ensure max air-flow.
The fact that when you pick it up the resonance remains, indicates it likely has nothing to do with where you typically have the NUC8i7HVK sitting or unevenness. But rather, more-than-likely coming from within the unit.
If you're have experience building PCs, my recommendation is after a canned air cleaning (if needed), to disassemble the unit and do a deep clean. This process will also give you the opportunity to tighten up any loose screws and determine if you have a faulty unit.
There is a great video on https://www.youtube.com/watch?v=-fiePyi6Xu0 youtube that takes you through that process.
Else, you could return the unit to where you purchased, if within purchase return and if beyond, within manufactures warranty, if you purchased a plan.
You may also have some luck getting a unit sent to you before you need to return the resonating one. It doesn't hurt to ask.
Good Luck and let us know!
I ran the same test and also see a 44 average RPM difference, representing just over 6%. The difference can easily set up a resonance. The audible (normal hearing) resonances based on the enclosure (depth, width, and height) are:
Generally, the resonance on these enclosures will be around the note A and C on the piano.
Go to Youtube and look for "Reference Frequencies" and select A and C, and play them. Are either of the notes close to the resonance you are hearing?
Other resonances can occur with the Mobo too, but less likely.
I also have resonance issues with my NUC8i7HVK.
To determine whether it's due to the difference in fan speed between left and right fans, you could do the following.
1. Increase fan RPM until the resonance becomes an issue, and note the two fan RPM at this point.
2. Go online to a tap tempo website and tap along to the frequency of the resonant peaks.
3. Increase fan speed by increments (across the range for which resonance is an issue), and repeat steps 1 and 2.
If the BPM of the resonance is equal to (or close to) the difference in RPM of fans, the difference in fan speed is likely to be the cause of the resonance. I plan to conduct this test myself this evening. Will report back.
Short update from my side, here's another video where I tried to capture this modem-like high-pitched noise: https://www.youtube.com/watch?v=7nh4xk-AGUY&feature=youtu.be hades canyon nuc8i7HVK noise - YouTube
To get rid of it, I tried several things, like
- directly connecting to a powerplug (instead of through a prolongation-cord) - no change
- connecting the machine on 3 different groups of outlets - no change
- letting the machine idle 'till the point where hwinfo shows all fans at 0 speed -> still high pitched resonance
- (slightly) pushing on the machine to see if the resonance goes away -> nope
- remove nvme drive + memory and boot -> noise still there (so 100% sure it's the nuc - already suspected that, but still better to verify).
- tried disabling all unused devices in the bios (lan-port, charge-usb, sound, integrated graphics, ... ) - no change
- For a different machine someone suggested to set the high-performance maximum to 99% instead of 100 - sound still there.
- Tried blowing in the holes (no canned-air available, and I wanted to be quick before any return-policy expired) - no change.
What I do notice is that
- the resonance seems to have a slightly change in it's 'chirp' each time the nvme-access led blinks.
- I can influence the noise by doing "something" on the machine, like refreshing a page in the browser - or starting an application.
At this point I'm pretty sure it's not related to the fans - specifically because hwinfo was reporting both fans to be at 0 RPM and the sound was still very audible. Putting in the nvme-drive / memory is no problem, but taking it completely apart and cleaning - on a brand-new device - goes beyond what I feel comfortable with, so that's not an options for me.
Considered if I could "ignore" this - but after sitting behind it in a quiet roomin the evening for a few hours and going to sleep my ears physically hurt. I seem to be super-sensitive to this. Only had mac's since ~2008 they never had anything like this. So I contacted the shop asking for an RMA/exchange and they were so nice to agree - letter with a return-stamp is on the way - so when I receive that I can stick it on the package and send it back. They say that usually a new machine would then be shipping within a week - which would be great .
I really hope that one then doesn't have the problem, because I really like this machine and it would actually be very sad/disappointing it this appears to be a structural issue.
I manged to do some quick tests.
I believe frequency is very low. it is most likely to 109Hz or even lower.
When I pickup unit even slightly on one side noise goes away. I can also feel slight vibration by my hands and same time noise goes away. Strongly rely on surface unit is sitting on. More thick table was more transferring noise compare to for example floor.
I wanted to replace thermal paste anyway so I will dismantle unit and check fans and any loose parts.
I will post more findings later.
Thanks for support Mike.
I was finally able to recreate the resonance issue.
Go into bios and set the boot performance to MAX TURBO
That's all it took. Upon bootup, the unit fans went to max and the resonance began and continued until the boot sequence completed and the unit settled down.
As former US NAVY Acoustic Analyst, I can hear that the resonance starts as a function of fans speed. The resonance itself, does not seem rotational, it sounds more like a mounting vibration.
I went back into bios and reset the boot performance to preserve battery (Mobile).
The resonance is not going away now and is a feature regardless of the boot performance setting.
That tells me that indeed, the resonance is part of the build and I believe, maybe the mobo mount or screws or something that needs to be tightened or loosened.
I initially thought it was LED film but ruled it out when, during the resonance, I removed the top cover and touched the LED film and it wasn't vibrating.
I've isolated the resonance 100% on my unit.
With the resonance sounding:
I'm disassembling the unit at some point in the next week and will report back exactly what is causing the resonance.
So, as a short update, got the replacement device and it seems to have the same issue (maybe slightly less annoying chirp - which a colleague listening to my verbal description said might be "coil whine"). I tried it in the living-room for VR and from a larger distance and it is acceptable.
For now I'll use it as my main PC and worst case I'll upgrade to something else in a few months and either sell or "demote" this machine to living-room machine. The idea behind the NUC is great, cpu is great and gpu-performance it close to what I expected and okay for my current needs - so hopefully I'll figure out some work-around in the meanwhile. Maybe increase the fan-speed to have some noise overpowering it or something. Posted a new youtube video (https://www.youtube.com/watch?v=gk5yXoE64nM hades canyon nu8i7hvk chirping noise - YouTube ) to demo the effect (best capture so far, also showing fans at 0) and see if other people also run into this issue and maybe have other ideas on what to try. Thanks for all the help so far... and if anyone has other ideas on what to try, please let me know.
I assembled my NUC today and also appear to be experiencing this annoying resonance sound from the left side of my unit.
Just the tap from setting the unit back down on the table made it go away, at least until the fan speed reduced and then again accelerated, upon which it returned. :-(