XPS 15 (Haswell) Owner's Lounge

Strange, very strange. I tried multiple cables and different USB ports as I indicated; none worked. Same cables work perfectly on a Windows 7 machine.

 

I am getting a lot of heat from the laptop, even when under a low load. HD and CPU is around 90F - 100F even when CPU usage is under 1%. I have the mid-tier version. Any ideas?

 

Very strange indeed. I really don't know what else to explore at this point in your case. And just so you know, my other 3 cables work on other machines as well (Macs). It's a very peculiar problem, no question.

 

Just an update - had my 3rd motherboard replacement today (i.e. it's my 4th MB now) and the coil whine is still there. Interestingly enough, though, the sound and behavior is slightly different with every MB I have had...

Another interesting thing - I can swear that when I plugged it up at first there was no or minimum (barely audible) coil whine, even when it reached full charge. I left it running a while and when I came back it was same old, same old.

I also tried locking the battery charging on on 90%, but the coil whine is now uniform, regardless of the charge level - always when the AC is plugged in.

 

In my opinion, the QHD+ is the best option. The pixels are so small that you can essentially use the display at any resolution that you want without having to worry about the screen looking blurry. I'm currently using a resolution of 2048x1152, and it looks great (I wanted a little more working space than 1920x1080).

 

That's really, REALLY good ! I don't know what sort of ambient temperatures you have where you live but 35-40C for the CPU is phenomenal. You have to bear in mind that this laptop is very thin and has small heatsinks. Additionally, it shuts down the fans at low loads so it's running on very limited, passive cooling...

 

I thought I was crazy too but I have exactly the same issue. My Nexus and Galaxy cables don't work (same connect/disconnect problem), an old HTC works, and the best is a logitech one that I got to charge my wireless mouse. It IS DEFINITELY the cables...

 

Hmm okay, it's not too noticeable, just a little warm when I use the laptop on my lap.

 

Here is the little piggi:
https://www.youtube.com/watch?v=X5n7T-3iZrc

 

cool man, good work - gonna be a pricey fix for Dell....looks like it's in a spot were it would affect mid- and top tier models equally as well...

 

The responsible Coil for whining
https://www.youtube.com/watch?v=X5n7T-3iZrc

Yeah, it is a poorly designed voltage regulation in those models. The responsible devleopement department uses a faulty module ... However i wonder if they are deaf at this department. As a manufacutrer i would shame myself if i had to sell a series of faulty notebooks O,o

However, maybe this could be a replacement coil: ??
http://www.conrad.de/ce/de/product/...1A92BE86520D718C6787BAFCE2.ASTPCEN07?ref=list

I need to understand what the things written on the coil mean:
2R2
339
422

2R2 stands for the inductivity of 2,2µF. But what does the rest mean? This type of coils can be also found in iphones and ipod. Somebody a guess?

 

Still not double-checked, but it seems like upgrading to the 3.14 kernel has finally solved the wifi issues with one of my routers, even with the 11n mode enabled.

 

That's great! Did you happen to test it with other units? Because I noticed a lot of variance in the quality and loudness of the sound from MB to MB. Will you be presenting Dell with this? I really wonder how they are going to react.

 

Awesome..

What does the waveform represent? Is it showing the noise level captured through microphone or is it some voltage from the component? If it is voltage, how does the waveform indicate that it is related to the coil whine?

 

How did you upgrade to 3.14?

 

i recieved my mid-teir xps on the 3/27
beautiful machine, thin, light, powerful, quiet (sometimes). i'm not sure but i dont' think have the coil whine, and if i do then its really minimal.
the touchpad is easy to retain smuges and very hard to clean. the palm rest with the soft cilcone is nice under the hands but not so easy to clean. i'm still trying to get used to the touch pad in the center of the laptop along with the "no number pad"
however, I have a little problem and question to see if anyone here can help me solve. the camera goes out if there's no light in the room or on me (this is the only laptop i've owned that does this). i've noticed what looks to be a sensor beside the camera. is there a way to disable this sensor?

 

I'm not sure how the guy in the video locate the defective unit. As far as I can tell, he's using a scope to measure voltage of a component. From the shape and size of the component, it's likely to be a power inductor. Unless he's an expert in regulator, I don't think he can tell whether the regulator is correctly designed or not just by the not-much-detailed waveform. From my experience, the reason why the scope does not show a steady waveform is because the frequency of the signal is not stable.

Furthermore, I don't think uF is the dimension for inductor. 339 can be a way to label the number the capacitance. Like 33*10^9pF. I'm not sure it's pF/fF or something else.

 

Not sure why, I was pretty sure I was using a generic 8.1 pro iso as a didn't trust Dell's.

So I did try the Dell DVD ported to usb and installed fine.

I still doubt Dell iso as the power option has no screen brightness! What kind of Dell 8.1 they gave me I have no idea.

 

I meant generic 8.1 pro

 

Thanks. I'm glad I got qhd then.
Just that everything looks so small. Mind you I need glasses I think

 

Well most XPS 15s come with 8.1 non-Pro, so unless you have one of the few that came with Pro, chances are the ISO you tried to use could only install Pro and not non-Pro, which would explain why you kept getting prompted for a product key that kept getting rejected.

You might have to click the "Show options that aren't currently available" link in the Advanced Power Options window to reveal adaptive brightness, fyi. But it's definitely on the Dell image since I found it before I did a clean install.

 

Turn up the volume in the video to very loud! You will hear exactly the interferences that are shown in the video. Take a SMD stetoscope and listen to the coils. This is the one making the noises.

2R2 is the usual description for the inductivity in SMD Coils. I used a scope to measure the frequencies in the five coils. Only this one has changing frequencies, that fit exactly to the whining noises.

The frequencies vary dependent on the load. If you have some read/write activity on the SSD or the battery stopped loading there is this activitiy on this coil.

I contacted them. They did not react. If i would have two more samples i would investigate two weeks to solve the problem. We should give it a try with some epoxy.

The waveform is the voltage on the coil. If i resize it you will see a digital signal. Some PWM (Pulse Width Modification) chip is generating this signal dependent on the load. (SSD activity, Battery charge/discharge, CPU state). Turn up the video volume and you will see exactly what you hear.

 

Yes, I can hear the noise. But I thought the scope you used was an electrical scope. So you get the conclusion that the inductor is causing the noise by the fact that there's a noise and the voltage on this inductor is non-ideal? I don't think the reasoning is that convincing.

But again, even if you are right that the frequency on that inductor is changing, it might be a symptom of some other design issue rather than indicating that this power inductor is itself defective. Thus, switching the inductor might probably not going to help.

 

This coil is causing the noise. As already written you can use a stetoscope to identify the noise. The osciloscope is only a second proof for my assumtption, as this is the only coil with changing frequencies that fit to the noises. I do not know if the frequencies are ideal or non-ideal. But from the view of an electrical designer it would be very simple to check this. The design of such a regulator is usually first done in a simulation. Then a prototype is checked vs. the simulation. If it doesnt. fit the simulation is wrong and the design aparment has to do some debugging with theyr circuit design programm reseller.

I do not think that the coils are faulty. I think DELL should have used epoxy filled coils. I also think it should be possible to fill this coils with epoxy afterwards. I i would know which coil this is exactly i would take this coil in parts. When i know how they are built i would open mine on my notebook and fill it with epoxy. This is only a fast fix for the user!

Further i think the electrical characteristics of the coil are good from the electrical designers view. But for some unlucky reason it is whining with this type of coil. I am not sure if the people at DELL are deaf. I do not think so.


My no.1 assumption for the real reason of coil whine is a mismanagement between DELL's design apartments. The electr. design appartment doesnt want to have the fault on theyr side. But i am reaally wondering how it could happen that such a noisy circuit design could go into mass production ... Somebody is sitting in the DELL's design apartment in a managment position which doesnt want to see ....

 

Some pics of that




It is the same signal captured with different frequencies.
The upper one with 2.5us and the second pic with 0.5ms.
The noise-resonating freqency of the coil is somewhere between
2khz and 14khz and more...

The best thing would be to take this coil as a single part and reprocude sound creation

 

Wait, so does this "coil noise" effect every laptop that Dell is selling?

I was just about to place an order.

 

It doesn't affect everyone. I am fortunate to have received an XPS 15 without any kind of coil whine. Perhaps I have just been lucky? It would be interesting to make a poll to see how prevalent this problem is. I receive my XPS in February (ordered in December).

 

I've had my motherboard replaced 3 times. Each one (4 in total) had coil whine. How unlucky can I be?

Had it been a rare problem, it wouldn't be an issue at all. Everyone would get their units repaired/replaced, and no one would go as far as to make all this fuss, invesitgation etc.

 

I received mine on 3/27 and i'm glad I don't have the coil wine either.....I have a mdi tier model.
however, I don't understand why the camera basically turns off in low light situations...it's a bit annoying as every HP that I've owned seems to work in even complete darkness.

 

Hmm, interesting. Maybe the mid-tier models are not affected?

 

Got mine on March 26th, top tier model and no coil wine.

 

Guys - you are 100% sure that you have no coil whine? Even when fully charged and the power cable plugged in?

I would be interested to hear from people who had the coil whine and then actually managed to fix it with repair/replacement.

 

well the sounds that I've heard people recorded and posted online i'm not getting any sounds like that. just regular fan noise when under some load or minimal fan noise when not doing much at all.
I wished I had opted for the top tier model because now I wish I had better battery life.

 

So envy you ; / I've gone through 4 motherboards and each one had the coil whine. I imagine it has something to do with the factory in which they were made. Mine were probably all from the same facility...

At this point I gave up on replacements, for the time being. I might try Spanish Dell, as I'm moving to Spain next week. The Polish technician said they might be using a different factory.

 

Btw, has anyone tried the "registry fix" for the coil whine that's been floating around the Internet? The command line from this site Stop error message on a Windows Server 2008 R2-based computer that has the Hyper-V role installed and that uses one or more Intel CPUs that are code-named Nehalem: "0x00000101 - CLOCK_WATCHDOG_TIMEOUT" is supposed to fix the problem, but I can't even use it, says "access denied"...

 

I was a Microsoft Store a few months ago and got to play with their demo XPS 15 top tier for a good 20 minutes. I spent a good 3 or 4 minutes lifting the laptop and putting my ear up against it and I couldn't detect any coil whine, the system was plugged in and fully charged. I'm normally very sensitive to high frequency noise, I can usually hear a coil whine in everything from my TV to the refrigerator.

 

Yeah, that's another interesting aspect. I heard about this too. Also I imagine the review copies must have been coil whine-free, as I read all the official reviews before buying the laptop and found hardly any mention of this huge problem. So did Dell intentionally hand-pick the few coil-whine free copies for promotional use and then dumped the faulty ones on us suckers? I know this reeks of conspiracy theory, but this seems like the only natural conclusion. Either this or the Internet is filled with lots of very unlucky guys.

 

I think someone tried that and said ti didn't work. It also may not apply to Win 8.1. But the reason you're getting access denied is because you're not using a Command Prompt that has elevated privileges.

 

Mid tier are affected (apt least mine and I remember others).
The s3 registry fix won't help. You can also play with restricting s-ststes with throttle stop under window or also with Linux kernel boot parameters. It does not work.
But I'm very interested if somebody already tired to put some epoxy on the suspicious voltage regulator. They could be the last resort!

 

Well, if Dell doesn't fix the issue somehow, I will demand a full refund or new model replacement...just when my warranty draws to its end

 

Ok guys. This is the good news! There are some versions of non noise XPS 9530!! As we found the responsible coil (and regulator unit) for the noise we can find an alternative coil maybe? As i found on notebookcheck there are different coils used in the power regulator unit!!



This is my regulator unit. Some high-res pics of a non-noise regulator unit would be extremely helpful.

e.g. here is also a different coil soldered https://www.youtube.com/watch?v=ZqpKcLv15Qo

And no, there is NO software solution to this problem. If you have coil whine it is a pure hardware problem. You could have some different noise with different load, e.g. turning off cpu power states, but why did you buy a powerfull powersaving processor.

 

I recently received a replacement unit for the XPS 15 due to some discoloration on the screen ( http://forum.notebookreview.com/del...l-xps-15-what-2-years-brings.html#post9608474) and found something interesting. The new replacement unit has a much different touchpad. It's super smooth and slippery and quiet while the older one has much more friction and noise when sliding and is more prone to smudges and discoloration. It looks like Dell either changed the trackpad material or they've been producing different trackpads on various models of the XPS 15. Anyone care to describe their trackpad?

 

I really appreciate the work for loacting the mis-designed voltage regulator, many thanks again.
Sending back my laptop is no option for me. So I might can fix the problem myself.

1) An important information is missing: What are the temperatures of the voltage regulator?!?!?! Do it get warm? What is about heat when stressing? - Just asking to get a feeling if one would decrease the stability/lifetime of the XPS.

2) Can you again post the part number of the regulator? Which of the three shown on the image is the most critical, e.g. most noisy and which the warmest?

I'm really considering to put some epoxy on it, but I'm afaraigt of thermal insulation. Otherwise if it gets's warm I may consider to get me some copper foams (disassebly old cellphones for example) and try to glue it with with electrically and glew this with a thermal glue on the regulator.

 

The noisy part of the regulator is the coil. It is the "2R2 339 422" coil next to the Battery connector (see arrow in pic). This is a 2,2µH Coil. In your XPS it could have a different Name but should be the same type.


High-Res pic check my previous post.

This coil is not getting warm. I did not check the temp. under heavy load, but i think it will remain cool. Putting it in epoxy is absolutely save. The noise will affect lifetime, out of the simple reason that noise means mechanical movement. If there is a coil part moving around with some frequency it will destroy itself after some time.

 

Ubuntu Kernel PPA: Index of /~kernel-ppa/mainline

Let me know if you need more detailed info.

 

Just note my 2 XPS's were bought in Spain, and both have coil whine (one stronger than the other).

 

Thank you for your fast reply.
At the moment at work (and using the XPS), therefore I can not take a look.

On the photos, I see no coil. Is it located under the regulator? Is it good visible when I open the laptop? Is it a copper winding?
I once had my XPS open to exchange the hdd with a sdd and remove the m-sata sdd from my mid-tier model. As far as I can remember, the regulators should be accessible just with removing the back plate and no need to remove the heatsink/fan unit? Right?
If I find that nasty part I will definitely give it silence by a expoy bath!!!!!! Yea! Hope I will work out.

 

The coil is inside this small grey box. "2R2 339 422".

 

I agree that the noise is probably due to that regulator but I don't quite think it's the coil.

Firstly, it's not necessarily mechanical movement that causes the noise. Radio signals can be audible as well when circuit is not properly designed , especially in wireless applications.

My guess is that the problem is probably related to driver that drives the power inductor. You should be able to find a chip near that inductor, probably DIP. Usually, to deliver different output power, the driver needs to alter its modulation frequency. In RF applications, signals interference with each other and one signal at a higher frequency can be modulated to a lower frequency by a nearby signal. When the modulated signal is at a frequency within human's hearing frequency range and strong enough, it becomes audible. It's a very common symptom in PCB design in wireless applications. Sometimes, the circuit even screams if something goes wrong.

In conclusion, if you want to dig deeper, I would recommend you to locate the driver chip that drives that inductor. It's certainly near that inductor and probably in DIP package. You can get its datasheet online. In the datasheet, there should be some explanation about how the chip changes its modulation frequency to deliver different power. If that's the case, I would say the cause is found. But as how to solve it, it's not that simple because it's neither the driver nor the inductor itself is faulty. Probably a different driver that works in a different way but achieves the same performance can solve the problem.

 

Without mechanical movement there is no noise. There is no physical phenoma to create sound or noise without mechanical movement. if you know one let me know, so i can have the nobel prize .

However you are right, the reason is a bad designed driver chip-coil constellation which makes the coil whine. As written i have checked the noises source with a stetoscope and verfied with the osciloscope. The mechanical noise source is this coil.