Buzz/high pitch sound in headphone output

I'll try to borrow that good headset again and give this another shot..

 

Actually, I can hear it from my speakers. (I spent several minutes hunting around the house for some before I remembered I had some plugged into my desktop. Go figure.) What I did was, set the sound as I previously had it, "calibrate" my hardware speaker volume level with some music to match normal hearing levels (ie, turn them way way up), and then turn off the music and muck around a bit. It's clear as day. Not a Michigan day, either. *coughcloudscough*

 

That figures.. Also figures that I can't get the sound out of the headphones still.. lol..

 

Please, i didn't mean to stop the investigation!
Just saying that if it is, and i think it is, a hardware issue it can't be solved by simple means..
But it is very interesting to try to find the origin of it. And i think the hard drive is the main causer of this noise, because i can ear the same noise but much more quiet in my Mp3 player which is HD based.

It would be nice to check if other OS make it also but right from the startup of the computer the noise is audible.

 

Actually if the noise isn't being introduced into the system through a chipset, you can normally "easily" fix it by improving the insulation around wires or devices, or by introducing a ground to draw the rougue signal out. The alternate would be that a ground needs to be moved because two items are being grounded to the same devices which inadversely would create a circuit and share the noise between devices.. Again, this isn't to hard to fix other then finding another grounding location inside the chassis can be tricky in notebook.

 

I noticed this thread lost alot of steam.. is there still reason to try and fix this? or should I stop?

 

I'm still interested...I just got a little sidetracked lately. I need to go back and read through what I need to do. I thought I'd answered all the questions I could but I will double-check.

 

I am still interested. I just have been distracted as well... it still strikes me as significant that it's only there for me on battery, after I resume, and at no other time. If I standby, resume, and hibernate, it's also still there. Rebooting makes it go away though... so I think software is involved.

edit: actually, any time i resume from hibernation it starts making noise. so hibernation and suspending, or resuming from those... either causes it or triggers is.

 

I'm still with 'ya...not much help from me this far, but I'm in

 

Any of you testers willing to give some trial and error for me?

Use RMclock for this...
On the advanced CPU settings page, Processor tab
try enabling and disabling each of the boxes one at a time (restart in between).
I've personally got them all checked.
On the platform tab uncheck everything (same as I have)
make sure to check the "Apply these settings at startup" is enabled.

I dont care what power profile you want to use.. but for simplicity lets just say "maximum performance" so we can pump as much voltage through the system for noise generation as possible.

Let me know the results.

 

I haven't used RM clock before. Is this it?
http://cpu.rightmark.org/download.shtml

There seem to be an awful lot of download options. Which do you recommend?

 

Sorry for not being more explicit, I forget how many tools Dmitri has created since I first started using RMclock...

CPU Rightmark is the full name of the tool.
http://cpu.rightmark.org/download/rmclock_215_bin.exe

 

I did some RM Clock testing. I am looking for the setting with the loudest noise.

Tests are performed by adjusting the settings and then restarting. The interference will be judged while doing activities known to coincide with the noise such as scrolling, accessing HD, utilizing CPU.

Platform tab:
Enter C4 on C3 Command - unchecked
Enable Popdown mode - unchecked
Enable Popup mode - unchecked
Apply these settings at startup - checked

Processor tab:
Default = all checked except for C1E, C3E, C4E

Test #1:
All unchecked...so far it doesn't seem like much of a change. I'll use this setting for a while. OK, doesn't seem like much difference.

Test #2:
Enable Thermal Monitor #1 (only) - checked
Possibly a slight increase in overall noise. Nothing really noticeable though. Still using it. OK, there is a slight increase in the noise with this setting.

Test #3:
Enable Thermal Monitor #2 (only) - checked
About the same as in Test #2.

Test #4:
Enable Enhanced Halt State (C1E) (only) - checked
Less noise than Tests #2 & #3, but by very little. About on par with Test #1.

Test #5:
Enable Enhanced Stop Grant State (C2E) (only) - checked
No noticeable gain in noise level.

Test #6:
Enable Enhanced Deep Sleep State (C3E) (only) - checked
About the same. Background interference/buzz may be slightly louder.

Test #7:
Enable Enhanced Deeper Sleep State (C4E) (only) - checked
No significant change.

Test #8:
Enable Hard Enhanced Deeper Sleep State (Hard C4E) (only) - checked
If anything it's a touch quieter than the others.

Test #9:
Everything checked
About on par with Test #8, maybe a touch softer.

Conclusions:
None of the above settings seemed to significantly add to the noise or reduce it. I've come to the conclusion that EMI is causing the noise you are hearing. This is quite common in notebook computers. It's not unlike the noise you would hear when listening to your walkman back in the 80's. Your notebook is far more complicated than a walkman, and has a lot more power running through more moving parts--creating the noise.

It's quite possible that a number of you don't hear it because the overall volume of this noise is very low. I had to look for it to hear it. And now that I know it's there, I still barely notice it.

For those that are bothered by the noise, it has been said that an external sound card will give you a cleaner audio source.

 

I've also got this on my HGL-30, I'm 100% sure it's coming from the hard drive, since it is louder when the hard drive indicator light is on.

 

When you say louder, do you mean really loud or just noticeable?

Also, what CPU does your HGL30 have?

 

Using an External soundcard doesnt seem to help mutch, we have tested a HEL80 with both a USB soundcard (M-BOX) and a firewire soundcard (M-Audio FW410) and in both cases the noise was present. also the noise seemed to be carried via the firewire cable to the speekers so that even if we turn all the faders down on a mixer through witch the sound from the soundcard was going, so that all the audio played from the soundcard was completely attenuated, the noise was present, we could hear the noise in the speakers even though there was no Audio coming through the mixer. wen bringing the faders back up though, the noise stayed at a constant level, as the audio meant to be comeing from the soundcard was became louder. And when the link somewhere along the line between laptop and speakers was severed, the noise ceased.

The noise also became softer when the AC adapter was unplugged, and the noise seemed to vary in intensity in accordance to things going on inside the machine, we didnt test this enough to pinpoint any contributing factors though.

Its a nasty little problem, and not at all fun when youre trying to make recordings. i hope there is a fix for this because i ordered a HEL80 today. and i want it to work properly

 

It's noticeable both when the HD isn't spinning and when it is spinning, but it get s loud enough to be heard easily when it is spinning.

It's a 2.0 GHz Core 2 Duo, 100 GB 5,400 RPM HD (ordered from powernotebooks.com, so whatever hard drives they put in there).

It's not a massive problem, games can be played fine and the music quality drowns out the sound, but it's a small problem in an otherwise perfect notebook... It'd be nice to get it fixed.

However, I do not know if this is the same problem as for the HEL-30...

 

Daniel, that sounds...umm...weird. If using an external soundcard doesn't solve the problem, then there is something seriously wrong with this computer...the only thing I can think of then is the motherboard. If an external soundcard doesnt work, then I think that Compal should do something about this problem.

 

Hi everyone! I posted the sound files in post #6, sorry for not participating after that! Seems this thread has caused quite a stirr, I`ve been trying to read up on all the posts here, but it was simply too much to read... Can anyone update me on any progress made with the noise-issue?

Daniel_Lacey and I (post #116) have found the noise very much evident when using external sound cards. This is a major problem for us, being laptop-musicians - since such noise is disturbing in live performances, and even more so in sound recordings.

If you have come up with suggestions for solving the problem, we are happy to try them out, and document the results here both in writing and in audio!

 

Perhaps Chrisyano can post the possible solution I PM'ed him. Unfortunatly I already deleted that email... If he doesn't have it, I'll post about it again..

Long story short... I've already explained that I'm very certain its EMI from the motor in the hard drive. Look at the location on your notebook comparing the HD and the sound output AND the PCMCIA slot.. since.. as you may are may not know the EMI is cast out in the direction or the spinning, you are hurling EMI directly at both locations from the HD..

Hard drives dont stop unless you enter a very low state of power. RESTATE: Even if your HD light isn't on, your HD is spinning.. it does this keep the access time faster. If it didn't always spin to some degree it would work more like your CD-ROM drives and have a delay at spinup-spindown. Not to mention it would shorten the life. Your HD light only comes on when your system calls the HD to access a datastream(i/o).

 

To Daniel_Lacey and Friends.....this is a major concern for me as well. I do a lot of production work in ProTools and I've been looking for a laptop to run ProTools with a FastTrack Pro USB in on the road with a MIDI keyboard. Does the noise get recorded onto Audio Tracks from external MIC's, line input gear like keyboards etc? Does it show up during mixdown do MP3's or WAV masters? Please reply as I am on the verge of a purchase....but I'll hold off if this issue is not resolved!!!

 

OK, looks like Chrisyano cleaned up his messages like I did, so I'll make this short and sweet...

Anyone that already owns this system and wants to give a try, please step up and give back your results.

:::THE SECRET:::
Aluminum is a great EMI sheild... And it just so happens most people have aluminum foil around their homes...

1) Open the laptop and pull out the HD.
2) Put a single SMOOTH layer of foil around all sides of the HD that don't have exposed circuitry, saudering points, chips, anything that isn't the sheetmetal case. If you are lucky, these drives has a thin plastic shell around them (for reducing SOME emi) so you can go the entire way around. JUST A SINGLE LAYER. Cut out around the connection cables for the HD also if you can without contacting anything bad.. (see above)
3) Build a small shield at the audio connectors if you can. EG: no exposed metal, circuits, chips.
4) Do this at your own risk. I've personally done this in several pedistal/tower servers that had high end scsi drives causing enough EMI to interfere with the video signal.

AGAIN... DO THIS AT YOUR OWN RISK. BE SLOW AND CAREFUL TO CHECK AND DOUBLE CHECK THE FOIL WONT CONTACT ANYTHING BUT PLASTIC IN THE CHASSIS.

For a little more info.. this only stops EMI... If there is actual voltage noise going through the chipsets, there is nothing to fix this other then possibly a class action lawsuit against either compal or the motherboard manufacturer.

If you want a more professional appearence and possibly greater EMI sheilding...
http://products3.3m.com/catalog/us/...mand_AbcPageHandler/theme_us_oemeletrical_3_0

This stuff is a bit pricey for just doing the HD and audio connection.. but you might be able to go somewhere like a hardware store or electical supplies store and get a small section of it cheap.

 

Pyro, you say you're pretty sure that the noise comes from the hard drive, but I don't think that's the case, for me at least, unless I'm missing something (which is very possible).

The reason is that the noise I hear is intermittent, doesn't seem to correspond to the hard drive activity, and only appears when I'm resuming from either standby or hibernate. In addition, it does seem to nearly always either start or increase in severity when I visit www.twit.tv -- I mention that website because it's the only one that I'm pretty sure does that consistently. I've tried Firefox, as well as Opera. Weird, ne?

The point of this post is: are you able to reconcile your hypothesis to my situation?

PS. I'll get playing with RMClock with the post you made a bit back. I haven't gotten around to it yet because it's just tedius enough, and I just busy enough, that I've been putting it off ^_^;

 

It would be interesting to find out if the noise only occurs or is quieter for specific hard drives or in general specific component configurations (memory type and speed, hard drive type, speed and capacity, WXGA+ or WXGA) etc.. I know this sounds strange but it may be that a Travelstar has more electrical noise than a Momentous drive or the memory needs to be shielded, perhaps Corsair memory has less EMI than say OCz etc.. I will say that I have heard very faint electrical noise whenever I have used headphones on any computer so I am used to a slight electrical noise in my life, hell some light bulbs are loud but we have learned to ignore it. Also as a side note my iPod also has what sounds like everyone is experiencing.

 

This noise is a very common occurrence in electronics. Especially when a smaller form factor is part of its design. I've heard this noise in a lot of other electronics equipment over the years.

The HEL80 is not the only notebook system that has some interference in its audio output. I found two posts a couple of days apart listing the same noise in two different notebook systems.

It is true that the EMI is probably also coming from other parts, but I would hypothesize that the HD is the most likely cause of most of the EMI causing the "noise". It's got the mass and speed to create a lot of EMI.

 

Has anyone had a reply from Compal about this yet?

Has Donald contacted Compal about this? 3 weeks ago tech support said he was working on getting a solution out there, is it posted somewhere?

This isn't good advertisement for the HEL80, hopefully the sellers and manufacturers will move to quickly resolve this, because it is rather annoying on my machine.

 

Hmm, I wonder, how can the harddrive's EMI cause noise in the output on external soundcards? In that case, the USB ports have to be affected as well, in which I think not since the signal is digital until it reaches the sound card (to my knowledge, a digital signal is not as much exposed to electromagnetical interference as an analog (?))

I don't think this has explicit connection to the noise problem, but I reinstalled windows here the other day, installed drivers, everything ok. But, it seems that the sound quality is lower than before (I've set equalizer to 'powerful'). Also, when you start small stereo's that has no amplifier 'protection' at startup, you will notice kind of a 'bopp' in the speakers. I have this 'bopp' every time a sound starts, and when I'm listening to music where there is a lot of thing going on and off, I hear this almost all the time. Anyone understand something of this?
Edit: Seems that the 'bopp' isn't that noticeable when disabling the EQ. I didn't have this problem earlier

 

Playing around with RMClock a bit, checking the "run HLT command when OS is idle" box removes the noise.

Also, I noticed something interesting. On the Monitoring page, when that HLT command thing is going, or when I freshly reboot, CPU Load and OS Load are more or less in sync. That is, they're out of sync, but they have a general correspondence.

When HLT instructions aren't being spammed at the processor during idle cycles, and I have resumed, the CPU Load is consistently a lot higher than the OS Load, and this is when the noise occurs.

 

I found information on another forum that also talks about RMClock and how to set it to reduce noise from the CPU.

http://forum.tabletpcreview.com/archive/index.php?t-3516.html

Also does using the HLT command cause RMClock to constantly use the CPU, instead of the CPU idle you get RMClock idle and CPU is contantly running RMClock generated HLT commands. Now if this is indeed true how does this affect, Battery Life, CPU Life, and wouldn't the CPU become very hot.

 

hlt is some kind of sleep instruction, but it did seem that my fan was running a lot more, so, with my little knowledge, my deduction is that battery life would have taken a significant hit. I'll try the thing in that link.

edit: actually, that was the post I saw earlier. I tried following the directions, but there wasn't any effect.

 

You have half a valid point on the digital signal thing... But I understand where you were going with it.

Digital Signals have a reputation for being better against EMI because its fairly easy to clean a digital signal using its inverse. I dont know of any usb connection on the planet that does line filtering. If you think about it..

__|---|_|--|_____ (poorly drawn digital signal) all you are looking for if you were the recieving end of this is either the 1 or the 0's.. (up/down)... if you send the inverse signal to this, you just look for where they match and you are 90% sure your signal is accurate if you switch at those points.

Just to clarify, this thread has been covering "atleast" two known sources. One being the voltage regulation near the CPU (while CPU scaling), and the other that is definatly EMI... I'm just tossing the HD out there as a theory because its a 12V electromagnet that is the #1 cause of EMI in any system. The smaller the form factor, the worse it gets.

As I've stated earlier, I dont have this problem to the best of my knowledge, so I just have been trying to be a guiding voice and share my idea's.

I remember working at a hard drive manufacturer years ago when they were trying to make prisms into data sources.. One of the great perks were the compounded sides you could read and write data from make the storage potential massive for its form factor, the other was lack of readible EMI because there were no moving parts.

 

You have a valid point here... but I think you lacked the key to why your point is valid.. different hardware has different power requirements. The more power you run through a device, the more noise it makes and more shielding it requires. Unfortunatly most lab tests are only using a sample of hardware to test with that is a fraction of what the market will potentially use.

I personally used the generic ram from powernotebooks since there are actually less then half a dozen ram manufacturers on the planet. All ram has lifetime warranties, and the speed difference from what I saw when I was buying my laptop didn't exist.. Timings for even the most expensive chips were the same as the cheapest. But, voltage requirements do change. The type of connection the ram chips use plays a large part of the exposure and exposing of EMI. KingMax is known for producing Tiny BGA chips, which have no exposed connections, so barely any EMI problems. Their memory tends to be a tid bit more expensive, but its used in alot of medical room devices, or anywhere that there is electromagnetic sensitive material.

I also went with whatever brand hard drive they stuck into the notebook, because I've got a 3 year warranty on the entire laptop, so I dont care who's drive it is.. I also went with a 5400rpm HD to ease my battery life and lower EMI.

 

I'm not sure if anyone has contacted Compal directly, I know I haven't, but then again I can't give them a system to test it on since as I've stated, I can't get the problem. You also need to keep forums in perspective to their audience. Most people find forums when they need answers, not because they are all happy and life is moving great... You also need to watch all the names on this thread, there is roughly 1 dozen names including myself, (and I dont count since I dont have the problem). Hardly a reason for compal to feel there is a huge problem. Also, compare the problems you can find on the internet from Compal as you will for a dell, and it may put things more into perspective

As for Donald and powernotebooks, I'm not aware he's participated in this thread, and I'm sure anyone who calls them he's working with directly. If you were contacting them, take some initiative and call them back to make sure the issue is still active.

 

I got your PM, and I hope you wanted my feedback on here. So here go's..

HLT command is a bios setting that works directly with a built in hardware feature of your processor to disable sections of the registers that aren't in use. Its sorta like the new motors that dont have to use all the cylinders unless there is enough load to warrant it. AMD users might know about this because we started using on the Athlons as a way to get the heat generation to go down. Dont worry about your system being slower though, because this all happens at the registers level which is extremely fast and in my year of using it can't say I've seen any negative effects.

As for it removing your noise... It could be the lowered voltage needed by your processor... thats just a guess.. Have you read all of Chrisyano's posts about fixing his voltage regulator noise using RMclock?

As for the results on the monitoring page: (taken from the documentation)
# CPU Load - shows actual CPU load (i.e., the ratio of "active" processor cycles to the total processor cycles per given amount of time). This value is measured via CPU-specific performance monitoring counters and represents true CPU load, i.e., the amount of time the CPU spent in "non-asleep" state.
# OS Load - shows CPU load level as reported by the OS, in a manner similar to Windows Task Manager.

More info from the documentation on that halt feature...
Run HLT command when the system is idle - this feature causes the utility to spawn external idle-priority RMClockHLT processes (as many as the number of system CPUs present, either physical or logical) which run HLT command in a loop. This will efficiently "steal" all idle time from the OS "System Idle Process" while retaining the CPU in a relatively low power state (C1) when the system is idle. This option can be used as an alternative to "Use OS load-based management" option described above on Intel Centrino platform, as well as in any other cases when you don't want the OS to manage the CPU idle time (e.g., when you want to prohibit the CPU to enter into deeper power saving states like C2, C3 or C4 if they're known to cause problems).
NOTE: since the OS will no longer receive any idle time, the OS CPU load indicated by Windows Task Manager and similar utilities will continuously remain at 100%. The OS load reported by RMClock, however, will still be correct, as the utility subtracts the amount of time the CPU spent executing RMClockHLT process(es) from total CPU busy time.

I personally use OS load management checked since that is recommended in the readme for centrino platforms.

 

I'm sure I read them all. I remember one in which he mentioned using RMClock to fix noise from his processor. I tried what he did and didn't notice any effect.

What gets me is that the disparity seems wrong... where is the cpu activity coming from if not a process the operating system is aware of? And, the noise is present only when it occurs. It seems that whatever is causing the one is causing the other... I suppose this will bear more fiddling...

 

Actually I used Notebook Hardware Control (but I'm sure you can do the same thing with RM Clock's utility) to undervolt the CPU when on battery power. This got rid of the voltage noise that only seems to be present on battery. On A/C power, there is no noise coming out of the CPU that I can detect.

 

We're talking about two seperate noises, aren't we?

 

Yes, RM clock lets you set what voltage and speeds you want for a processor.

 

I'm not THAT familiar with processors, but I know from an old thread that I was talking to Dmitri (RMclock dev) about that same issue, and he said that the OS has overhead due to monitoring services running from the OS that are registered by default as "activity". This feature toggles that overhead reporting off.

 

I'd put money on it that it's the onboard sound. If you turn up the mic volume the hissing will get louder. A problem I once had on my desktop that was cured with a nice sound card. Maybe this got posted b4 but just letting you know.

 

If the microphone input is open and you turn the microphone output volume way up (non-muted) of course you will get tons of feedback... I can do this with my SBlive 5.1... and I can do this with my SB Audigy 2 zs.... my dell inspiron...

MS even address this "problem" when XP came out because this confused many users.

 

Well, I know a test I can perform...I could unplug my harddrive and boot an operating system from a cd and do whatever (if that's not done before)...that will definately settle if it's harddrive EMI noise or not at all.

 

That would make sense. Btw, for anyone else that is having the problem, I think they should use a Live CD also. In doing so, we can really determine if it is the hard drive.

 

Then, why is the large disparity there only when I resume from standby or hibernate? It'd make sense that the OS would discount some of its own background services... perhaps these services are glitching up and causing trouble with the processor? -- some mismatch between windows w/acpi and... the state of the processor or something...

*I did hear this in Vista RC 1 as well, though that doesn't necessarily mean anything...

 

You can always buy this and solve all of your troubles...

http://www.creative.com/products/product.asp?category=1&subcategory=205&product=10769

 

Are you saying this because youve tried and it has solved the high pitch buzz on your compal hel80 ? Because this problem persists with external usb or firewire devices, however if you heve solved this problem using a PCMCIA card then maybe we could get pcmcia firewire or usb adapter to eliminate the noise on our external sound cards

 

I have no idea if it would work or not. I just assume bypassing the port all together it'd solve our problem.

 

I'm just tossing this out on a limb... but in my years of fixing computers I've never known suspended states to work 100%... I dont trust any feature for this that isn't directly part of the processor.. Waking up requires the OS to remount partitions, reload and unpack data from the swap, reiniitialize hardware and drivers... There are just way to many points of failure involved... I personally dont turn my machines off unless transporting it for more then I'm walking. Otherwise, my processors throttle all the down, devices go into a lower power state (that support it like HD's, and wireless nics, processors), and I turn off the screen by closing the lid.. if the lid is open, I just have a blank screen saver kick on.

Something to consider though.. my setup is VERY bad if you toss your laptop in a bag without turning it off.. it will overheat and potentially cause damage.. but like I said.. I only turn my system off when taking it somewhere that I am not simply walking to.

 

for those of you talking about using a liveCD.. make sure you literally have your HD unplugged, because it will still be initialized by the bios and run in a lowered power state. Also remember that your CD is ALSO a spindle device, and therefor has a motor that will cause EMI in almost the same location..

I'm starting to believe the EMI is either being created, or entering through one of the host bridges since it effects many ports. But I'm interested to see the liveCD results..

I'd REALLY like to see a small bootable pen drive's results.. Get rid of all the spindle devices.

 

Something interesting I found in the advanced RMclock tweaks file:

// CPU Load "noise filtering" feature when performing performance/power transitions.
//
// Default: 1 (enabled), Valid values: 0 (disabled), 1 (enabled)
//
"NoiseFilter"=dword:00000001