Need ideas: eliminating interference from hard drives

[c0d3h4x0r]

The problem:

While viewing or recording from TV Tuner via coax cable jack, notable interference appears on various channels. The interference is clearly generated by hard-drive activity. The interference looks worse in real-time (looks like frame sync problems) than in still screen captures (just looks like a faint, regular series of horizontal light/dark bars), so providing a screenshot of the interference would be pretty unhelpful.

I never had this problem with my old MSI KT4 Ultra mainboard, but I recently had to replace it with a new ASUS A7N8X-E Deluxe mainboard, and now I have this problem. At the time I replaced the motherboard I also moved everything into a new aluminum case, which positions/mounts a lot of components differently than the old case did. So it's possible this could be an issue of component positioning inside the case.


My system:

• Windows XP Pro SP2 + all latest updates from Windows Update
  All latest BIOS and drivers
  Lian-Li PC-60 Aluminum case
  ASUS A7N8X-E Deluxe motherboard (nVidia nForce2 chipset)
  Athlon XP 1700+ processor
  PowerColor ATI RADEON 9600 series AGP 8X
  2x Seagate Barricuda 200GB SATA drives in a RAID-1 (mirrored) configuration, with Windows installed on this array as the C: drive
  One Seagate Barricuda 120GB PATA drive on IDE0 as Master via C/S
  One LITE-ON DVD-RW ATAPI drive on IDE0 as Slave via C/S
  1GB RAM
  Using the integrated SoundForce/AC97 sound
  Using the integrated nVidia LAN/network jack

Factors I've already ruled out:

• This is not an interlacing problem. I know what interlacing is and what it looks like, and this is not it. If you even try to claim it's an interlacing problem, be prepared to be unapologetically flamed.
  
  The coax cabling/splitting/etc is not the problem. Already had it checked out and fixed up by COMCAST, including a signal amplifier and proper filter on the cable-modem branch. And like I said earlier, this problem only happened when I moved to a new motherboard and case.
  
  BIOS or drivers are not the problem. Like I said already, I'm running all the latest ones.
  
  PCI Latency is not the issue. I've already installed PowerStrip and tinkered extensively with all the PCI Latency settings for the TV capture card, the video card, and the SATA-RAID on-board controller, and it makes no difference whatsoever.
  
  Other timing/bus/caching/spread-sprectrum settings in the motherboard BIOS are not the problem. I've already tinkered with all of these, one at a time and in just about every possible combination, and none of them have any effect on this issue.
  
  It's not a bad TV capture card or a bad graphics card. I've tried both in other machines and this issue doesn't manifest.
  
  It's not a problem of dust caked up on components inside the PC. As I just replaced the case and motherboard, I completely blew/dusted off all components before reassembling, so it's spotlessly clean.
  
  It's not bad RAM. I've already exhaustively run all my RAM through memtest86, memtest86+, and Prime95 torture tests, and it's all good.
  
  It's not a ground loop problem. If it were, I would see the problem at all times, not just during hard drive activity.
  
  It's not an overclocking/overheating problem. I do not overclock my system at all, and I have 3 case fans, a CPU fan, and a PSU fan all running, and I monitor all the fan speeds and temperature levels with Motherboard Monitor and it's all fine.
  
  It's not a problem of saturating the bandwidth of the PCI bus or SATA bus, because I see this problem when simply watching (not recording) TV via my capture card, and doing nothing particularly disk intensive (for instance, just running a "dir /s" at the root of my C: drive while watching TV is enough to trigger the interference).
  
  It's not an issue of sloppy cabling running everywhere inside the case. All my cabling is neat and tidy using tie wraps, with power cables kept away from other signal-carrying cables, and signal-carrying cables (such as the SATA cables) kept away from each other and from the PCI cards.

Factors I've not yet ruled out, but intend to test:

• Hard drive positioning. The Lian-Li PC60 case has an internal hard drive cage that by default mounts the drives vertically. I suppose it's possible that this generates a plane of magnetic or EM interference which is then 90-degrees different from the PCI cards, and that might cause a problem. In my old case these same drives were mounted horizontally, in line with the plane of the PCI cards.
  
  I haven't yet narrowed down my repro to see if it only happens with disk activity of the SATA array or whether it also happens with disk activity of the IDE drive also. I'll have to format the IDE drive and copy some files to it so I can test with disk activity on it.
  
  I haven't yet tried puting ferrite rings/beads along signal-carrying cables inside the case. I suppose I could try putting ferrite rings around the SATA cables, for instance, to see if it cuts down on interference.
  
  I suppose it could be a bad motherboard, but I really doubt it. It's brand new, has been properly treated (in a static-safe manner) since I bought it, and it exhibits no other problems.

I'm looking for any other substantive troubleshooting ideas anyone here has to offer. If you've struggled with your internal TV capture card picking up interference from hard drives, and you've managed to solve the problem to an acceptable point, I'm interested in hearing your experiences.

And please, newbies with no real experience on this issue, or with nothing more insightful to offer than "make sure you're running the latest drivers" or "it must be interlacing", stay off this thread. You aren't helping.

[SatStorm]

No much to do... Did you try "RF stoppers"? Some times those magnets do help on this.

[Richard_G]

There have been a number of people reporting this type of problem. The one common factor has been nVidia chipset motherboards with SATA hard drives. Some have found that different (not necessarily the latest) chipset drivers have cured it. If you search the forums you may find the answer to the problem, you aren't the first and almost certainly won't be the last to suffer.

That's why I'm sticking with VIA chipset and UDMA hard drives for video work!

[nic2k4]

You've pretty much covered everything, you could try running with the HD's out of the case, try different positions, see if anything changes. Also, disconnect any drives you don't need, then reconnect them one at a time (could be just one bad drive), you'll have to break the array.

Otherwise, you've answered your own question " It's not a bad TV capture card or a bad graphics card. I've tried both in other machines and this issue doesn't manifest.

It's not a problem of dust caked up on components inside the PC. As I just replaced the case and motherboard...", points to the motherboard not cooperating.

[c0d3h4x0r]

There have been a number of people reporting this type of problem. The one common factor has been nVidia chipset motherboards with SATA hard drives. Some have found that different (not necessarily the latest) chipset drivers have cured it. If you search the forums you may find the answer to the problem, you aren't the first and almost certainly won't be the last to suffer.

That's why I'm sticking with VIA chipset and UDMA hard drives for video work!

You say other people have reported this issue with nVidia chipsets and SATA hard drives. Can you point me to such reports or any particular threads where the issue is being discussed? I've already searched the VideoHelp forums and Google exhaustively and haven't been able to find anything about this.

I'm willing to throw money at an external, USB 2.0-based TV capture unit if that's what it takes to get TV capture working correctly again with my new motherboard... but I don't want to throw money at a stab in the dark. I would like to have some reasonable level of confidence that such a purchase would indeed solve the problem, so I would like to discuss it with other people who have had the same issue.

[Richard_G]

Took some finding but I knew I'd seen them somewhere
see here
https://www.videohelp.com/forum/viewtopic.php?t=200093
and here
https://www.videohelp.com/forum/viewtopic.php?t=189059

[SatStorm]

I read that too.
There is an issue with the first generation controlers of the SATA drivers in general (not only with nvidia chipsets).
Later SATA controlers fixed this, but we simply don't know what is "cured" and what is not!

[jlietz]

Total stab in the dark...
Did you use the ATI card in your previous system? Does this card receive power directly from the power supply? Perhaps it is a problem with the power supply giving the video card "noisy" power. I'm not an expert in this area, but perhaps what you are seeing is similar to what happens to a TV when someone else in the house turns on a blender. In this case, could the additional power draw of the active hard drive be introducing the noise?

[Soopafresh]

Could the case be acting like an antenna ?

[SLK001]

Is this interference visible only on the screen preview, or is it also present in the recorded capture?

[edsmith77]

There is one other possibility that has not yet been discussed. That is low signal strength on the cable system such that external suprious noise becomes a factor in overall signal strength. Which would be masked or not as noticeable with a higher strength signal.
Loose coax fittings not the part that fits onto the chassis connector but the part that connects to the cable itself such that during hard disk activiy the additional vibration (resonant freg) being such that it adds to the cable signal.
Aside from these two not being contributors, I would tend to belive that a hard disk failure is in your future.

My trouble shooting mantra is that when the obvious is not the culprit then go with the dammned obvious.
Good Luck.

[ronnylov]

Instead of using the internal tuner on the capture card you can try connecting the antenna cable to a VCR and connect the VCR to the capture card via composite or S-VIDEO cable and see if this interference is still there.

By the way this is the method I use because the quality of the TV-tuner in my JVC SVHS VCR is much better than the quality of the tuner in the capture card (or maybe it is less affected by interference because it is not inside a computer).

[Richard_G]

As it seems everyone else is missing the point and you have already proved that the interference is caused by hard drive activity, you need to do something about that. Try the fix in the previous thread first. If that doesn't work, try disabling the RAID array and only use a single SATA drive and try using just the IDE drive.

That should narrow the cause down to SATA controller or RAID controller.

[SCDVD]

One possibility is the power supply. One area where power supply manufactures cut cost is in the filtering in their power supplies. If you have a "sweet deal" power supply (aka cheap -- <most of the case makers use the cheapest supplies they can find>) it may not be adequately filtering the electrical noise generated by the hard disc when it is active. I had a similar problem that went away when I put a real power supply in my system.

[ozymango]

The problem:

While viewing or recording from TV Tuner via coax cable jack, notable interference appears on various channels ...

It's not a ground loop problem. If it were, I would see the problem at all times, not just during hard drive activity.

Where's edDV when you need him?

My experience is with RF interference in VCRs, not computers, but there's a lot of overlap between the two so let's try some stuff.

First, you say the problem appears during "hard drive activity"? What exactly do you mean by that? You power your computer on, your hard drives are "active" right away, so is it that you notice some kind of visual pulse in the interference/noise when you hear the drive heads "moving"?
If so, then there really is no need to troubleshoot any farther because you've ID'd the problem -- radio interference coming from the hard drive(s). And you could have a ground loop problem still, it's not necessarily an "all or nothing" thing when it comes to RF interference. A lot of factors come into play here.

Major starter -- you replaced both the mother board and the case, I read, and for troubleshooting you only want to futz with one piece at a time, because either one/or both could be contributing to the problem. A very slight shift in hard drive orientation can cause all sorts of changes.

Frankly my hunch (based on lots of experience with this sort of problem with VCRs) is some combination of weak signal and not-fully-shielded tuner in your capture card. I know you said Comcast came out and did the full service treatment but I've had them do that for some RF issues in the past and they don't always necessarily have the most experience technicians in the universe (no offense intended!).

And putting a tuner (a very RF sensitive piece of equipment) in a PC is like putting an ice cube in the oven -- unless you pack it very carefully, it's a really hostile environment for RF signals. So yeah, anything to get the tuner away from the computer is probably a good idea.

Because image quality is obviously very important to you and you've tried and are willing to try tons of things, I definitely think you should look into external tuner capture device (usb2 or firewire, not usb1), or (as somebody else wrote) use an external VCR or other tuner and plug that into your capture card. But if you're looking for maximum tuner control (from the PC) you'll probably want the former, not that latter. Somebody else may have better luck with an internal tuner but there are so many freakin' variables that can affect your signal, I personally like to "isolate" the tuner from all my other hardware as much as possible. My own setup is various VCR tuners plugged into my capture card -- the image is better on some channels on one VCR than the other, so depending if I wanna record "CSI" or "Monster House," I may use one or the other.

Good luck!

[c0d3h4x0r]

Thanks everyone for the ideas and posts.

I returned the ASUS A7N8X-E Deluxe (nForce2-based) motherboard and instead got an MSI KT6V that utilizes the newer VIA VT8237 Southbridge with integrated SATA-RAID. I did a LOT of reading up on the subject of onboard SATA-controllers over the last 24 hours and it turns out that nearly all non-VIA socket A boards suffer from a bad design choice (the SATA controller chip, typically Promise or Silicon Image/CMD, rides directly on the PCI bus, consuming PCI bandwidth).

I'm happy with the new motherboard. Unfortunately, my TV card seems to have been permanently damaged by the time it spent on the ASUS board, because now no matter which computer I plug it into, the computer fails to boot or issues "Plug & Play Device Error" POST messages on bootup. So I went to the shop this morning and got a new Hauppage/WinTV PCI capture card which I will try to install tonight or tomorrow. I'll post back with results from all this when I have anything useful to report.

Oh, and to those who suggested faulty coax or poor signal strength on the cable line, you didn't read my original post very carefully, did you? I said the cabling is all good and I've already got a signal amplifier, and I never had the interference on the original motherboard using the same cabling.

To whoever suggested a possible PSU problem -- good idea, but I should have listed that in the "already ruled out" list. I'm running a 450-Watt Antec PSU and I monitor all my voltages and it's fine.

[ozymango]

Oh, and to those who suggested faulty coax or poor signal strength on the cable line, you didn't read my original post very carefully, did you? I said the cabling is all good and I've already got a signal amplifier, and I never had the interference on the original motherboard using the same cabling.

And here's to your new tuner card working perfectly and not giving you any problems -- 'cause if it doesn't, you may find yourself right back here with more non-careful readers spending their time trying to be of some assistance. Too bad for you, I guess. 8)

[c0d3h4x0r]

Bought a new Hauppauge WinTV-PVR-250 card and installed it with the VIA KT6V motherboard, with Windows installed on the SATA-RAID-1 (mirror) array as the C: drive... all problems are solved. I can watch/record video smoothly, no interference or choppiness even with massive hard disk activity (i.e. reads and writes, not idling).

A number of factors probably combined here to solve my problem:

• This new TV card has hardware-based MPEG-II encoding whereas my AverTV Stereo did not and had to rely on CPU horsepower and a software codec for encoding.
  
  My AverTV Stereo card might have been gradually dying or partially damaged all along... no way to know for sure.
  
  This motherboard utilizes the VIA VT8237 Southbridge chip, which integrates the SATA-RAID controller and most other integrated devices (audio, ethernet, etc) into the Southbridge, completely bypassing the PCI bus, leaving PCI cards free to smoothly operate even while all the integrated devices are in use.

As for the attitude I exhibited in my last post here: I'm not annoyed for people trying to help. On the contrary, I'm thankful for it. But it is still disrespectful for someone to reply without first completely reading what I took so much time and effort to write. I wasn't dilligent in my first post just because I like typing; I did it to avoid wasting everyone's time with an unhelpful complaint. If I had just posted, "My system suxors, its broke d00d, help me!!!" you would have been annoyed too, I'm sure. So I'm really just asking for the same level of dilligence in return.

Richard_G, thanks for the helpful info and links about SATA-RAID and nForce2 boards. Based on further research I did from that starting point, I found out this design flaw isn't specific to nForce2-based boards, but is in fact pretty common to all earlier Socket-A boards that do not utilize a Southbridge chip that has SATA-RAID integrated onto it. VIA-based boards that utiltize the VT8237 Southbridge, or nForce2-based boards that utilize the MCP2 Southbridge, should not experience this issue. On the other hand, VIA-based boards that utilize the VT8235 Southbridge, or nForce2-based boards that utilize the original MCP Southbridge, will probably exhibit this issue to some degree or another.

[ozymango]

As for the attitude I exhibited in my last post here: I'm not annoyed for people trying to help. On the contrary, I'm thankful for it. But it is still disrespectful for someone to reply without first completely reading what I took so much time and effort to write. I wasn't dilligent in my first post just because I like typing; I did it to avoid wasting everyone's time with an unhelpful complaint.

Fair enough (and I for one appreciate your diligence). But at the same time, I don't agree that it was "disrespectful" of those who responded by saying it might be a problem that you had already "eliminated." This isn't to say you were wrong, just that sometimes we (all of us) might come to certain conclusions based on certain info, and somebody else may catch something we didn't.

Related context: You initially wrote that the problem had nothing to do with your video capture card, based on your evidence. But then you wrote afterwards that maybe the video capture card could have had some issues that you weren't aware of.

My own response to you about RF interference possibly being something to consider (which you had seemed to eliminate, based on your experience) was based on my experience with strange RF interference problems that seemed to be coming from one place -- but weren't. So it's not that I wasn't reading you carefully and I'm guessing that most others also read you carefully -- but they made some tentative conclusions based on their experience. Not a mark of disrespect against you or your experience, just mentioning some possibilites that you may not be aware of.

Also this is a free and public board, and in the spirit of a true democracy I personally find that depite any pleas I have from people here to answer me in a certain way, at a certain time, with certain information ... it's kind of catch-as-catch-can, ya know? Very often you get great, solid advice, and equally often you get all kinds of off-the-cuff or whatever reactions, some helpful, some not, some amusing, some not, but if we play nice with each other we (hopefully!) are more likely to get the actual help we need in the future -- everybody has on-days and off-days, and if somebody offers me a "not useful" piece of advice or commentary, it's pretty easy to just ignore/skip it and go on to the next. Assuming we're not getting 42,000 responses! :P

And last but not least, and absolutly no offense to you (or anybody else), one thing I encounter a lot in my own experiences as a computer tech and my past experience as a car/motorcycle repair person -- there are a lot of people out there who "know" what the problem is as to why something is working or isn't working in a certain way, and then it turns out that they were ... mistaken. To paraphrase Mark Twain (I believe) -- "It's not what people don't know that hurts 'em. It's what they know, that just ain't so."

This goes for me, too!