radio interference from radeon?

[Harabecw]

Hello! I haven't encountered this problem with my Ti4200...
when shutdown, everything on the radio is clear. once I start up the comp, I get some static. I assume it is from the radeon because the static noice changes when the monitor is either off(just it, the computer is not in standby) or I'm changing frequencies(after playing a game and going back to desktop).

Any idea why this happens and how to solve it?

 

[vss1980]

Firstly, if it is coming from the card you may wanna get that changed because it is clearly not meeting the specifications laid out for it.

But I doubt its from the Radeon card specifically. Its more likely to be coming from interference generated by the monitor because (A) in relative terms the Radeon is operating at a very low power level and its transmission medium is the cable going from the card to the monitor. Although there may be RF leakage it I am doubtful it would be that powerful. And (B) once the graphics signal hits the monitor it will be processed and amplified in various ways so that the image can be drawn which means running a high frequency to a bunch of magnetic coils wrapped around the monitor to deflect the beam. I see this as the most likely source.

Are you using different refresh rates with the card compared to those used on the ti4200? If so, set everything the same and see if the problem is still there or quickly dig out your old card even just to make sure it goes away with the ti4200 and then comes back when you put the radeon in.

Have you actually moved the radio closer to the monitor or computer to see where the source is? The only real solution is to move the radio further away from the source, with the second stupidly extreme option being to shroud your computer kit in a metal case that blocks out EMI (which I doubt is really practical).

 

[Harabecw]

using the same 85hrz in windows & games.
this particular station uses a relatively weak signal lately. we get interference at work when we use coronas in the printing machines (if you know what I mean
Don't have the 4200 anymore...oh well.

EDIT: get static at both 60hrz and 85hrz. 85 gives more but not a lot of difference. the radio & monitor have been in the same place for about 5 years. btw, its a radeon 9700.

 

[vss1980]

Just clarify one thing:

"the static noice changes when the monitor is either off(just it, the computer is not in standby) or I'm changing frequencies(after playing a game and going back to desktop)"

When you turn the monitor off, does the noise go away or just deviate a little?

 

[ScottMac]

More likely it's your power supply.

If you're talking AM radio, and it's ~890Khz (i.e., WLS AM in Chicago), then it's 99% chance it's your power supply.

I had one that (standard with the case) that killed AM 890 for a hundred feet or more. I bought a new supply (from the same place I bought the case) .....same thing. I bought another supply (Antec) and the noise was gone.

The variance could be load related.

I agree that another likely culprit is the monitor (poor / bad / defective shelding on the cable or case).

The easy way to check is to move the radio around the case & monitor and see how the noise changes.

If you find that the monitor cosr is leaky, you can get Ferrite Beads at Radio Shack or other component electronic stores.

Good Luck

Scott

 

[Peter]

Well you'd have to identify the exact frequency of the unwanted noise emission, find out what component in the machine works at this frequency (or a fraction of it), and eliminate the problem.

Doing this properly requires an EMI lab and probably a redesign of the culprit piece of hardware.
(This is one of the things that separates complete, branded systems from DIY attempts - at least here in Europe, as a system builder you HAVE to certify the entire assembly for EMI compliance ... or at least claim compliance and stand in for it if proven wrong.)

 

[Peter]

End users might want to try enabling Spread Spectrum clocking to flatten out the peaks in the EMI spectrum, make sure the case is properly closed (no missing slot brackets, lid on, no silly fan cutouts or glass windows). Cables with a ferrite bead near the connectors also help shield EMI.
Conductive copper tape in the right places also works wonders sometimes - but then again you'd need the EMI lab equipment to figure out where these spots are.

 

[Harabecw]

when I turn the monitor off the noise goes away, so I don't think it is the power supply.
The station is at 91.80MHz. I live in Israel btw
I'll try to find a new cable for the monitor and see how that goes.