Hi,
I have a problem with my sound card it starts producing high pitched noise as soon as it gets power (not software issue) it can be heard inside pc case but whats worse its that this sound transfers to headphones (all outputs), I think it might be a dying capacitor but i have no idea how to chceck which one is it. I've tried with different psu / outside pc case so its not interference issue.
Recorded noise: https://www.dropbox.com/s/p6z1foou40vwqz7/noise.wav?dl=0
I have a problem with my sound card it starts producing high pitched noise as soon as it gets power (not software issue) it can be heard inside pc case but whats worse its that this sound transfers to headphones (all outputs), I think it might be a dying capacitor but i have no idea how to chceck which one is it. I've tried with different psu / outside pc case so its not interference issue.
Recorded noise: https://www.dropbox.com/s/p6z1foou40vwqz7/noise.wav?dl=0
have you read my post? i said its not a driver issue, noise starts as soon as the card gets power it doesnt even need to be in pci slot just when the additional power connector is connected..
The DX supposedly uses solid (polymer) electrolytic caps, I wouldn't expect those to go bad so easily unless you got unlucky. But knowing Murphy, who knows.
The most likely component for physical vibration would be L25, the DC/DC converter inductor. A small microphone for probing around would be handy, or one of these DIY listening devices with a mic and headphone amplifier.
Here's what they seem to be doing in the power section, looking at pcitures of the card:
There is a DC/DC converter employing an ITE CAT7105 chip, I'm guessing either for digital supply (+3.3 V or thereabouts) or for converter supply (+5V) since the chip is advertised as a high-efficiency downconverter. L25 is the choke, D6 the rectifier, output filtering would have to be nearby. There are a few compensation-related components, possibly the two guys right underneath the chip. The converter is supposed to be operating at a fixed 410 kHz, but PWM being what it is, it certainly is not impossible that audible tones can be generated depending on output loading. Those certainly should not be ending up in the output voltage though...
There also is an 78M08 linear regulator, presumably used for analog supply.
Both of these seem to be fed from +12V, at least the underside of the card shows two separate SMD fuses.
The Xonar D1 used the same kind of setup, actually.
What I still don't quite get is how they're avoiding ground loops on a card like this (which tends to be another option for unwanted noise to come in). They've got ground coming in over the floppy Molex, PCIe (presumably), and slot bracket, and I don't see how you'd be accomodating all three of these. This actually applies to all sound cards with extra supply connectors. Maybe they kinda "brute-forced" it with a ground plane?
I would have to poke around with a multimeter (and suitable probes) to say anything more, sorry.
The most likely component for physical vibration would be L25, the DC/DC converter inductor. A small microphone for probing around would be handy, or one of these DIY listening devices with a mic and headphone amplifier.
Here's what they seem to be doing in the power section, looking at pcitures of the card:
There is a DC/DC converter employing an ITE CAT7105 chip, I'm guessing either for digital supply (+3.3 V or thereabouts) or for converter supply (+5V) since the chip is advertised as a high-efficiency downconverter. L25 is the choke, D6 the rectifier, output filtering would have to be nearby. There are a few compensation-related components, possibly the two guys right underneath the chip. The converter is supposed to be operating at a fixed 410 kHz, but PWM being what it is, it certainly is not impossible that audible tones can be generated depending on output loading. Those certainly should not be ending up in the output voltage though...
There also is an 78M08 linear regulator, presumably used for analog supply.
Both of these seem to be fed from +12V, at least the underside of the card shows two separate SMD fuses.
The Xonar D1 used the same kind of setup, actually.
What I still don't quite get is how they're avoiding ground loops on a card like this (which tends to be another option for unwanted noise to come in). They've got ground coming in over the floppy Molex, PCIe (presumably), and slot bracket, and I don't see how you'd be accomodating all three of these. This actually applies to all sound cards with extra supply connectors. Maybe they kinda "brute-forced" it with a ground plane?
I would have to poke around with a multimeter (and suitable probes) to say anything more, sorry.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.