Playing MP3 From Laptop

I have a large collection of mp3's on my laptop and play them through my home stereo system (Harmon Kardon receiver) with a Y jack which I have to plug into my laptop headphone jack. The problem is a loud humming noise coming out of the floor speakers with this configuration.

Is there another way to connect the laptop to the stereo without getting the humming. I think the problem arises from the amplifier in the headphone jack.

Thanks

 

Does your laptop have an spdif connection? If not, you could always buy an external sound card.

 

Haha, I know the EXACT problem and the EXACT solution.

http://www.ebtechaudio.com/humxdes.html

Make sure your laptop is plugged into that and your problem will be gone.

 

Humming noises are usually cause by not grounding your installation. I would start by checking your cable/satellite coax connection and work your way to the rest of the cables.

 

Does the external sound card have a built in amplifier? If so, it would cause the same problem. If not, that might be the most economical solution.

I believe everything is grounded properly but I will double check that as well.

Thanks

 

Thanks for all the advice. I did find the solution that cost me a grand total of $18 + tax. For anyone else that has the same problem the solution is to buy a ground loop isolator. Can be purchased at Best Buy in their car audio installation dep't.

 

Yep, told you that was the solution. Go me.

 

New problem. Ground loop isolator works great at eliminating any humm. After installing however, I am pulling in a very faint radio station coming through my floor speakers. Doesn't matter if I put jacks in CD, Tape Monitor, etc., station comes in when hooked up to lap top. Very strange. Is there any type of filter to eliminate this?

 

BTW...thanks for pointing me in the right direction Warrior

 

Update......found a better solution to my problem (see above). After doing exhaustive research on the net I found a simple, cheap and effective solution.

To eliminate the hum all one needs to do is spend $.69 (that's 69 cents!!!!) on a plug attachment that converts a 3 prong plug to a 2 prong. By eliminating the center ground plug there is no ground loop formed. Purchase the adapter (usually grey in color) at any hardware store.

 

i'm having this problem, but i question what happens if you eliminate the ground? is it safe for the laptop or the user??

 

It is against electrical code to remove the ground.
The proper solution is to use an isolated data connection. An isolation transformer can work, as can an optical connection. In your case, the absolute best is S/PDIF that uses a laser and fiber.
To check if your sound card has S/PDIF, open a volume mixer application (I use KMix) and look for something labeled S/PDIF or IEC958.

And if your amp has a digital input and DSP, it most likely uses digital technology to amplify audio. Do not use the analog inputs since converting from digital to analog and back to digital will degrade the quality.

 

If you two haven't gotten it yet, welcome to the forums both star and realtor. And in the case of star, it's always nice seeing more Linux users around these parts.

Anyway, if you don't have an S/PDIF output on the laptop, the best and safest solution would be an external sound interface. Quality of the electronics doesn't really matter since it's all digital, but driver quality does, and having the option between optical and coaxial S/PDIF is always nice. This Zalman part looks pretty decent at first glance, but I've by no means researched it.

Just to provide a contrary view to star882's post, a lot of audiophiles complain about jitter with optical interconnects, and tend to prefer coaxial. I myself am using optical with my desktop just because I don't have a spare coax handy and haven't experienced any problems.

 

Fiber is superior to electrical wire if used correctly. Just make sure the ends are clean and avoid sharp bends. (Yes, it is more fragile, and yes, it costs more, but you get an extremely good isolated connection that is immune to EMI.) With long cable runs, it also requires fewer repeaters.
Also, fiber carries digital data so if it works, it works. To simulate a weak signal, simply loosen the connector. You may be surprised how loose it can be before the link drops. The noise level in a fiber is virtually constant (virtually 0 if connections are lightproof) so it's hard to get an intermittent connection that only works some of the time. However, a marginal signal can cause the link to suddenly stop working as the laser diode ages, although I have never encountered this in practice even with cheap diodes (used in very cheap motherboards) that are rated for even as little as 500 hours of operation.
BTW, an easy way to check the signal level of a fiber is to point one end at your hand (never look directly into a fiber) and look at the brightness of the spot. That only works for visible wavelengths. You'll need to use a signal meter for infrared. You can make one using a small solar cell and a multimeter that can read microamps.

 

You might be able to see the infared with a digital camera. Their CCD's pick up infrared light unlike the human eye, and the nonprofessional models often do not filter out infrared light.

 

Infrared fiber optic lasers are often deep infrared and might not show up very well on a regular digital camera, even one with the IR filter removed.

 

I can't say I'm well schooled in optical transmission, but here's a rip from an article I found here.

I don't know how much of current complaints with TOSlink are just audiophile bickering (the same audiophiles who paint the top of their CD's to improve sound quality, perhaps), and how much of an issue still exists. But it does make sense, with digital links you don't have to worry about signal degradation, but you'll always be fighting data quantity and time itself.

 

5MHz is too low for modern S/PDIF. At 192k samples per second, 24 bits per sample, and 2 channels, the data rate is about 9.2 Mbps. That does not include start and stop bits or ECC data. It would be roughly 10MHz, which is very slow for a fiber connection. Fiber connections easily go up to many GHz so 10MHz is very easy in comparison.
Most UART chips actually sample the signal at several times per bit to minimize error. They are very tolerant of jitter compared to what was used in the past. In any case, a loopback test could be done by reserializing the deserialized input and sending it back to the computer, which deserializes it, buffers it, and then reserializes it to send it back to the amp. Then the same data is sent through the loop many times and checked to make sure it does not change and also check for any ECC errors. (My amp actually has a port for that but I don't have the necessary cable and software to test with.)
But it definitely beats converting the data to analog and back to digital. Just like how DVI is better than VGA, S/PDIF is better than line level analog.

BTW, modern DSPs do a much, much better job of converting PCM to PWM than old DSPs. Some modern amplifiers also modulate the supply voltage for better resolution at low volumes. And of course, they're much, much more efficient than the old ones. (Some new amplifiers can attain as much as 90% efficiency at low volume as opposed to 30% or less for old amplifiers.)