Problems with channel imbalance. Possible culprits?"

[KurskKnyaz]

1. What is the decibel loss over a single foot of 8 ohm 14awg speaker wire?
2. Is the relationship of db/signal attenuation to a unit of wire distance linear or logarithmic?

 

[Born2bwire]

Originally posted by: KurskKnyaz
1. What is the decibel loss over a single foot of 8 ohm 14awg speaker wire?
2. Is the relationship of db/signal attenuation to a unit of wire distance linear or logarithmic?

1. What kind of cable? Either way, over a foot, practically nil.
2. In terms of dB, linear.

 

[krotchy]

I dont know that this question really makes sense to be honest, since modeling speaker wire as a transmission line is way more work than needed. Looking over the numbers, a 3dB loss (1/2 Power) would require a wire length of about 1600 feet, so knowing the dB relationship is pretty pointless.

The R of the speaker = 8 ohms (though this isnt truly valid as its impedance changes over frequency, but for simplicity its a perfect 8 ohm resistor)
The R of the wire = 2.525mOhm per foot according to wikipedia (Im lazy sue me), or .00252 ohms per foot, and you need to double that since every foot of wire conducts both directions.

Assuming an audio receiver than transmits 100 Watts perfectly to a speaker regardless of load impedance (I would figure most are Voltage limited not power limited in reality, but for simplicity constant power is nice)
With this case your power to the speaker comes down to a simple voltage divider where Speaker power % =100 * 8/(8+(2* .00252 * feet)). So 1000 feet of speaker wire = 61.3% Power to the speaker, or in decibels, you have a loss of -2.12dB. Using decibels in this case really tells you almost nothing IMO, and if you have a 1000 foot run of speaker wire, you might want to rethink your setup.

 

[Born2bwire]

Originally posted by: krotchy
I dont know that this question really makes sense to be honest, since modeling speaker wire as a transmission line is way more work than needed. Looking over the numbers, a 3dB loss (1/2 Power) would require a wire length of about 1600 feet, so knowing the dB relationship is pretty pointless.

The R of the speaker = 8 ohms (though this isnt truly valid as its impedance changes over frequency, but for simplicity its a perfect 8 ohm resistor)
The R of the wire = 2.525mOhm per foot according to wikipedia (Im lazy sue me), or .00252 ohms per foot, and you need to double that since every foot of wire conducts both directions.

Assuming a special receiver than transmits 100 Watts perfectly regardless of impedance (I wouild figure most are Voltage limited not power limited in reality, but for simplicity constant power is nice)
With this case your power to the speaker comes down to a simple voltage divider where Speaker power % =100 * 8/(8+(2* .00252 * feet)). So 1000 feet of speaker wire = 61.3% Power to the speaker, or in decibels, you have a loss of -2.12dB

A purely real characteristic impedance is lossless. You can't use 8 ohms to be the resistive loss of the cable, it's the root of the ratio of the inductance and capacitance. And besides, where's your engineering spirit? Any EM modeling suite should be able to easily find the wave numbers for the primary mode of a simple transmission line like in twisted pair or especially coaxial. ADS Momentum has a toolkit that does just that for you. This man came looking for a simple answer and by God we're going to obfuscate it to such an extent that he'll never bother us again.

 

[KurskKnyaz]

Thank you for your response. What I got out of the technical jargon is that cutting a speaker wire 2' shorter would not make no difference. The reason I'm asking is that I own a pair of speakers and the left channel is estimated (by ear, I set output to mono and attenuated the left channel in software until the sound seems to be centered) seems to be 5db louder. The left channel is the slave speaker so I figured that because I cut the cable shorter from the master (which contains the amp) this caused my left channel to be louder. I assumed that the circuitry in the master is designed to compensate for the signal having to travel a longer distance to the slave channel and when I cut the cable I lowered resistance so it became louder.

The next culprit would be an un-aligned stereo potentiometer. I took the master apart and tested the logarithmic potentiometer with a multimeter and the resistance is even on both channels. Does anyone have any idea what else could be causing this? The source is out of the question. This is a problem with the amp in the master speaker. The pot is fine, the soldering seems fine everywhere on the amp. Should I look to see if the connection from the amp to the crossover in the master channel is bad due to poor soldering (i didn't look at the soldering on the crossover). I'm sure I had this problem for a long time but I just noticed this now.

MODS: can you please change the tread title to "problems with channel imbalance. Possible culprits?"

 

[krotchy]

Originally posted by: Born2bwire

Originally posted by: krotchy
I dont know that this question really makes sense to be honest, since modeling speaker wire as a transmission line is way more work than needed. Looking over the numbers, a 3dB loss (1/2 Power) would require a wire length of about 1600 feet, so knowing the dB relationship is pretty pointless.

The R of the speaker = 8 ohms (though this isnt truly valid as its impedance changes over frequency, but for simplicity its a perfect 8 ohm resistor)
The R of the wire = 2.525mOhm per foot according to wikipedia (Im lazy sue me), or .00252 ohms per foot, and you need to double that since every foot of wire conducts both directions.

Assuming a special receiver than transmits 100 Watts perfectly regardless of impedance (I wouild figure most are Voltage limited not power limited in reality, but for simplicity constant power is nice)
With this case your power to the speaker comes down to a simple voltage divider where Speaker power % =100 * 8/(8+(2* .00252 * feet)). So 1000 feet of speaker wire = 61.3% Power to the speaker, or in decibels, you have a loss of -2.12dB

A purely real characteristic impedance is lossless. You can't use 8 ohms to be the resistive loss of the cable, it's the root of the ratio of the inductance and capacitance. And besides, where's your engineering spirit? Any EM modeling suite should be able to easily find the wave numbers for the primary mode of a simple transmission line like in twisted pair or especially coaxial. ADS Momentum has a toolkit that does just that for you. This man came looking for a simple answer and by God we're going to obfuscate it to such an extent that he'll never bother us again.

8 Ohms isnt the loss of the cable, it is the resistance of the speaker itself. I was just simplifying the problem to a voltage divider consisting of 2 resistors, the speaker and the wire.

 

[Muadib]

Originally posted by: KurskKnyaz
MODS: can you please change the tread title to "problems with channel imbalance. Possible culprits?"

You can change it yourself. Edit your first post in the thread.

As to your problem, are you using a pc, or is it a receiver? Did you swap your left & right speakers to rule them out?

 

[biggestmuff]

Originally posted by: KurskKnyaz
Thank you for your response. What I got out of the technical jargon is that cutting a speaker wire 2' shorter would not make no difference. The reason I'm asking is that I own a pair of speakers and the left channel is estimated (by ear, I set output to mono and attenuated the left channel in software until the sound seems to be centered) seems to be 5db louder. The left channel is the slave speaker so I figured that because I cut the cable shorter from the master (which contains the amp) this caused my left channel to be louder. I assumed that the circuitry in the master is designed to compensate for the signal having to travel a longer distance to the slave channel and when I cut the cable I lowered resistance so it became louder.

The next culprit would be an un-aligned stereo potentiometer. I took the master apart and tested the logarithmic potentiometer with a multimeter and the resistance is even on both channels. Does anyone have any idea what else could be causing this? The source is out of the question. This is a problem with the amp in the master speaker. The pot is fine, the soldering seems fine everywhere on the amp. Should I look to see if the connection from the amp to the crossover in the master channel is bad due to poor soldering (i didn't look at the soldering on the crossover). I'm sure I had this problem for a long time but I just noticed this now.

MODS: can you please change the tread title to "problems with channel imbalance. Possible culprits?"

Did you try adjusting the volume between each speaker? Hint: "balance" control.

What brand and model of speakers?

 

[KurskKnyaz]

Originally posted by: Muadib

Originally posted by: KurskKnyaz
MODS: can you please change the tread title to "problems with channel imbalance. Possible culprits?"

You can change it yourself. Edit your first post in the thread.

As to your problem, are you using a pc, or is it a receiver? Did you swap your left & right speakers to rule them out?

I am using a PC, the source is out of the question. I have swapped channels.

 

[KurskKnyaz]

There is no balance control on these speakers

 

[Muadib]

Are you certain there is no balance control for your soundcard? My old card had it, but my onboard sound doesn't. If yours doesn't, I think you are going to have to live with it.

 

[PurdueRy]

Windows has built in balance control. Most sound cards should support this. In XP its in the advanced speaker properties area.

 

[Muadib]

Originally posted by: PurdueRy
Windows has built in balance control. Most sound cards should support this. In XP its in the advanced speaker properties area.

You are correct. Vista has it too, as I just found it. It's under the levels tab though, which is why I didn't see it before.

 

[KurskKnyaz]

Originally posted by: Muadib
Are you certain there is no balance control for your soundcard? My old card had it, but my onboard sound doesn't. If yours doesn't, I think you are going to have to live with it.

There is balance control on my sound card. I used it to determine that the slave speaker was 5 db louder. I want to know what is causing the slave speaker to be 5db louder or master 5db quieter in the hardware. When I say hardware i mean circuitry within the master speaker.

 

[PurdueRy]

Originally posted by: KurskKnyaz

Originally posted by: Muadib
Are you certain there is no balance control for your soundcard? My old card had it, but my onboard sound doesn't. If yours doesn't, I think you are going to have to live with it.

There is balance control on my sound card. I used it to determine that the slave speaker was 5 db louder. I want to know what is causing the slave speaker to be 5db louder or master 5db quieter in the hardware. When I say hardware i mean circuitry within the master speaker.

What are you talking about when you say "slave" speaker?

Are they the exact same speaker? Have you tried swapping the speakers to see if the location is affecting the volume?

 

[Born2bwire]

Originally posted by: krotchy

Originally posted by: Born2bwire

Originally posted by: krotchy
I dont know that this question really makes sense to be honest, since modeling speaker wire as a transmission line is way more work than needed. Looking over the numbers, a 3dB loss (1/2 Power) would require a wire length of about 1600 feet, so knowing the dB relationship is pretty pointless.

The R of the speaker = 8 ohms (though this isnt truly valid as its impedance changes over frequency, but for simplicity its a perfect 8 ohm resistor)
The R of the wire = 2.525mOhm per foot according to wikipedia (Im lazy sue me), or .00252 ohms per foot, and you need to double that since every foot of wire conducts both directions.

Assuming a special receiver than transmits 100 Watts perfectly regardless of impedance (I wouild figure most are Voltage limited not power limited in reality, but for simplicity constant power is nice)
With this case your power to the speaker comes down to a simple voltage divider where Speaker power % =100 * 8/(8+(2* .00252 * feet)). So 1000 feet of speaker wire = 61.3% Power to the speaker, or in decibels, you have a loss of -2.12dB

A purely real characteristic impedance is lossless. You can't use 8 ohms to be the resistive loss of the cable, it's the root of the ratio of the inductance and capacitance. And besides, where's your engineering spirit? Any EM modeling suite should be able to easily find the wave numbers for the primary mode of a simple transmission line like in twisted pair or especially coaxial. ADS Momentum has a toolkit that does just that for you. This man came looking for a simple answer and by God we're going to obfuscate it to such an extent that he'll never bother us again.

8 Ohms isnt the loss of the cable, it is the resistance of the speaker itself. I was just simplifying the problem to a voltage divider consisting of 2 resistors, the speaker and the wire.

Ok, I missed that. Well crap, I even had a big ass response to all this too until I went back and reread what you said.