I have my DAC (output 75 ohms) going to my preamp Ayre K-5xeMP Input Impedance of 20kohm per channel. I have my sound processor going to my Ayre K-5xeMP. Due to the location in my room of my speakers(I'm not able to move) the balance is off in my main listening position. The sound processor adjusts the balance so I get a nice center image with movies. I would like to acquire the same thing with my DAC. The sound processor calculation says it's 1.5 dB off. I was planning on taking a pair XLS cables (to keep length the same) and cut the one where it's off and add some resistors to decrease the voltage so that the it matches. I was looking at the goldpt attenuator site and noticed that they have different ones 10K, 20K, 25K, 50K, 100K, 250K. Not sure what that means.
How To Get More Usable Steps
R2 up to 6 DB up to -4 db. So I'm thinking I don't need to worry about that.
So my question if I just get a few resistors and make a voltage divider with R2 as 75K So what should I use for R1 values. Do I base my calculation on 20K of the pre-amp for attenuation?
Also it's balance(xls) so I need double of everything correct?
How To Get More Usable Steps
R2 up to 6 DB up to -4 db. So I'm thinking I don't need to worry about that.
So my question if I just get a few resistors and make a voltage divider with R2 as 75K So what should I use for R1 values. Do I base my calculation on 20K of the pre-amp for attenuation?
Also it's balance(xls) so I need double of everything correct?
You only need a series resistor before the louder preamp input. It will work against the
20k input impedance. To reduce the signal by 1.5dB, use the equation (ignoring the 75R):
-1.5 = 20 x log (20/(R+20)) so R = 3.77k, then use a standard 3.74 1% type.
For balanced system, use a resistor of half that value (1.87k) in series
with each signal polarity in the louder channel.
Last edited:
You need to use half that value for each signal polarity to get the same attenuation
as for the single ended case. A 1/4W metal film resistor is appropriate here,
and is best placed near the preamp input.
Take simple case. 2:1 attenuation. 1K load.
For unbalanced, a 1K in the cable against the 1K load gives the desired 2:1 division.
For balanced, ideally you want two equal resistors, one each leg. They should still add-up to 1K, to get the desired division. So this case is two 500r resistors in the cable.
Yes, in series with the signal [cable center conductor], but not connected to ground.
In this illustration, Z is the added resistor, and Y is the preamp input impedance.
The bottom conductor is the audio ground.
https://upload.wikimedia.org/wikipe.../1179px-Attenuator,_general_L-section.svg.png
Last edited:
one does not need to use E92 resistors.
A 3k6 (into 20k) will give 1.44dB of attenuation compared to 3k74 giving -1.48dB
One will struggle to measure the difference of ~0.05dB
Well, I received my resistors on Saturday, Spliced them into my XLS cables. They did absolutely nothing. I pulled out my trustee RadioShack sound meter and ran pink noise.
I purchased two 1.87K .1% TT Electronics 1/4 W Metal Film Resistors. If it was too low I expected at least 1/2 DB change.
Input impedance: 20k ohms unbalanced, 40k ohms balanced. So I'm guessing I should have used 40K and then split it in 1/2 for each side of balanced. So I'm calculating 6K34 split in 1/2 for each side for R1 and 212K for R2(which I don't really need). So 3K17 for each side. I would have expected some decrease in volume at 1k86 on each side.
I purchased two 1.87K .1% TT Electronics 1/4 W Metal Film Resistors. If it was too low I expected at least 1/2 DB change.
Input impedance: 20k ohms unbalanced, 40k ohms balanced. So I'm guessing I should have used 40K and then split it in 1/2 for each side of balanced. So I'm calculating 6K34 split in 1/2 for each side for R1 and 212K for R2(which I don't really need). So 3K17 for each side. I would have expected some decrease in volume at 1k86 on each side.
My experience of using balance controls is very limited, but that does not stop me having an opinion.
I would be hard pressed to notice a 0.5dB imbalance on a pure dual mono signal coming from a pair of speakers.
With stereo I would find it even more difficult.
But why did you buy 0.1% resistor tolerance?
Post9 gives details and the 3.8% difference between 3k74 and 3k6 results in ~0.05dB change.
And also confirms that 3k6 or 3k74 gives the attenuation you are looking for.
You could even use 3k9 1% metal film 50ppm/C 600mW from Farnell or Rapid @ ~3p each. (yes, they have massively increased the cost of axial resistors).
Last edited:
From my understanding a balanced at the other end the inverse signal on each line are compared and the difference is considered error. I would think you would want the resistors to be as close as possible so each side is pretty close. That is why I chose 0.1%. I was only buying couple of resistors so they were only a $1 to $2. Shipping was more then the parts.
one does not need to use E92 resistors.
A 3k6 (into 20k) will give 1.44dB of attenuation compared to 3k74 giving -1.48dB
One will struggle to measure the difference of ~0.05dB
This was into 20K. I think I really needed into 40K. 20K for each side of the balance connection. So the values should have been 3k74 for each side. I split the 3K74 for each side and purchased 1k87. But I would have thought I would have gotten some decrease in DB that I could see on my Radio Shack sound meter.
__________________
one does not need to use E92 resistors.
A 3k6 (into 20k) will give 1.44dB of attenuation compared to 3k74 giving -1.48dB
One will struggle to measure the difference of ~0.05dB
This was into 20K. I think I really needed into 40K. 20K for each side of the balance connection. So the values should have been 3k74 for each side. I split the 3K74 for each side and purchased 1k87. But I would have thought I would have gotten some decrease in DB that I could see on my Radio Shack sound meter.
__________________
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.