Skip to main content

hey everyone-

just picked up a pair active monitors, and am looking to build a simple passive volume control for them. Running 1/4" balanced TRS out of my interface into the monitors currently. Don't want to use the software master faders, as I'll lose the all important bit-depth of my digital audio when I have the fader dialed down...:mad:

I was thinking a dp12t rotary switch(maybe more positions if possible) with matched resistors for equal reduction on each channel, however, since the signal is balanced, with a positive and inverted line per channel, things get complicated, well maybe at least to my knowledge. Am I gonna need to convert to unbalanced first, introduce the switch, and then convert back to balanced? Please tell me there's a better way to do what I want...I'd like to keep my software master faders at unity as well as my monitors.

Comments

jimmys69 Sat, 06/25/2011 - 19:22

What interface are you using? Many get away without using balance monitor cables. As long as you are not crossing AC lines and there is no crazy interference going on, you should be fine with unbalanced. I would be more concerned with the rotary switch degrading the quality of your reference than noise. I would purchase a monitor controller before DIY. But that is me.

tele Sat, 06/25/2011 - 20:28

Thanks Jimmy!

I'm using an Echo Audiofire 4, which supports bal/unbal. I'd like to stick balanced. I trust my cables and it gives me peace of mind knowing what's going on between my DA and monitors. Maybe it's silly, but the less things I have to think about the better.

What about a simple stepped passive attenuator do you think would degrade my signal, using correctly matched metal film resistors? I'd like to think that something that simple if done right would be fine...

Many of the monitor controllers have features I don't need, and therefore not worth the price to me...however I did see a passive attenuator meant for my application for $60, the Nano Patch Plus, which does only what I want, passive attenuation. However this I believe uses a potentiometer, which my instincts tell me would be inferior to a switch, but I may very well be wrong. Even if the differences are minimal, I'd enjoy building my own, even if the price ends up the same.

Boswell Mon, 06/27/2011 - 02:54

The SM Pro Audio Nanopatch is a good little unit, and you don't need to worry that it uses a continuously-variable potentiometer. The inter-channel matching is good down to around 40dB - 50dB of attenuation, but can go off for high attenuations.

If you want to build something, Glassware Audio makes a [[url=http://[/URL]="http://glass-ware.s…"]stepped balanced attenuator kit[/]="http://glass-ware.s…"]stepped balanced attenuator kit[/]. It has a design flaw in that it's only accurate driving high-impedance loads, but you can use it perfectly well with standard monitor medium-impedance inputs if you are interested in achieving reproducable attenuation settings rather than being precise about what actual attenuation each setting represents.

tele Mon, 07/04/2011 - 10:15

I appreciate your help Boswell, I'm considering buying that attenuator kit.

Is there a way I could adjust the resistor values to be precise about what attenuation each setting represents? Anyone come across a good article explaining how I would derive the right values for attenuation driving a medium impedance load? I'd probably want to order the resistors separately anyways so that I can test/match them well.

Boswell Mon, 07/04/2011 - 10:27

In principle, yes, but it may come down to solving some equations or doing some spreadsheet work. I don't think you have specified what monitors you have, so I can't tell what input impedance you are dealing with. "Medium" could be anywhere from 2K to 30K Ohm, and a range like that would make a significant difference to the low attenuation resistor values.

tele Tue, 07/05/2011 - 09:13

Inputs are 10K (KRK R6G2). I'm willing to make a spreadsheet to derive values, but could use a push in the right direction as to which formulas to use.

Also, as I originally mentioned, attenuating both +/- signals with separate resistors seems like walking on thin ice, is it worth being concerned about? How closely matched should these resistors be to avoid problems?

Boswell Wed, 07/06/2011 - 10:19

I've done a quick Open Office spreadsheet for the calculations. Here is a text output of the result for 1K input resistors and a 10K load (all resistances in Ohms):

2xR1 are the series resistors
R2 the shunt resistor
R3 the input resistance of the driven device.
R1=1000
R3=1.00E+004

dB R2
1.6 open
05 3458.54
10 1019.22
15 452.14
20 227.27
25 120.61
30 65.74
35 36.34
40 20.24
45 11.32
50 6.35
55 3.56
60 2
65 1.13
70 0.63
75 0.36
80 0.2

tele Wed, 07/06/2011 - 20:09

Thanks for your help! Any chance I could get that spreadsheet, or even just see the formula relating R1, R2, R3(Z?), and dB?

Also, is this for a U-pad? switching out the shunt resistors for each attenuation value? Would this match with the PCB of the Glassware Audio stepped bal. atten. you suggested?

Boswell Thu, 07/07/2011 - 03:13

The quick spreadsheet I did was not aimed at the Glassware boards, for which I do not have any schematic details, although it could probably be adapted easily enough.

In the spreadsheet, the attenuations in dB were in cells A9..An, and the shunt resistor formulae were in the corresponding B cells. This is the formula for cell B9:
=(2*$A$3*$A$4/((10^(A9/20))-1))/($A$4-(2*$A$3/((10^(A9/20))-1)))

The input resistor values were in A3 and the load resistor (speakers) in A4.

Whether the shunt resistors are a ladder or individual values is up to you. I know the Glassware boards had a coarse-fine switching arrangement, presumably in which various series resistors formed the coarse values and the shunt resistors the fine values.