How does an attenuator impact on a system ?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey all.

Wasn't really sure where to post, so I figured here's as good a place as any.

I understand that basic concepts of impedence, and that you want the ouput impedence of the source to be significantly lower than the input impedence of the next component.

Now, suppose I have a preamp with an output impedence of 100 ohms, and an amplifier with an imput impedence of 100k ohms. Then, I whack a 100k potentiometer between them. Since the pot is in parrallel with the two devices, have I essentially turned that 100 ohm into 99.9 ohm (or unchanged), and the 100k ohm into 50k (halved it !)?

What's the mathematics in chosing a pot given an output and input load ?
 
AX tech editor
Joined 2002
Paid Member
lordvader said:
Hey all.

Wasn't really sure where to post, so I figured here's as good a place as any.

I understand that basic concepts of impedence, and that you want the ouput impedence of the source to be significantly lower than the input impedence of the next component.

Now, suppose I have a preamp with an output impedence of 100 ohms, and an amplifier with an imput impedence of 100k ohms. Then, I whack a 100k potentiometer between them. Since the pot is in parrallel with the two devices, have I essentially turned that 100 ohm into 99.9 ohm (or unchanged), and the 100k ohm into 50k (halved it !)?

What's the mathematics in chosing a pot given an output and input load ?


Yes, in the max level setting. As Richieboy noted, the values vary with the wiper setting. Assuming that the Zout of the source and the Zin of the amp are linear (resistive), not varying with level, freq, etc (and in a competently designed component that will be the case), the only effect of those varying impedances will be that the attenuation is not exactly what you would calculate with zero Zout and infinite Zin.

For example, with zero Zout and infinite Zin, if you put the wiper half way, you get exactly half the level. With 100Ohms Zout and 100k Zin, that will be a bit less than half level, but this is a non-issue, if you pot is at least 10 times the Zout, and not more than the Zin, preferably lower. In your case, 10 to 20k pot would be OK.

Jan Didden
 
Hi,
I would select a pot at or below Zin/10.
This keeps the source impedance presented to the following amp @ <Zin/40.
Take your Zin=100k and pot=10k then Rs<=2k5, but this varies down to near 0r0. The variation in source resistance causes changes in the power amplifier.
As usual, the lower the Rs, the less the power amp is affected.
If your pre-amp/source can drive a lower value pot then go down to Zin/100 i.e. 1k0 and Rs then varies from 0r0 to 250r (much better).
Caution:- a 1k0 load driven at upto 2V requires the source to send peak currents upto 3mApk. Some (many?) sources will not play well at this level of output current.
 
Hi,
the source resistance seen by the receiver depends on the position of the pot wiper.

when at the bottom (=zero sound) Rs=0
when at the top (=zero attenuation) Rs=potvalue//Rs source~=Rs source.
When between these positions the Rupper and Rlower set the source resistance.

If the wiper is in between then the receiver sees Rupper and Rlower, effectively in parallel.
at -20db Rupper =10k-1k=9k, Rlower =1k,
Rs=1k//[9k+Rs source]~=900r (if Rs source=100r, Zin=100k, then attenuation = -20.16db).
As the wiper moves towards max attenuation, Rs falls.
As the wiper moves towards -6db Rs rises.
At -6db position, Rupper=Rlower =5k. Rs=5k//[5k+Rs source]~5k//5k1=2k52, or just about Zin/40. (attenuation = -6.32db)
as the wiper moves from -6db to -0db Rs falls again.
Note that Rs is seen by the amplifier and the HF filter that should be fitted will vary in turn over frequency as Rs varies. -0db and -60db present a low Rs and the filter is set very high letting through a lot of unecessary crud. At -6db, the high Rs sets the frequency at it's lowest and chops off the upper audio band. The pot and amplifier MUST be designed together to get a passive volume control to work near effectively.
Most users just fit to the Zin/10 formula and usually get it wrong i.e. variable filter with volume. For those who insist on using DC coupled amplifiers then there is the variable output offset to contend with (a DC servo is designed to compensate for amplifier offset, NOT for variable input offset).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.