Help in choosing a potentiometer as a "Passive preamplifier"

[coldcurrents]

Am I likely to hear a difference between 10k and 50k inputs using 6ft cables?

 

[BigE]

I see red and green khozmos available.... is there a difference?

 

[canonken]

Thanks for this!!!

I have actually wondered about this, and wish it was 'real'. I feel better about needing a preamp despite playing with a super-simple system.

 

[ranshdow]

Am I likely to hear a difference between 10k and 50k inputs using 6ft cables?

Here are some RC filter calculations, using two different types of interconnects: a) a generic RCA cable (328pF/m) and b) a fancier RCA interconnect (82pF/m)



10k Z + 656 pF = 24.26kHz rolloff point (you may hear this)
50k Z + 656 pF = 4.85kHz rolloff (you will hear this)
10k Z + 164 pF = 97.05kHz rolloff point (you won't hear this)
50k Z + 164 pF = 19.41kHz rolloff (you may hear this)


Calculations can be done here: RC Low Pass Filter Calculator



Thread with measured capacitances of RCA cables: Measuring the capacitance of some RCA cables

 

[Bigun]

remember that a 50k pot doesn’t present 50k to the load, worse case it is 25k

 

[rayma]

Worst case for a 50k pot is 12.5k Thevenin equivalent source impedance.

 

[Bigun]

that’s what I said, just was using my British accent

 

[ranshdow]

Is there a simple explanation for the whole Thevenin equivalent source impedance concept, because I don't get it, at least as it applies to a pot in parallel between a source and a load.

 

[rayma]

Thevenin source impedance is a general concept, but it is simple in most audio circuits.
For a pot, short the input (top terminal of the pot) to ground. Then calculate the impedance
from the output node (wiper) to ground. That would be the top section of the pot in parallel
with the bottom section of the pot.

The maximum source impedance would be when the pot is set at the electrical center (both sections equal).
Then for a pot resistance equal to R, we have a maximum source impedance equal to (R/2) // (R/2) = R/4.
For a 50k pot, that is 12.5k maximum source impedance.

For the general (any circuit) case, set all sources equal to zero volts or amps.
Then calculate the impedance to ground for the desired network node.
This can also be done between a pair of network nodes instead of from one node to ground.

There is also a Thevenin equivalent source voltage or current. Think of a passive mixer, for example.

Here is a graph of the output impedance of a 10k pot. Notice that it peaks at -6dB (electrical half) output.

 

[ranshdow]

Ah, that makes sense to me now. Both halves of the pot are in-line and parallel simultaneously, just the dividing point moves.

 

[rongon]

Recalculating for the two pot values of 10k and 50k:

10k (Zout = 2.5k) + 656 pF = 97kHz rolloff point (you won't hear this)
50k (Zout = 12.5k) + 656 pF = 19.4kHz rolloff (*you may hear this*)
10k (Zout = 2.5k) + 164 pF = 388kHz rolloff point (no-one can hear this)
50k (Zout = 12.5k) + 164 pF = 77.6kHz rolloff (you won't hear this)

20k (Zout = 5k) + 656pF = 48.5kHz (you won't hear this)

A 20k pot might be the best compromise for a 'passive preamp'. It won't load down weak sources as much and it won't cause too much of a HF rolloff with reasonable length cables. No?

 

[DNic]

Hi, You should try the 10K/ 20K version of this. They sound exceedingly good for the price, way better that Alps pot, (except the Black Beauty, that is)
https://www.ebay.com/itm/2750716874...BfbHlDzeNw%3D%3D|clp:2334524|tkp:BFBMxObU48ZfCheers

 

[rayma]

Yes, I have several of those in 10k values. They are well made and sound good.
It's amazing they can make and sell them for that price with multiple fingers, no noise.
Too bad they only use thick film resistors, but are precision 1%.

 

[Shinjitsu Audio]

This is the best I have found:

DACT Type SMD Stepped Attenuator 21 step​

 

[mkane77g]

link no good^

 

[GM]

Stepped preamps

 

[mkane77g]

Thats plenty thanks

 

[ginetto61]

thanks for the very interesting thread I wonder if the fact that a power amp input is not a pure resistive load can have any influence
Some amps have also input capacitance Maybe not that low to be neglectable ?
Of course an active preamp with a robust output stage could drive almost any power amp

 

[rongon]

If you use a 50k ohm volume control, its maximum Zout will be 1/4 of that, or 50,000/4 = 12.5k ohms.
If the input impedance (Zin) of the amplifier is 10k ohms, this will be too high.

If you use a 10k ohm volume control, the max Zout of the that volume control is 2.5k ohms.
If the Zin of the amplifier is 10k ohms, then this is acceptable (although Zin of 25k ohms would be more optimal).

Yes, the input capacitance (Cin) of the load (amp input plus cable capacitance) is a concern.

If the Cin of the amp+interconnect cabling is 1000pF (not far-fetched) then that results in an RC lowpass filter of Zout in series with Cin in parallel with Zin.

The output impedance (Zout) of the volume control/potentiometer will combine with the Cin of the load to create an RC low pass filter.

Zout _____
          |
        Cin+Zin
          |
         GND

That will mean 2.5k ohms in series with 1000pF//10k ohms in parallel, which will introduce some loss of gain from the 1:4 Zout:Zin ratio, and a low pass filter that's -3dB at 63.7kHz.

So, you will need to find out what the resistances and capacitances are for the various bits you plan to use.

 

[Chris Daly]

thanks for the very interesting thread I wonder if the fact that a power amp input is not a pure resistive load can have any influence
Some amps have also input capacitance Maybe not that low to be neglectable ?
Of course an active preamp with a robust output stage could drive almost any power amp

As can a line level source, where power amp input sensitivity sensibly matches.