About stepped attenuators (i.e. am i hearing things ?)

[gino1961]

Good morning everyone
Always looking for a definitive pre-line I made some kits with ok but not exciting results
I really like the sound developed in 3 dimensions (which I say 4 because a very good live recording can take you back in time)
Jokes aside... a friend of mine passed me a series type 20k log mechanical attenuator built using Holco resistors
I put it in a box with RCA ins and outs like a passive preamp and connected it in front of an old Bryston 0.4B preamp that I had around and ... my goodness.
By setting the preamp volume to maximum and controlling the level with the attenuator the sound became much clearer and more detailed
The difference between a knife that doesn't cut well and a scalpel
Question... is it autosuggestion or is there an abyss between attenuators and common potentiometers?
Unfortunately it's quite big and doesn't fit inside the case
So I was thinking about a DACT but they aren't cheap
Thanks for any kind advice

 

[somebodyelse]

Given the good measurements we've seen from headphone amps with cheap pots as an attenuator there's nothing to worry about with common potentiometers, except a bit of channel mismatch sometimes. Stepped attenuators will avoid this mismatch, but beyond that shouldn't be audibly or measurably different unless one of them is faulty. Since you seem capable of a bit of DIY you could make up a switch box to compare them. Levels should be matched better than 1% (multimeter, sine wave <400Hz should do) and arrange for the switching to be done without you knowing which you are listening to. This could be automated, or by someone else doing the switching without accidentally giving the game away. It's quite a bit of effort to go to to do it properly, but quite revealing if you've not tried before.

 

[gino1961]

Given the good measurements we've seen from headphone amps with cheap pots as an attenuator there's nothing to worry about with common potentiometers, except a bit of channel mismatch sometimes. Stepped attenuators will avoid this mismatch, but beyond that shouldn't be audibly or measurably different unless one of them is faulty. Since you seem capable of a bit of DIY you could make up a switch box to compare them. Levels should be matched better than 1% (multimeter, sine wave <400Hz should do) and arrange for the switching to be done without you knowing which you are listening to. This could be automated, or by someone else doing the switching without accidentally giving the game away. It's quite a bit of effort to go to to do it properly, but quite revealing if you've not tried before.

Hi thank you very much for your kind and valuable reply
You are right about measurements But with the same friend we compared the attenuator to a very decent Alps blue velvet 10K log pot a very common pot that can be found almost everywhere in preamps also not cheap
The attenuator was clearly better More details and also dynamics
It was difficult to go back to the pot after hearing this perfomance
I agree completely on the cheap pots unbalance

 

[somebodyelse]

The attenuator was clearly better More details and also dynamics
It was difficult to go back to the pot after hearing this perfomance

Now try it level matched and blind I say this having built attenuators with switch wafers and relays. No harm using them if you got them, and the Blue Velvet may be more reliable in the long term. May feel nicer too. I just can't kid myself that they sound different any more.

 

[gino1961]

Now try it level matched and blind I say this having built attenuators with switch wafers and relays. No harm using them if you got them, and the Blue Velvet may be more reliable in the long term. May feel nicer too. I just can't kid myself that they sound different any more.

Hi again Your words have convinced me Pot is the way After all i see these pots on very expensive preamps
Just one thing about impedance
Could a lower value volume pot be better choice ? like a 10K log instead of a 50-100k log ? i will put always a cap after the pot
as a reference the input unbalanced impedance of the very excellent Topping Pre90 line preamplifier is 10K from the specs

 

[RayDunzl]

I used a pair of 10K pots between the CD player and Amplifier for years, because that (and speakers) was all I had or needed.

Decided on a pair so I wouldn't be bothered by channel imbalance that I couldn't correct. They're audio taper, but not stepped.

No box, just some old curtain hanging hardware screwed to a wooden thing for setting hot food upon.

 

[somebodyelse]

It may or may not make a difference depending on the details - output impedance of source, input impedance following it, and capacitance of cable in between. For most modern hardware the output impedance will be low and input impedance high enough that there probably won't be a difference if the pot is near the following stage.

From what I remember the cap should go before the pot. This avoids any DC at the wiper/switch which could cause noise while changing volume.

 

[gino1961]

I used a pair of 10K pots between the CD player and Amplifier for years, because that (and speakers) was all I had or needed.

Decided on a pair so I wouldn't be bothered by channel imbalance that I couldn't correct. They're audio taper, but not stepped.

No box, just some old curtain hanging hardware screwed to a wooden thing for setting hot food upon.

View attachment 360397

hi thanks a lot for the valuable advice I tend to prefer a kind of universal preamp that can drive any power amp around
I have one with just 5kohm of input impedance and i wonder why so low actually
i see the line preamp as the corner stone of a audio system I would like to set on one and keep it for ever
Active preamps are more tricky but more universal than passive ones

 

[gino1961]

It may or may not make a difference depending on the details - output impedance of source, input impedance following it, and capacitance of cable in between. For most modern hardware the output impedance will be low and input impedance high enough that there probably won't be a difference if the pot is near the following stage.

Thanks i see I have another doubt about buffering the inputs Most preamps dont buffer the inputs No one tube preamp
for solid state i am watching some schematics and they have indeed a first buffer - volume - preamp stage
i prefer to stay as simple as possible like pot + buffer or pot+preamp stage

From what I remember the cap should go before the pot. This avoids any DC at the wiper/switch which could cause noise while changing volume.

i had this issue but then i was suggested to put a small resistor of 100ohm at the pot out before the dc blocking cap It seems to work
pot out - 100ohm - cap

 

[DVDdoug]

have one with just 5kohm of input impedance and i wonder why so low actually

Lower output impedance is better, up to a point, because it makes the output voltage independent of the load. i.e. The input impedance should be much greater than the output impedance. And lower impedance makes the connection more immune to electromagnetic noise pickup.

The output of a (active) preamp might be in the ballpark of 100 Ohms or less. But that doesn't mean it's designed to drive a 100 Ohm load. If they give you an output impedance spec it's likely the minimum recommend load impedance. Just like with power amps... It might be rated for 4 or 8-Ohms, but the actual output source impedance is usually a fraction of an Ohm and if there's a spec it's "damping factor".

 

[gino1961]

Lower output impedance is better, up to a point, because it makes the output voltage independent of the load. i.e. The input impedance should be much greater than the output impedance. And lower impedance makes the connection more immune to electromagnetic noise pickup.

The output of a (active) preamp might be in the ballpark of 100 Ohms or less. But that doesn't mean it's designed to drive a 100 Ohm load. If they give you an output impedance spec it's likely the minimum recommend load impedance. Just like with power amps... It might be rated for 4 or 8-Ohms, but the actual output source impedance is usually a fraction of an Ohm and if there's a spec it's "damping factor".

thanks a lot for the precious advice As i said i much prefer a line preamp able to drive any power amp around
then it must be solid state and low Z out like you say 100 Ohms or less

 

[AuricGoldfinger]

Hi gino1961,

Make recordings from the output of your preamp with and without the pot in front if it, then compare them after level matching in an audio editing suite.

Stepped attenuators make perfect sense when the goal is excellent channel matching and precise incremental attenuation from a strictly analog hardware-based solution. They are switches and can be excellent. I don’t begrudge people for choosing them, but the claims that get repeated online are problematic.

Let us put the stepped type aside for the moment and look at the Alps RK27, a conductive plastic pot that is affordable to most DIYers and is easily sourced. From Mouser or Digikey that is, not a knockoff from elsewhere.

One of the (many) audiophile claims that doesn’t make sense is the one that the Alps RK27 reportedly sounds “inferior” to other audiophile pots. The kicker here is the RK27 has possibly the best thermal coefficient next down from a stepped resistor attenuator among the supposedly superior sounding pots they claim to have compared it to. Carbon track pots are usually 700 to 1000 ppm, and cermets are often 500 ppm. The RK27’s thermal coefficient is about one-quarter to one-third that. Another type they get excited over are the TKD brand pots, which like Alps are good quality products with ppm of about 100, but cost 2 times that of the RK27.

Tests of the Alps RK27 place its noise + distortion at “pretty much immeasurable” by most analyzers.

 

[gino1961]

Hi thank you for the very valuable advice
When I did the test I am quite sure that the level was higher than usual
Usually what sounds louder seems better
It's true that very top 10kUSD preamps use Alps blue the best clue of their quality
I will live with it
They are much more convenient than more complex stepped attenuator
Speaking of tests I would like to buy a signal generator and a scope to play a little
I believe in square wave tests

 

[Doodski]

Hi thank you for the very valuable advice
When I did the test I am quite sure that the level was higher than usual
Usually what sounds louder seems better
It's true that very top 10kUSD preamps use Alps blue the best clue of their quality
I will live with it
They are much more convenient than more complex stepped attenuator
Speaking of tests I would like to buy a signal generator and a scope to play a little
I believe in square wave tests

OOch... be careful.... DC output de jour there in square waves input to audio amp...

 

[gino1961]

My interest is only in line preamps with low gain and tested alone
My long term search is for a very transparent solid state line preamp to become the corner stone of my system
I would like to use max 100mV test signals
SW is a very challenging signal
I have seen a video in which the sw test was able to show problems even in the BNC to BNC cable connecting the sw signal generator to the scope
It was a very cheap cable to be honest but I was impressed by the sensitivity of the test

 

[Doodski]

My interest is only in line preamps with low gain and tested alone
I would like to use max 100mV test signals
SW is a very challenging signal
I have seen a video in which the sw test was able to show problems even in the BNC to BNC cable connecting the generator to the scope
It was a very cheap cable to be honest but I was impressed by the sensitivity of the test

Yes the fundamental frequency of the square wave rising and falling edge is infinite if it is a 90 degree pure angle. BNC connectors will have limits to frequency but usually it is very good bandwidth and I have seen bad BNC connectors but it is rare and I always kept spares.

 

[gino1961]

Yes the fundamental frequency of the square wave rising and falling edge is infinite if it is a 90 degree pure angle. BNC connectors will have limits to frequency but usually it is very good bandwidth and I have seen bad BNC connectors but it is rare and I always kept spares.

thank you for the valuable advice The nice thing is that any issue in the signal chain can be clearly seen Then the job is to find the weak point
If i will want to check audio units i will have to use bnc to rca adapters of proper quality
I would be curious to check how an high end rca cable passes a sw at different levels and Hz Maybe it ruins the sw

 

[Doodski]

i will have to use bnc to rca adapters of proper quality

I used those all the time and they are very reliable but keep extras.

I would be curious to check how an high end rca cable passes a sw at different levels and Hz Maybe it ruins the sw

I have done this. The RCA cables pass the square wave well.

 

[gino1961]

I used those all the time and they are very reliable but keep extras.
I have done this. The RCA cables pass the square wave well.

thank you very much again Good to know that unbalanced cables are just fine
Next dark item are coupling caps
DC coupling is a very nice idea but very challenging to do it properly Many line preamps are AC coupled anyway
if i had a line preamp that shows some issues with sw at audio frequencies i would not be fully satisfied

 

[Doodski]

Next dark item are coupling caps
DC coupling is a very nice idea but very challenging to do it properly Many line preamps are AC coupled anyway

I have made several DC decoupling AC coupling adapters for those car audio high level outputs that float at 1/2 VCC (~7 volts DC.) and adapted to a RCA level termination . They work fantastic and in a sound comparison I could not detect if they where in or out of circuit. Very nice they are.