I have a 100K 24step stepped attenuator
the differences are fairly simple to define.
A volume pot has an infinitely variable resolution, vs your
attenuator which has 24 discrete steps.
A volume pot suffers from dirt acting between the wiper and the track to make noise , the attenuator has simple contacts which use higher contact forces, and tend to be self-cleaning - something which would wear away a pot track.
The attenuator probably uses high tolerance resistors for the elements, these are much closer matched than most volume pots, leading to better channel matching.
Volume pots tend to use carbon tracks unless you get expensive ones, attenuators tend to use closely matched metal film resistors, which are less noisy than carbon.
so basically you have 4 factors which lead you to use the attenuator rather than a volume pot:
1) better channel matching
2) better noise performance
3) better long term stability and reliability
4) you already have the attenuator.
I hope this helps,
Ray
the differences are fairly simple to define.
A volume pot has an infinitely variable resolution, vs your
attenuator which has 24 discrete steps.
A volume pot suffers from dirt acting between the wiper and the track to make noise , the attenuator has simple contacts which use higher contact forces, and tend to be self-cleaning - something which would wear away a pot track.
The attenuator probably uses high tolerance resistors for the elements, these are much closer matched than most volume pots, leading to better channel matching.
Volume pots tend to use carbon tracks unless you get expensive ones, attenuators tend to use closely matched metal film resistors, which are less noisy than carbon.
so basically you have 4 factors which lead you to use the attenuator rather than a volume pot:
1) better channel matching
2) better noise performance
3) better long term stability and reliability
4) you already have the attenuator.
I hope this helps,
Ray
..not quite the answer
I've just realised after posting my last msg that I've answered a slightly different question
I thought you asked 'volume pot VS attenuator'.... my bad, sorry.
To try to answer your original question, I need to ask you whether you have a chain-select type of attenuator , or a matched-pair one?
the chain type uses a series of resistors, the input comes in at one end, the other is grounded, and the attenuator selects the point on the chain where the output is taken from.
problem is that each resistor adds noise to the signal, and 23 resistors add most noise at the worst point - i.e. minimum volume setting.
the matched-pair type typically has 2 'decks' of switching elements and resistors and is arranged so there is never more than 2 resistors involved at any setting , and only one in the signal path , the setting simply selects the pair of resistors which provides the appropriate setting.
which type do you have?
Does this help answer the question , or have I simply muddied the waters ?
Ray
I've just realised after posting my last msg that I've answered a slightly different question
I thought you asked 'volume pot VS attenuator'.... my bad, sorry.
To try to answer your original question, I need to ask you whether you have a chain-select type of attenuator , or a matched-pair one?
the chain type uses a series of resistors, the input comes in at one end, the other is grounded, and the attenuator selects the point on the chain where the output is taken from.
problem is that each resistor adds noise to the signal, and 23 resistors add most noise at the worst point - i.e. minimum volume setting.
the matched-pair type typically has 2 'decks' of switching elements and resistors and is arranged so there is never more than 2 resistors involved at any setting , and only one in the signal path , the setting simply selects the pair of resistors which provides the appropriate setting.
which type do you have?
Does this help answer the question , or have I simply muddied the waters ?
Ray
Are you talking about thermal noise ..
no, thermal noise won't vary from resistor type or number , as you point out.
I was thinking more of 'contact' or '1/f' noise. it's a fairly marginal thing , but we are aiming for the best possible sound, i expect.
I suppose you could use non-inductive wirewounds for the attenuator , then you would pretty much eliminate all resistor noise except the thermal component.
pricey too.
ray
no, thermal noise won't vary from resistor type or number , as you point out.
I was thinking more of 'contact' or '1/f' noise. it's a fairly marginal thing , but we are aiming for the best possible sound, i expect.
I suppose you could use non-inductive wirewounds for the attenuator , then you would pretty much eliminate all resistor noise except the thermal component.
pricey too.
ray
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.