pan pot impedance

[tomugli]

could those with more knowledge help one with less in answering a little question i have?

is it possibe, and if so what value, to put a reverse log dual pot (as pan control) straight after a 10k fader without a buffer in between them.

if not; could anyone suggest a suitable fet to use in a simple voltage follower (an alternative to the 2sk170?) for the buffering application?

thanks for any help you can offer,

tom

 

[JohnRoberts]

could those with more knowledge help one with less in answering a little question i have?

is it possibe, and if so what value, to put a reverse log dual pot (as pan control) straight after a 10k fader without a buffer in between them.

if not; could anyone suggest a suitable fet to use in a simple voltage follower (an alternative to the 2sk170?) for the buffering application?

thanks for any help you can offer,

tom

While the pan law would not likely be constant power or constant voltage you can cascade passive pan after a unbuffered fader by scaling up impedances.  Depending on how many sources are being combined I would be inclined to take the loss rather than add distortion from a too simple buffer.

This is a pretty widely discussed topic and it will depend somewhat on what pot tapers you can source and other factors.

JR

 

[tomugli]

thank you for the quick reply!

I have read that it should be roughly a 1:10 impedance ratio. If this is correct, i believe that this would mean using a 100k antil-log pan pot after the 10k fader, but i am unsure as to what kind of effect this would have on the audio signal, and as to whether there would be any benefit to this kind of topology over using some kind of either fet or ic buffer. I am hoping eventually to combine a total of 16 sources.

thanks again for any help you can offer.

tom

 

[abbey road d enfer]

You would have to use a dual 100k pot with 220 kohms bus injection resistors. This is not the way to optimise the noise structure. The expected benefits of not having a buffer would be countered by the cons. IMO, a buffer between fader and panpot has many pros and few cons. Why are you considering a single FET buffer? Again, the expected benefits are ruined by the impossibility to achieve really high performance.
Have you actually tried to source log/antilog dual pots?

 

[skipwave]

I'm prepping a little DIY 16 channel mixer project. Going to have buffered aux sends, but was going to do without the buffer before the panpot based on comments in the following thread (which is missing posts from he-who-shall-no-longer-be-named-here):

http://www.groupdiy.com/index.php?topic=1490

As I understand it, using the Orban panpot implementation, the impedance doesn't vary too much allowing buffer-less input.

http://www.groupdiy.com/index.php?topic=7537.msg89400#msg89400

 

[abbey road d enfer]

I'm prepping a little DIY 16 channel mixer project. Going to have buffered aux sends, but was going to do without the buffer before the panpot based on comments in the following thread (which is missing posts from he-who-shall-no-longer-be-named-here):
http://www.groupdiy.com/index.php?topic=1490

The absence of the pariah's posts make the thread difficult to understand. Anyway I haven't seen anything in this thread that recommends NOT using a buffer...

As I understand it, using the Orban panpot implementation, the impedance doesn't vary too much allowing buffer-less input.

The pan-pot law will not be affected by the fader's position, but the fader's law will, and the attenuation will make your mixing-bus operating-level quite low, inducing noise.
[/quote]

 

[tomugli]

in terms of pot sourcing, that was going to be a bit of an experiment based loosely around something i read about swapping carbon tracks in dual pots, but judging by the responses here there may perhapps be more sound approaches! maybe something to try at a later date.

i saw the NYD pan pot schematic and i plan to experiment with that topology, also having found a schematic showing a proposed implementation with a 10k fader and no buffer. I suppose really, i am searching for the most minimalistic approach at implementing fader-pan-summing bus i can. (without it sounding awful!)

cheers, tom

 

[Svart]

Does it have to have the pan pot after the fader?  Are you trying to go with a single track fader and that is why you want the panpot after it?

 

[tomugli]

Svart,

i must admit that this is something which had not crossed my mind! i could go with a dual fader. do you happen to have a link to any schematics i could have a look at? is this as simple as it sounds?! 

 

[Svart]

I don't have any schematics, it was just an idea so that your low impedance driver is driving the variable impedance panpot and the panpot will drive a known load(fader configured as a voltage divider).  I'll look around and see what I find, I don't even know if it's feasible or worth looking at.

 

[skipwave]

I'm prepping a little DIY 16 channel mixer project. Going to have buffered aux sends, but was going to do without the buffer before the panpot based on comments in the following thread (which is missing posts from he-who-shall-no-longer-be-named-here):
http://www.groupdiy.com/index.php?topic=1490

The absence of the pariah's posts make the thread difficult to understand.


Anyway I haven't seen anything in this thread that recommends NOT using a buffer...

This post: http://www.groupdiy.com/index.php?topic=1490.msg18927#msg18927

Says: "For a truly minimalist implementation, you could eliminate the JFET Q1 as well as C1, C2, R1 and R2. The input impedance of the panpot is high enough that you don't really need the buffer. Input impedance will range between 6K and 10K depending on control settings, and most gear can drive that without problems."

Reference to: Schematic

 

[abbey road d enfer]

With this arrangement, the busses operate at -28dBu. With 5 inputs, as on the schematic, bus noise will stand at ca. -78dBu. With 16 inputs, it will increase to -69dBu. Hardly acceptable. There are limits to a minimalist approach.

 

[skipwave]

Thanks for the explanation. I suspected there would be a noise penalty. I'm trying to figure out how to pull off the lowest parts count for 16 inputs with fader, pan pot, and two (buffered) aux sends.

 

[abbey road d enfer]

Why are you concerned about the parts count? The cost of buffering is a tiny fraction of the cost of pots, faders, connectors, metalwork, PSU.

 

[skipwave]

Oh, certainly I don't object to buffers. The FET source follower shown in the schematic is particularly appealing as such, because it is minimalist in the sense of a single, discrete device being the simplest active circuit. However, thinking the buffers could be eliminated, I also have no desire to retain any unnecessary devices (I refrain from saying components, as the grey area encompassed by "unnecessary, but still quite useful" becomes considerably larger).

 

[tomugli]

The FET source follower shown in the schematic

the specified 2sk170 proves seemingly elusive. is there a suitable alternative you are using skipwave?

 

[abbey road d enfer]

http://www.mintech.co.uk/News/linear_lsk170.pdf