Pot tapers, always confused
martin griffith
just looking for some rev log pots for a S+K HP filter, I can never
remember which curve is which. But does law C look wrong
http://www.omeg.co.uk/lawe16ct.htm
when compared to
http://www.bourns.com/pdfs/gapno_panel_controls_taper.pdf
martin
After the first death, there is no other.
(Dylan Thomas)
Walter Harley
news:o6tc61pakr3g6jb0u...@4ax.com...
> duh, one of those days.
> just looking for some rev log pots for a S+K HP filter, I can never
> remember which curve is which. But does law C look wrong
> http://www.omeg.co.uk/lawe16ct.htm
> when compared to
> http://www.bourns.com/pdfs/gapno_panel_controls_taper.pdf
Unless your confusion has infected me as well, the Bourns chart looks like
it uses different terminology than I've seen from other mfrs; and Omega's C
is what I would call reverse log. I'm used to thinking of A as log, B as
linear, C as reverse log, and D as a steeper variant on A.
Panasonic makes some reverse-log pots, but in the Digikey catalog I don't
see dual ones; just combined A/C, to be used as a stereo balance control.
Noble sell them, but I imagine you'd have to order 10,000.
Joel Kolstad
...does anyone have an equation for what the relative resistance of an audio
taper pot is supposed to be vs. position? Empirically I came up with
R(pos)=Rmax*(1.4^pos-1)/(1.4^10-1), where "pos" ranges from 0-10, but I
imagine somewhere this must be defined.
martin griffith
Do you mean relative resistance or attenuation, as in Volume Control?
Its difficult to define because of the bad cut off at the bottom of a
typical pot. Do you consider the 40dB,60dB or80dB atteuation point as
the other measuring point, compared with 0dB att.
Thats what I like about having a linear pot driving a log VCA, you can
so easily define the end points!
I cant honestly see any point in having any attenuation, below -60, it
makes the scale too cramped. A 40dB range would be better
but i'm sure J.Woodgate will know
Joel Kolstad
"martin griffith" <marting...@XXyahoo.co.uk> wrote in message
> Do you mean relative resistance or attenuation, as in Volume Control?
I meant resistance; I'm using one to tune some op-amp parameters rather than
as a volume control.
> Its difficult to define because of the bad cut off at the bottom of a
> typical pot.
Yeah, I was hoping for the theoretical curve. For my purposes, if a 1k pot is
still 1 ohm at the far end, that's close enough to zero for mean even if it
only is 60dB attenuation.
---Joel
martin griffith
John Woodgate
<marting...@XXyahoo.co.uk> wrote (in
<u83d61tj2ua9mbiu2...@4ax.com>) about 'Pot tapers, always
confused', on Wed, 20 Apr 2005:
>but i'm sure J.Woodgate will know
That's very flattering, but WHAT will I know? The relevant IEC standard
is IEC 60393-5, but I don't know whether it defines non-linear
resistance 'laws'.
Note that the 'log' laws are actually exponential!
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk
Pooh Bear
In practice, the usual inexpensive pots are made by screening 3 layers of
resistive ink, so merely approximating to a log / whatever curve. See the
manufacturers' curves, the change of slope is very obvious.
There are better ways to get good log accuracy but they cost more.
Graham
Pooh Bear
Joel Kolstad wrote:
Depends on the exact law. JIS audio tapers A ( volume control ) and C ( 'reverse
log' ) have 15% of total track ressitance at the central position. In practice
anything between about 10% and 20% actually.
That 1 ohm is normally called the 'hop on' resistance.
Other laws are available - usually desired for the 'reverse log' taper. e.g 10C -
means 10% total R @ centre. Manufacturers haven't defined these curves well and
they may vary from one manufacturer to another. The 'RD' curve from Alps is
perhaps best knows - I think it's similar to the 10C. I use a 'Y' taper from
Taiwan Alpha - or 05C pot in one of my mic pre-amps. If you're sweeping frequency
with a pot used as a series element - the ideal Centre value depends on the range
of control. E.g. I use a 20C type part for a swept mid frequency EQ circuit.
You'll only get these parts when making at least modest quantites though.
Graham
Precious Pup
martin griffith wrote:
>
> have a look at one of my defunct websites...
> http://www.wavefront.mcmail.com/pot.htm
Dude,
That kicks ass. Thanks.