Taper of Potentiometers

[olds...@tubes.com]

I have always known pots to either have an audio or a linear taper.
I'm looking at a schematic which has numerous pots. Some say they are "W
taper", the others say they are "K Taper". I have not seen them listed
in this way, ever.

What do these letters mean?
Are there other letters used?

Anyone have a chart that shows and explains them?

Thanks

(This schematic is for an early 1990's era semiconductor preamp, using
op-amps, and transistors).

[Fox's Mercantile]

On 1/23/18 5:08 PM, olds...@tubes.com wrote:
> Anyone have a chart that shows and explains them?

There aren't any.
Each manufacturer has their own idea of -suffix for what
kind of taper.

--
"I am a river to my people."
Jeff-1.0
WA6FWi
http:foxsmercantile.com

[pf...@aol.com]

On Tuesday, January 23, 2018 at 6:09:10 PM UTC-5, olds...@tubes.com wrote:
> I have always known pots to either have an audio or a linear taper.
> I'm looking at a schematic which has numerous pots. Some say they are "W
> taper", the others say they are "K Taper". I have not seen them listed
> in this way, ever.
>
> What do these letters mean?
> Are there other letters used?
>
> Anyone have a chart that shows and explains them?

W-taper pots are center-detent (center-null), typically used for tone controls, where center is Null.

K-taper pots typically will show a number ahead of the K, meaning *number* K-ohms.

K-pots may be linear or audio. W-pots are typically linear.

Your questions are getting better and better!

Peter Wieck
Melrose Park, PA

[olds...@tubes.com]

On Tue, 23 Jan 2018 18:01:50 -0800 (PST), "pf...@aol.com" <pf...@aol.com>
wrote:

I know that K ohms means 1000 ohms. But this is written as follows
(directly from the schematic) (except they use the ohm symbol and have
the +/- symbol printed one on top of the other)

Pot, 20K (ohm symbol) +/- 20%, K-taper

The other ones are
Pot, 50K (ohm symbol) +/- 20%, W-taper


There are 4 of the W-taper. so that makes sense about the tone controls
(center null). Because there are two per channel.

But there are a shitload of the K-taper ones, and this is a preamp
mixer, so lots of volume control pots.

This is an Altec Lansing preamp/mixer. It has a broken shaft on one of
the volume pots, so I need to replace that pot. I found a schematic for
it on this Altec Lansing unofficial website.

http://alteclansingunofficial.nlenet.net/proelectronics/rackmixers/index.html

This is the model to download:
(It says Click to download the .ZIP file)
Altec Lansing 1692A Mixer Literature Sheets - All

The zipfile contains a bunch of .JPG images.
The parts list where these pots are listed, is 1692a-4s.jpg

[Phil Allison]

olds...@tubes.com wrote:

------------------------

>
> I have always known pots to either have an audio or a linear taper.
>

** Tube radios and TVs using carbon track pots used either linear "lin" OR logarithmic "log" taper pots. The log pot would always be for volume.

Log pots had a very gradual rate of increase in resistance as you turned it clockwise.

Audio taper pots arrived with hi-fi amplifiers and gave a quicker rise in volume than a log type, so was preferred by many makers and also guitar amp makers.

The code letters used varied alarmingly, A originally meant linear and C =log. Pots coming from Asia used A = audio taper with B = linear and D = log.

The code letter "E" referred to a "reverse log" pot, useful in instrumentation where the actual gain of a amplification stage must be varied smoothly.

Strangely, I have not come across a reverse linear pot anywhere.


.... Phil

[N_Cook]

I had a go at rewinding a wire-wound log pot one time, it survived a
year and then something went wrong.
3 sections of former, like a large saw blade with 3 teeth stretched out,
bent into a ring. Each section had a different gauge of wire.
The wiper ran around the non-toothed edge of the "saw".

[jurb...@gmail.com]

>"Strangely, I have not come across a reverse linear pot anywhere. "

You probably have, in instrumentation somewhere. Just not marked as such. Like the intensity control of a scope or some shit like that. I could see there being applications like that. And some things just give you a part number and that's it.

I do bet they're rare though.

[jurb...@gmail.com]

>"I had a go at rewinding a wire-wound log pot one time, it survived a
year and then something went wrong. "

I repaired a variac a while back. Obviously it was linear. Luckily I caught it in time as it was rubbing the wiper over a winding that was popped out of place and it would have broken soon.

But don't confuse a variac with a pot, a variac actually works like a transformer. Now if one were to feed it DC it would be nothing but a pot, and likely burn up unless you kept the input voltage down.

But a little bit of glue took care of it, good thing too as it was one of the ones that go up to 150 VAC.

[tabb...@gmail.com]

What you been smokin?

[Mike Coon]

In article <gvsf6d960l0o7p6cr...@4ax.com>,
olds...@tubes.com says...

>
> I know that K ohms means 1000 ohms. But this is written as follows
> (directly from the schematic) (except they use the ohm symbol and have
> the +/- symbol printed one on top of the other)
>
> Pot, 20K (ohm symbol) +/- 20%, K-taper

Multiples of 1000 are written "k" not "K". There is an erroneous belief
that capital multipliers are >1 and lower case means divisors. This rule
works OK except for 1000...

Mike.

[Terry Schwartz]

The log pot would always be for volume.
>
> Log pots had a very gradual rate of increase in resistance as you turned it clockwise.
>

Logarithmic pots were typically used in volume controls for two reasons:

1) The log taper provided a finer control of listening volume at low levels, where users need that fine control, and

2) The log taper provides what is essentially a "dead band" at the low end of the pot, therefore accommodating the installation of an on/off switch on the back of the pot, without sacrificing low level volume control. The switch often requires a 15 or 20 degree rotation in order to activate -- using up part of the pot travel.

Interestingly, some of the aftermarket modular pot/switch manufacturers weren't that great about following those "rules" and when you installed such a replacement switch control into an old radio, the volume at switch activation was already too high for comfortable listening.

[pf...@aol.com]

On Wednesday, January 24, 2018 at 7:21:20 AM UTC-5, Mike Coon wrote:

> Multiples of 1000 are written "k" not "K". There is an erroneous belief
> that capital multipliers are >1 and lower case means divisors. This rule
> works OK except for 1000...

As already noted, there are almost as many codes as makers. In my (very) old Centerlab Catalog, I did find references to W-Taper pots, but nothing conclusive at all on K or k taper as a specific reference.

With the additional hint of this being a slider, further googling got me to drill bits, but nothing on pots, linear or rotary.

[Ralph Mowery]

In article <743d1aa8-fd64-414d...@googlegroups.com>,
tschw...@aol.com says...

I always though the audio taaper was because the way the ear responds to
sound in sort of a log function.

[Mike Coon]

In article <MPG.34d26d73...@news.east.earthlink.net>,
rmower...@earthlink.net says...

>
> I always though the audio taaper was because the way the ear responds
to
> sound in sort of a log function.

E.g. https://en.wikipedia.org/wiki/Weber%E2%80%93Fechner_law

Mike.

[Terry Schwartz]

Yes -- I imagine that is true, hence the need for finer control at lower volume levels.

Terry

[Terry Schwartz]

On Wednesday, January 24, 2018 at 10:59:48 AM UTC-6, Terry Schwartz wrote:
> Yes -- I imagine that is true, hence the need for finer control at lower volume levels.
>
> Terry

Actually, I think the ear responds in an anti-log fashion. Small changes in audio source output levels are perceived as large changes at lower volume levels. Not so much at higher levels, where the ear essentially "saturates" and cant tell the difference between 90dB and 100dB (numbers pulled out of nowhere as examples) So the log curve on the pot compensates for this.

[tabb...@gmail.com]

AIUI log pots were never really log. They normally used 2 resistance zones to give a very crude approximtion of a log law. It was good enough for audio, where the aim was to avoid everything happening down the bottom end of travel.


NT

[Gareth Magennis]

wrote in message
news:aa7420bf-47db-4c33...@googlegroups.com...

*************************************



Pah, mine goes to over 260 VAC.



Gareth.

[olds...@tubes.com]

On Wed, 24 Jan 2018 06:02:37 -0800 (PST), "pf...@aol.com" <pf...@aol.com>
wrote:

That's exactly what I found on the web. In other words, NOTHING
regarding the K-taper. But I did find the W-taper and it was explained
the same as you said. I found some forum where they were discussing pot
tapers and some mentioned even more letters, such as N, S, M, D, and a
few more. But not K. And yea, I managed to bring up drill bits too.
That forum is here:
http://www.diystompboxes.com/smfforum/index.php?topic=92918.0

This pot is R60, (master volume control). It's a unusual type of pot in
the sense it's mounted to the PCB and mounted backwards, meaning there
is a small piece of metal (ring) attached to it's rear, that is soldered
to the PCB and the three tabs are bent downward, where they are soldered
to the PCB.

I have not yet taken it off the board. The pot does work, but the shaft
is broken below the front panel surface. My first thought is to contact
Altec Lansing and see if I can get an exact replacement. If not, I may
try to saw off the remaining plastic shaft so it's flat, and make up a
"sleeve" for it, (a piece of metal tubing) and glue another shaft to it
with the tubing to reinforce it. This shaft should be about 1.5 inch
long, (before being broken) so I have room to work, since there is
still a half inch to work with. If none of that works, I'll have to
Mcgyver some other pot in there, and will choose an audio taper type.

With any luck, Altec will have a replacment part. That would be the
easiest fix.

A while back, I found out that Peavey electronics does have parts for
all their old stuff. I hope Altec Lansing does too.

For reference, there is a lot of useful info about pots on this website.
http://www.potentiometers.com/potcomFAQ.cfm?FAQID=29

[pf...@aol.com]

On Wednesday, January 24, 2018 at 4:24:24 PM UTC-5, olds...@tubes.com wrote:

> I have not yet taken it off the board. The pot does work, but the shaft
> is broken below the front panel surface.

JB Weld is your best friend. Get a soda-straw or heat-shrink tubing about the same diameter as the shaft, either use the broken piece or a suitable bit of solid material to extend the shaft to the correct length. Mix up the *SLOW* JB Weld epoxy, and use the straw/tubing to hold the shebang together and vertical, but out of the pot, until the epoxy sets up. Let the epoxy fill any voids. The result will be stronger than the original.

[olds...@tubes.com]

On Wed, 24 Jan 2018 05:48:52 -0800 (PST), Terry Schwartz
<tschw...@aol.com> wrote:

>2) The log taper provides what is essentially a "dead band" at the low end of
>the pot, therefore accommodating the installation of an on/off switch on the
>back of the pot, without sacrificing low level volume control. The switch often
> requires a 15 or 20 degree rotation in order to activate -- using up part of the
> pot travel.

This is something I never even thought about, but it makes a lot of
sense. I recall buying replacment pots for old tube gear, some 45 or
more years ago, and you bought them without a switch, and bought the
switch separately, which then attached to the rear of the pot.
Apparently those pots were made to be used with a switch, but also
worked without one. I never noticed any "dead spots" on them, or had
problems where I could not turn the volume down low enough. Apparently
that was all taken into consideration when they were made. But back
then, it was a lot simpler. You chose audio and linear taper, and the
correct resistance, and shaft length. That was about it. I recall
getting many with LONG shafts that had to be sawed off and a flat spot
filed onto the shaft.

The good thing back then is that the shafts were all metal and did not
break off.

[bruce2...@gmail.com]

Have you ever dipped the tip of a cigarette in gasoline and tried it?

[Phil Allison]

Terry Schwartz wrote:

------------------------

>
> Actually, I think the ear responds in an anti-log fashion. Small changes
> in audio source output levels are perceived as large changes at lower
> volume levels.
>
>

** Like other senses, hearing responds to *percentage" changes in sound level.

The ideal volume control produces similar percentage changes for similar rotations or small movements. Good "log" pots have three linear sections that approximate the ideal fairly well.


.... Phil

[jurb...@gmail.com]

>"Pah, mine goes to over 260 VAC. "

Don't try to pull that cotton/polyesther wool substitute over my eyes. Your nominal line voltage is 240. So that means it is like me having one that goes to 130. Yours would have to go to 150 to be equivalent.

What impresses me (some) is how some of these modern power supplies can handle auto-swicthing between line voltages.

[MOP CAP]

As soon as my son teaches me how to scan on my new printer, I will post
an old catalog page with many tapers shown. I will try to put in the
drop box. I think the page was from an old Centralab catalog.
CP

[jurb...@gmail.com]

>"I will try to put in the
drop box."

Too bad. Dropbox doesn't want to display for me. It might be because I used to be a member ad gave it up. I am currently looking for free hosting that is compatible with my browser.

I don't mind an ad or two but I don't want people to have to sign up or sign in and that is what it is doing for me.

[Ralph Mowery]

In article <37301d83-be37-4233...@googlegroups.com>,
pf...@aol.com says...

All that is what makes things 'interisting'. I believe the K is for
Kelvin temperature and that made them use the k for 1000.

Too bad that many companies seem ot use their own code for part numbers.
I really hate the companies that use their part number on standard parts
and if you need to replace them, you are almost forced to go to the
origional company or one of their repair man. Friend in the auto
repair business told me that often a luxulary car part would be the same
as a less expensive model. They use a different number and higher price.

[rickman]

If you give it a thought, you will realize a linear pot is its own reverse.
The reverse doesn't refer to whether the resistance goes up or down, it
refers to which direction the log curve goes, a log taper or an exponential
taper. A linear taper is the same either way... it's linear.

--

Rick C

Viewed the eclipse at Wintercrest Farms,
on the centerline of totality since 1998

[whit3rd]

On Wednesday, January 24, 2018 at 9:05:17 PM UTC-8, rickman wrote:
> Phil Allison wrote on 1/23/2018 10:29 PM:

> > Strangely, I have not come across a reverse linear pot anywhere.
>
> If you give it a thought, you will realize a linear pot is its own reverse.
> The reverse doesn't refer to whether the resistance goes up or down, it

> refers to which direction the log curve goes...

Log taper means mimicing log(d) where the deflection (d) goes from
1 to 100 or somesuch. Inverse log taper means mimicing log(1/d) = - log(d)

So, linear taper can mimic 'd', and its inverse can be '-d', as you say.
Or, linear taper can be 'd', and its inverse can be '1/d'.

The '1/d' taper would be a way to make an adjustable attenuator, linear in inverse gain
going from '1' meaning full signal, to '2' meaning 1/2, ... '10' meaning 1/10

[rickman]

Then it is no longer linear.

[Fox's Mercantile]

Congratulations, not only are you ignorant, you have no sense of humor.



--
"I am a river to my people."
Jeff-1.0
WA6FWi
http:foxsmercantile.com

[tabb...@gmail.com]

I would very much expect he didn't need to 'give it a thought' to work out something so obvious.


NT

[pf...@aol.com]

On Wednesday, January 24, 2018 at 9:37:24 PM UTC-5, jurb...@gmail.com wrote:
> >"I will try to put in the
> drop box."
>
> Too bad. Dropbox doesn't want to display for me. It might be because I used to be a member ad gave it up. I am currently looking for free hosting that is compatible with my browser.
>

Dropbox does not support Samsung devices for what that is worth. Or so they tell me.

[rickman]

And yet, he mentions he has never seen a reverse linear taper.

[pf...@aol.com]

On Thursday, January 25, 2018 at 8:06:29 AM UTC-5, rickman wrote:

> And yet, he mentions he has never seen a reverse linear taper.

Back when I was a journeyman electrician, and therefore was privileged to haze the apprentices, I asked one to fetch me (amongst many other things) a combo-plate from the warehouse like this: https://cdn.gescan.com/media/catalog/product/cache/1/image/9df78eab33525d08d6e5fb8d27136e95/i/b/ibv97532-2.jpg

And when he brought me this:

https://images.homedepot-static.com/productImages/5849af02-96ef-4c79-9192-37a8464488aa/svn/stainless-steel-legrand-pass-seymour-combination-wall-plates-sl18cc5-64_1000.jpg

I sent him back, stating I needed the receptacle on the left.

He got as far as the door before he caught on. And it never happened to him again. Others got all the way to the counter at the warehouse.

[amdx]

I have repaired many broken plastic parts.
I have a kit with 0.025" and 0.035" drill bits and
matching steel rods. I put the parts together, then drill
through where inserting a steel pin will give it the most strength.
I epoxy the joint and the pin and put it all together.
I use a Dremel tool to drill the hole and and cut the steel pin off,
after the epoxy sets.
The epoxy I have is Huntsman Fastweld 10, (ex. Ciba-Geigy Araldite)
It is a 5 minute set 24 hour cure.

Mikek

[amdx]

In my late teens a father bought his daughter a Volkswagen Beetle,
the first thing he did was send her to go to the gas station* and have
them check the water in the radiator.

Mikek

* back when they had service at gas stations.

[Fox's Mercantile]

On 1/25/18 7:06 AM, rickman wrote:
>
> And yet, he mentions he has never seen a reverse linear taper.

Jesus you're fucking dense.

It was God damned joke you moron.

[tabb...@gmail.com]

On Thursday, 25 January 2018 18:12:03 UTC, Fox's Mercantile wrote:
> On 1/25/18 7:06 AM, rickman wrote:

> > And yet, he mentions he has never seen a reverse linear taper.
>
> Jesus you're fucking dense.
>
> It was God damned joke you moron.

bizarre isn't it.

[Phil Allison]

Phil Allison wrote:

----------------------

> >
>
> ** Like other senses, hearing responds to *percentage" changes in sound
> level.
>
> The ideal volume control produces similar percentage changes for
> similar rotations or small movements. Good "log" pots have three
> linear sections that approximate the ideal fairly well.
>
>

** FYI:

A simple way to get *very repeatable* "log" type curve for audio is to wire the two sections of a dual linear pot in series. The wiper of the first half goes to the top of the second and its wiper becomes the output. Bottom ends are both grounded.

At the centre (50%) position, this reliably gives -15dB attenuation when loading effects are taken into account.

At the 75% setting, attenuation is -6.5dB, at 25% it is -25dB while down at 10%, attenuation is -40dB.


..... Phil

[Clifford Heath]

On 26/01/18 15:45, Phil Allison wrote:

> Phil Allison wrote:
>> ** Like other senses, hearing responds to *percentage" changes in sound
>> level.
>>
>> The ideal volume control produces similar percentage changes for
>> similar rotations or small movements. Good "log" pots have three
>> linear sections that approximate the ideal fairly well.
> ** FYI:
>
> A simple way to get *very repeatable* "log" type curve for audio is to wire the two sections of a dual linear pot in series. The wiper of the first half goes to the top of the second and its wiper becomes the output. Bottom ends are both grounded.
>
> At the centre (50%) position, this reliably gives -15dB attenuation when loading effects are taken into account.

Hah! Square-root taper :)
Probably good enough for rock-n-^H^H^H^H^Haudio

If you know the source and load impedance of the pot,
wiring a linear pot with the input to the wiper and
output from the top gets you somewhere close to the
same effect.

[jurb...@gmail.com]

>"And when he brought me this:"

They got me once at the TV parts store. I needed a 223" color CRT for a Motorola. The guy says he only had 25" CRTs right now. I said "How am I going to fit that ?" and he replied "A cabinet stretcher". I said "OK, where do I get one of those ?".

Once and only once...

[jurb...@gmail.com]

>"A simple way to get *very repeatable* "log" type curve for audio is to wire the two sections of a dual linear pot in series. The wiper of the first half goes to the top of the second and its wiper becomes the output. Bottom ends are both grounded.

At the centre (50%) position, this reliably gives -15dB attenuation when loading effects are taken into account.

At the 75% setting, attenuation is -6.5dB, at 25% it is -25dB while down at 10%, attenuation is -40dB.


..... Phil"

That is what high end audio equipment does. Have seen it a few times. It does make for a nice smooth taper. Also in preamps it enhances the S/N ratio at lower volume settings. Usually the first pot is somewhere near the tone control circuit, if any. The second pot is right at the preamp outputs.

Actually if a volume control is "too fast" a resistor between the wiper and ground side makes a pretty decent taper.

One time a guy broings in a receiver, an upgrade from the one he had but oth the same brand. He asked why he had to turn it higher to get the same volume level even though it was more power. I had to explain it so I used the accelerator in a car. I told him they could make the linkage so it is just about floored when you give it half a pedal, but does that make the car go faster ? Nope, it has the same horsepower (kilowatts for youse across the pond) no matter how the accelerator linkage is designed. I think he understood. I told him it gives him more accurate control.

[Phil Allison]

Clifford Heath wrote:

----------------------

>
> > A simple way to get *very repeatable* "log" type curve for audio is
> > to wire the two sections of a dual linear pot in series. The wiper
> > of the first half goes to the top of the second and its wiper becomes
> > the output. Bottom ends are both grounded.
> >
> > At the centre (50%) position, this reliably gives -15dB attenuation
> > when loading effects are taken into account.
>
>
> Hah! Square-root taper :)
>

** Is that what it's called ?

> Probably good enough for rock-n-^H^H^H^H^Haudio
>

** It's so repeatable, unlike with some batch of low cost log pots, you could have the -dBs figures printed on a front panel and they would always align with the pot.

> If you know the source and load impedance of the pot,
> wiring a linear pot with the input to the wiper and
> output from the top gets you somewhere close to the
> same effect.

** Horrible idea - it shorts the signal at low settings and sends the output impedance high just when you want it low for noise/hum reasons.


..... Phil

[jurb...@gmail.com]

>"** Horrible idea - it shorts the signal at low settings and sends the output impedance high just when you want it low for noise/hum reasons. "

Isn't that how some guitar pickups are wired ? And don't ask me why.

[Phil Allison]

jurb...@gmail.com wrote:

-------------------------

>
>
> >"** Horrible idea - it shorts the signal at low settings and sends the
> > output impedance high just when you want it low for noise/hum reasons. "
>
>
> Isn't that how some guitar pickups are wired ? And don't ask me why.
>

** Only very rarely.

That method has just *one* advantage where there are multiple PUs, each with its own volume pot. It forms a resistive mixer that allows signal to pass from any pickup to the output regardless of the other pot settings.

The usual method requires no control be zeroed when PUs are switched in parallel or the instrument becomes silent.

The wiring schemes used in most electric guitars are primitive, full of dodges and compromises - good examples of how NOT to do it.


See wiring for early Gibson models.

http://archive.gibson.com/Files/schematics/lespaul2.gif




.... Phil

[jurb...@gmail.com]

>"Strangely, I have not come across a reverse linear pot anywhere. "

Now I wonder what kind of taper is used in the pots in a graphic equalizer. Double reverse anti-log or some bizarre shit like that ?

[Fox's Mercantile]

For fuck's sake. Am I the only one here that got the joke?

[tabb...@gmail.com]

On Friday, 26 January 2018 14:33:49 UTC, Fox's Mercantile wrote:
> On 1/26/18 7:48 AM, jurb...@gmail.com wrote:
> >> "Strangely, I have not come across a reverse linear pot anywhere."
> >
> > Now I wonder what kind of taper is used in the pots in a graphic
> > equalizer. Double reverse anti-log or some bizarre shit like that ?

I don't know, but would expect linear.

> For fuck's sake. Am I the only one here that got the joke?

someone hasn't been paying attention


NT

[Clifford Heath]

On 26/01/18 19:19, Phil Allison wrote:

> Clifford Heath wrote:
>> Hah! Square-root taper :)
> ** Is that what it's called ?

No idea what it's called, but if a log pot can be
called that (even though the intended operation is
exponential, i.e. opposite of log, to compensate
for the log response of the ear) then this should
be square-root, because the actual curve is the
square function.

Clifford Heath.

[Phil Allison]

Clifford Heath wrote:

-----------------------

>
> >> Hah! Square-root taper :)
> >>
> > ** Is that what it's called ?
>
>
> No idea what it's called, but if a log pot can be
> called that (even though the intended operation is
> exponential, i.e. opposite of log, to compensate
> for the log response of the ear) then this should
> be square-root, because the actual curve is the
> square function.
>

** At least one source calls it a "square law attenuator" or SLA.

http://www.bernacomp.com/elec/og2/square_law_attenuator.gif

A SLA closely follows the formula:

Gain = 40log(position ratio) where closed =0 and open =1



.... Phil

[Phil Allison]

jurb...@gmail.com wrote:

-------------------------

>
> Now I wonder what kind of taper is used in the pots in a graphic
> equalizer. Double reverse anti-log or some bizarre shit like that ?
>
>

** The vast majority use standard, linear slider pots - which tend to crowd most of the dB adjustment range in the last 25-30% of their travel.

A few special types have a small dead band in the middle of the track accompanied by a centre detent.

I recall seeing just one, up market graphic that had sliders with symmetrical non-linear tracks to compensate for the end crowding effect and give a linear result.


.... Phil