Audio VS. Linear

[ScottB]

On my other post where I was kicking around logging shunts, I mentioned something that might be relevant and I'd love to hear some speculation.

Is the impedence of the whole coil and volume control itself causing a linear tome control to act like a reverse log pot? Does this effect, if I haven't got it wrong, account for some of the preferences one way or the other?

[FF_Pedals]

Originally Posted by jonesy

I use all CTS audio taper pots for volume and tone in all my wiring harness's, you will get a smooth taper and even roll off that way.

I have found linear pots will roll off abuptly when used for volume pots, wether it is 50's or modern wiring.

Not sure where you are getting your info FFpedals but I diagree with a lot of the information I have seen you posting lately

I'm not getting my info anywhere other than around here and from playing my Les Paul.

I may have shit the bed on this one. There is something wrong with the audio taper pots I have though. I changed the pots and wiring at the same time so I'm not sure really what's causing it to mess up but it's got the on-off switch effect between 8-10 on the volume pots. One of them is from Alpha and the other is from ebay with no label. I'm just assuming Linear would work better for me. I've got new CTS pots coming so I should be all set up soon.

Please feel free to call me out on anything.

[ScottB]

Originally Posted by FF_Pedals

I'm not getting my info anywhere other than around here and from playing my Les Paul.

I may have shit the bed on this one. There is something wrong with the audio taper pots I have though. I changed the pots and wiring at the same time so I'm not sure really what's causing it to mess up but it's got the on-off switch effect between 8-10 on the volume pots. One of them is from Alpha and the other is from ebay with no label. I'm just assuming Linear would work better for me. I've got new CTS pots coming so I should be all set up soon.

Please feel free to call me out on anything.

Could be if you reversed the wiring to the pots it would not only fix the rotation direction but it would also accomodate the log curve built into the pot (which is probably not symmetrical on the true log curve).

Sounds like you have a left hand audio taper pot wired into a righty guitar or else a right handed pot wired as a lefty. Just speculating.

[jonesy]

Originally Posted by korus

if tone pot/cap is connected to input of volume pot - modern wiring
if tone pot/cap is connected to output of volume pot - 50s wiring

if inputs are connected to outer lug - dependent wiring
if inputs are connected to wiper lug - independent wiring

so there are 2*2=4 options

1. modern dependent
2. 50s dependent
3. modern independent
4. 50s independent

I think it is rather simple, though it's just me.
I consider independent wirings as tone killers.
50's wiring is unpredictable for those who use volumes below 5 (I do).

So, for me, modern dependent it is. With optional treble bleeds. YMMV, that makes life interesting.

HTH

My 04' SG is wired up 50's dependent and even with the stock Gibson pots (Russian PIO .047's) both volume and tone controls roll off very evenly all the way from10-1

I had 06' Studio LP wired modern dependent with stock pots (no volume bleeds) and all controls tapered off nicely as well with no sudden cut-off.

Some guitars sound better one way or the other, seems to vary from guitar to guitar IMHO

[FF_Pedals]

Originally Posted by ScottB

Could be if you reversed the wiring to the pots it would not only fix the rotation direction but it would also accomodate the log curve built into the pot (which is probably not symmetrical on the true log curve).

Sounds like you have a left hand audio taper pot wired into a righty guitar or else a right handed pot wired as a lefty. Just speculating.

But doesn't a log taper have most of the variation in the upper end? I've double checked the wiring, it's 50's independent volume. The fact that one is push pull and the other regular helps as there are two different construction types which helps rule out the chance of having got lefty pots.

The pickups are GFS. I'm getting a new bridge pickup soon, we'll see what happens when I rewire everything.

Sorry if what I posted in this thread confuses anyone or pisses anyone off who's trying to sell pots. My responses are just based off of my experience with this one guitar with these particular parts. The problem could potentially be due to a cheap approximation to the log curve used in the import pots.

[KP Willits]

I have 2 300K Linear Gibson pots and 2 CTS 250K Audio pots. After a bit more research; I set it up with the linears for volume and the audios for tone. Used Russian PIOs from Jonsey; a .022 for the neck and a .047 for the bridge. I am sure that many of you will find this to be an odd combination; but I am really liking it. Wish I could play better so I could give you tone clips. I guess my message is "Don't be afraid to try something new or unusual.".

[jonesy]

Originally Posted by KP Willits

I have 2 300K Linear Gibson pots and 2 CTS 250K Audio pots. After a bit more research; I set it up with the linears for volume and the audios for tone. Used Russian PIOs from Jonsey; a .022 for the neck and a .047 for the bridge. I am sure that many of you will find this to be an odd combination; but I am really liking it. Wish I could play better so I could give you tone clips. I guess my message is "Don't be afraid to try something new or unusual.".

Nice to hear you are tinkering around and trying new things bro. So many people just "Talk" about stuff but don't really try things out. IMO that is the only way you will know what works for you A+

[FF_Pedals]

Originally Posted by KP Willits

I have 2 300K Linear Gibson pots and 2 CTS 250K Audio pots. After a bit more research; I set it up with the linears for volume and the audios for tone. Used Russian PIOs from Jonsey; a .022 for the neck and a .047 for the bridge. I am sure that many of you will find this to be an odd combination; but I am really liking it. Wish I could play better so I could give you tone clips. I guess my message is "Don't be afraid to try something new or unusual.".

Sounds intriguing. What kind of guitar is it in? What kind of pickups?

[ScottB]

Originally Posted by FF_Pedals

But doesn't a log taper have most of the variation in the upper end? I've double checked the wiring, it's 50's independent volume. The fact that one is push pull and the other regular helps as there are two different construction types which helps rule out the chance of having got lefty pots.

...

That's why I suggested maybe you have CCW (lefty) pots, but unless both pots came from the same source then I don't see much chance of two random leftys in one build. Curious.

[ScottB]

The more I think about this, you have independent wiring, yes?

Remember pointing out to me the impedence of the coil?

If you have the pot wired backwards, wouldn't this, in addition to the long vs short side of the audio "curve" make it work not only backwards but little better than a linear, maybe worse?

Can't rule out a wiring anomaly but I'm going to assume everything is correct.

[FF_Pedals]

Okay, just busted out the multimeter to figure this one out.

ScottB, as per our discussion yesterday I posted your diagram for the 50's wiring with independent volumes:



When looking at the volume pot, from top to bottom we clearly have lugs, 3, 2, and 1 (ground).

The output impedance looking into the output jack of the guitar depends on which pickup selector you are in. Note, all the measurements are DC of course. When I pick a humbucker with all the volumes and tones cranked (tone has no effect at DC), the reading is the DC reading of the coil, around 8kohms. Put the selector in the middle for neck and bridge in parallel, it drops to ~4kohms. Now when you look at it, you would expect the output to see the 500k from the volume pot, but this resistance is in parallel with the coil resistance so you just see the coil resistance when at full volume. As the volume is turned down, resistance is added in series between the output and coil. Between the wiper and ground the impedance is the coil impedance in parallel with the resistance of the pot between the wiper and ground.

When measuring the output impedance for my import log pot, it jumps significantly when turned between 10 - > 8. When measuring the series resistance that's getting added in, I measured between lug 3 and lug 2, where the difference jumps from 0 ohms to several hundred kilo ohms in a small fraction of the turn.

Reading this article from R.G. Keen about pots: The Secret Life of Pots

you will come across this graph of potentiometer curves:


Copyright R.G. Keen Geofex.com

From looking at the logarithmic approximation you can see how a really sharp knee for a 2 segment approximation could give the problems I'm describing.

This would leave me to believe it is a bad idea to use import or cheap audio pots. You need audio pots with a smoother taper (CTS, Bourns, etc...). I imagine this is why some import guitars like Agile put linear taper pots for the volumes.

Hopefully my experience clears things up as to why we need high quality potentiometers if we want to use audio tapers.

[ScottB]

That is an excellent post, I need to chew on this for a little while.

I see what you are saying about the log "ramp" vs the true log curve, I'm still struggling because it seems if the pot were backwards that is exactly what you would get.

[ScottB]

Or it could be, just as you point out, a cheap substrat construction that "jumps" from 80% to 100% total resistance (or 0%-20%)

[FF_Pedals]

If it was reverse, the jump would be at 0-2 or something.

[FF_Pedals]

Man, I think I just figured out how to fix it. If you put a resistor in parallel with lugs 2 and 3, at full volume it has no effect because they're shorted. As you turn down, the resistance between lugs 2 and 3 shoots up because of the bad log taper. I put a 33k resistor in parallel so it never flies up into the hundreds of kilo ohms range. It totally smoothed out the taper...

[ScottB]

If the circuit turns off at 10 then we know one of two things:

at 10 (CW) the coil is grounded out, or

at 10 (CW) the wiper falls out of the circuit and opens the coil.

So let me ask because I have been making assumptions, but are you saying the pot works normally up to 8, with 8 as the loudest, and then beyond 8 just shuts off?

If that is the case then you have option 2 happening IMO

[ScottB]

Originally Posted by FF_Pedals

Man, I think I just figured out how to fix it. If you put a resistor in parallel with lugs 2 and 3, at full volume it has no effect because they're shorted. As you turn down, the resistance between lugs 2 and 3 shoots up because of the bad log taper. I put a 33k resistor in parallel so it never flies up into the hundreds of kilo ohms range. It totally smoothed out the taper...

In that case, sounds like the pot wiper is opening up at full.

[FF_Pedals]

No you are stuck on the backwards thing. It's not backwards, trust me. At 10 CW there is full volume. As you turn down there is a huge jump between 8 and 10, then a smooth taper down from 8 to 0. This is indicative of the two segment approximation. It's very annoying when you're trying to use a high gain pedal, especially a fuzz pedal. As you turn down the pot it sounds like you're flicking a boost switch off, rather than just gradually decreasing the volume.

[FF_Pedals]

When a normal log pot is at 10 (CW), lugs 2 and 3 are shorted. This is normal.

[ScottB]

I see. As if the bottom linear leg is from 0-8, and the top linear leg is 8-10. And the top is way too sharp.

Interesting. So back to what you and I discussed on my thoughts yesterday re: the log resistor, have you changed your mind somewhat? I should be in the same boat with a linear 1M and a 200k (or whatever) shunt.

Any new thoughts?

[korus]

Originally Posted by FF_Pedals

But doesn't a log taper have most of the variation in the upper end? I've double checked the wiring, it's 50's independent volume. The fact that one is push pull and the other regular helps as there are two different construction types which helps rule out the chance of having got lefty pots.

The pickups are GFS. I'm getting a new bridge pickup soon, we'll see what happens when I rewire everything.

Sorry if what I posted in this thread confuses anyone or pisses anyone off who's trying to sell pots. My responses are just based off of my experience with this one guitar with these particular parts. The problem could potentially be due to a cheap approximation to the log curve used in the import pots.

And that is exactly why if you (or anyone else) should use custom taper pots like RS Superpots (not afiliated). And it does not matter if it is 500$ or 5000$ guitar, whether pickups are 800$ per set or 125$ per set.

The signal chain (and any other chain for that matter) is as weak as its weakest component so do not let cheapo pots diminish everything else.

Once you try them you'll always have them spare, just like me.

[FF_Pedals]

Your proposed diagram used the resistor in parallel with the coil, hanging off the wiper to ground. My suggestion uses the resistor in parallel with lugs 2 and 3, which are in series with the (wiper + coil) node and the output.

You would have to plot some graphs to see the responses. In the link to Geofex, R. G. Keen shows plots for how to make log and reverse log pots from linear pots. What I'm trying to do is add the effect he uses to get reverse log effects to my cheap log pot to smooth out the effect of the sharp knee.

All in all, I will be replacing my pots with CTS pots.

[korus]

Originally Posted by FF_Pedals

Your proposed diagram used the resistor in parallel with the coil, hanging off the wiper to ground. My suggestion uses the resistor in parallel with lugs 2 and 3, which are in series with the (wiper + coil) node and the output.

You would have to plot some graphs to see the responses. In the link to Geofex, R. G. Keen shows plots for how to make log and reverse log pots from linear pots. What I'm trying to do is add the effect he uses to get reverse log effects to my cheap log pot to smooth out the effect of the sharp knee.

All in all, I will be replacing my pots with CTS pots.

For 500k audio (log) pot using 220k or 150k resistor connected with input and output (non grounded lugs) of the volume pot will bring log pot closer to linear taper. Try that while waiting for CTSs. 220k is closer to linear than 150k.

[FF_Pedals]

Cool, I'm psyched to get this sorted out. I have to take the whole thing apart and do a final buffing, should be nice and cured by now. I should have it all done in two or three weeks. It sounds great though right now, don't get me wrong...

[ScottB]

Originally Posted by FF_Pedals

Your proposed diagram used the resistor in parallel with the coil, hanging off the wiper to ground. My suggestion uses the resistor in parallel with lugs 2 and 3, which are in series with the (wiper + coil) node and the output.

You would have to plot some graphs to see the responses. In the link to Geofex, R. G. Keen shows plots for how to make log and reverse log pots from linear pots. What I'm trying to do is add the effect he uses to get reverse log effects to my cheap log pot to smooth out the effect of the sharp knee.

All in all, I will be replacing my pots with CTS pots.

I see, and because it doesn't matter (current is bi-directional) you made his "type II" reverse log circuit.

So you actually see functionality with a reverse log taper on the volume...I wonder what else is affecting this. Could just be the pot "taper" I guess.

[FF_Pedals]

Originally Posted by ScottB

I see, and because it doesn't matter (current is bi-directional) you made his "type II" reverse log circuit.

So you actually see functionality with a reverse log taper on the volume...I wonder what else is affecting this. Could just be the pot "taper" I guess.

No, what I meant was I used his configuration (getting a reverse log from linear) to smooth out the regular audio taper I have. The idea being 'adding' the reverse log effect to the regular log effect brings you closer to linear. I think it smooths out the knee, pulling it closer towards the linear case. I'd have to see plots though.

[ScottB]

That was kind of what I was putting together, sorry i didn't make it clear.

I'm curious now, to avoid confusion you said you put the shunt resistor between poles 2 & 3. Can you clarify the pinout?

Poles 1, 2 &3

Coil, Switch or ground.

[korus]

Originally Posted by FF_Pedals

No, what I meant was I used his configuration (getting a reverse log from linear) to smooth out the regular audio taper I have. The idea being 'adding' the reverse log effect to the regular log effect brings you closer to linear. I think it smooths out the knee, pulling it closer towards the linear case. I'd have to see plots though.

What they are doing here The Secret Life of Pots is using linear pot and putting resistor to it :
1. ground and output (wiper) to create audio taper
2. input (outer non ground) and output (wiper) to create reverse audio taper

What I am suggesting to you is to 2. but with AUDIO taper pot. That will bring it closer to linear or custom taper vol pot. Use aligator clips and you do not even need to solder.

edit : I see you already did that, so sorry to bother. btw what were the measured values of the pot and resistor ?

[ScottB]

Would a log shunt on an audio taper actually create a wavy-line log curve?

[FF_Pedals]

Lug 3 connects to the output and the tone section for 50's wiring. Lug 2 connects to the pickup. Lug 1 is ground.

I am using a 500k, cheap log pot. Just on a whim I connected a 33k resistor between lugs 2 and 3. Normally, when you turn down the volume, a series resistance is added between the output and the pickup as the wiper travels down the pot towards ground. With my crappy taper pots, the series resistance added between the pickup and the output shoots up from zero to hundreds of kilo ohms very fast. Note, with a 500k pot, when you are at full volume, the pickup is in parallel with the pot, so you only get like 8kohms output resistance. When the pot is fully turned down, the pickup is grounded so volume is off, the output resistance is the parallel combination of the added resistor and 500k.

Originally Posted by ScottB

Would a log shunt on an audio taper actually create a wavy-line log curve?

So two pots? What for?