Resistor in parallel with potentiometer

[Big_Mike]

I just finished a single-ended Plexi clone, and having been putting it through the paces the last few days.  The circuit has a master volume, 500K audio pot, with a 470K resistor and a cap between lugs 2 and 3.  The wires from the board for the master volume then attach to lugs 2 and 3.

What is the purpose of having the resistor connecting the two lugs, and how does it affect resistance?  One thing I am trying to play with is the taper of the pot to allow for better control at lower volumes.  I am not sure how the resistor added to the pot influences the circuit.

[pompeiisneaks]

It works just like any resistors in parallel, it just adds that 470k so that at max the resistance is less than 470k.  That's the rule of parallel resistors, the resistance is always less than the lesser of the two in parallel. 

As you lower the pot in resistance for the path of the signal, it goes lower of course.   (maybe the original designer wanted a 400k or 450k instead of 500k, that was the easiest way to get it.)

~Phil

[John]

There is another reason for the parallel resistor: it provides a path to ground for the power tube grid should the wiper lose contact with the track of the pot.,

[jjasilli]

Not clear from your post; schematics are always best.  Presumably the MV pot is wired as a voltage divider: 1) Input; 2) Output; 3) Ground.  The fixed resistor changes the taper of the pot; and is in parallel with 2+3 only, not the whole pot. 


There are formulas for pot taper as affected by a parallel fixed resistor, but you can measure the pot/resistor resistance with a meter at various points of pot rotation to see exactly what's happening.


The cap is likely bleeding treble to ground. You can temporarily remove the cap form the circuit to ascertain its effect.

[tubeswell]

... One thing I am trying to play with is the taper of the pot to allow for better control at lower volumes.  I am not sure how the resistor added to the pot influences the circuit.

Pot tapering discussion starts 1/2 way down this page:


http://www.geofex.com/article_folders/potsecrets/potscret.htm

[Big_Mike]

Not clear from your post; schematics are always best.  Presumably the MV pot is wired as a voltage divider: 1) Input; 2) Output; 3) Ground.  The fixed resistor changes the taper of the pot; and is in parallel with 2+3 only, not the whole pot. 


There are formulas for pot taper as affected by a parallel fixed resistor, but you can measure the pot/resistor resistance with a meter at various points of pot rotation to see exactly what's happening.


The cap is likely bleeding treble to ground. You can temporarily remove the cap form the circuit to ascertain its effect.

You are correct - lug 1 goes to ground, lugs 2 and 3 are input and output.  The pot is a 500K audio pot with a 470K resistor across the 2 and 3 lugs.

[jjasilli]

I think you have your lug nomenclature reversed.  When looked at from behind, i.e., inside the chassis, with the lugs facing up so you can see & get at them, the numbering is 3, 2, 1 from left to right.  Normally 3 goes to ground. 

[Big_Mike]

I think you have your lug nomenclature reversed.  When looked at from behind, i.e., inside the chassis, with the lugs facing up so you can see & get at them, the numbering is 3, 2, 1 from left to right.  Normally 3 goes to ground.

Sorry about that, yes my nomenclature is backwards.  Lug 3 goes to ground, lugs 1 and 2 are input and output, with a 470K resistor between lugs 1 and 2.

[jjasilli]

OK, good.  The fixed resistor between lugs 1 & 2 is more normal.  If the cap is also between 1 & 2 then it is preserving treble as the pot is turned down in volume.  As the resistance of the pot is lowered, vol goes down because more signal bleeds through the pot to ground.  But, treble bleeds to ground first.  So treble tends to disappear at low vol settings.  The addition of the cap keeps treble in the signal chain when vol is turned down on that pot.  With a high vol setting the cap is superfluous.  This circuit is also used to modify guitar vol pots - the treble bleed mod.


The cap is in parallel with a resistance.  This is a type of RC circuit (R= Resistor; C = Cap) called a notch circuit.  It passes a certain band of frequency; the band varies as the resistance value is change with pot rotation.  The simple explanation is that the cap passes highs at low vol settings, as stated above.

[2deaf]

One thing I am trying to play with is the taper of the pot to allow for better control at lower volumes.

From the way I understand the connections you have, the parallel fixed resistor is from the hot side of the pot. to the wiper.  If so, this is doing the opposite of what you want.  A parallel fixed resistor from wiper to ground will make it so that you have to turn the knob more to get the same change at low levels.  I would try taking the parallel fixed resistor off altogether and see how that works.

A resistive voltage divider by itself doesn't affect frequency response.  In your situation, there is capacitance at the tube that the wiper is going to and that will affect the frequency response.  The capacitor from the hot side of the pot. to the wiper counters the high frequency roll-off caused by the tube input capacitance.  Frequently, the capacitor on the potentiometer way overcompensates in which case it is a "bright" feature.

You don't see capacitors on master volume pot.'s very often because they usually feed a phase inverter or a pentode.  Phase inverters and pentodes have much lower input capacitance than a typical preamp triode.     

[Big_Mike]

Thanks for all the information.  Attached is the layout of the amp, you can see how the master volume is wired up.  I removed the resistor this afternoon, and the amp seem to have more "sparkle", especially on solo notes.  Not sure why the designer of the amp added the resistor and cap combination to the master, is it just to preserve more treble as the volume is decreased?

[2deaf]

There is a 220K resistor from the Master Volume wiper to ground.  The input capacitance on the EL34 is negligible.  So the original setup will have an audio taper that is slightly closer to linear and that will become brighter as the volume is turned down.

[kagliostro]

A schematic is always more readable than a layout





more easy to understand

Franco

[Big_Mike]

Thanks to everyone for all the replies.  I am going to do an A/B test with and without the 470K resistor across lugs 1 and 2 and see which one I like better.  This amp sounds really good, by the way, I highly recommend it if you want Plexi tone at reasonable volumes.  It can still get quite loud, even clean.

[Tony Bones]

Here is the taper of the composite potentiometer as draw.
I've also attached the Excel sheet for anyone that wants to play around with different values. You should be able to load it into any version ofExcel as well as free spreadsheet programs like Libre Office or Gnumeric. (Gnumeric is more compact and faster than Libre Office.)

[Tony Bones]

To save a little trouble, here's the taper without the 470k resistor. (I used the same spreadsheet, I just put in a really big number for the top fixed resistor.)

[Big_Mike]

Is there a particular reason why the second fixed resistor in the circuit has to be 220K?  In playing with the excel chart, it seems like I could have better low volume control if I lowered that resistor value.  Don't know if there is any impedence or safety issue with needing a 220K resistor in the circuit at that location.

[jjasilli]

The topic is complicated, because often a component has multiple functions.  The 220K resistor is often called a grid leak resistor, because the power tube's signal grid (G1) NEEDS to be connected to ground in order to maintain bias.  Without the grid-ground connection the tube would have an internal short circuit and fail catastrophically.  Per Ohms' Law, because there is virtually no current, the value of the grid leak resistor for that purpose is arbitrary, unless the resistance gets way too high.  See the tube specs. G1 sees the 5.6K grid stopper + the 220K grid leak resistor on the way to ground.  The combined value must not exceed the tube specs.

Also, the 220K grid leak resistor has another function.  It lifts signal voltage from ground to charge G1, while completing the signal circuit at that point in the amp. If you lower the value of the grid leak resistor, signal voltage will decline -- there will be less gain; & vice-versa. 


N.B.: the 220K grid leak resistor does not equal the grid leak resistance!  That's because the 220k grid leak resistor is in parallel with the 500K pot which is partially in parallel with its 470K bypass resistor.  So the actual grid leak resistance is variable (because of the pot) but never more than 220K.

[Tony Bones]

Mike, the only risk I see with making the 220k grid leak too small is that it will load down the driver tube when the master is at max. The load that V2b sees (with MV at max setting) is its own 100k plate load in parallel with both the 500k pot and the grid leak. The 500k pot is big enough that it doesn't have much affect. Even 220k // 100k is 68k, so not too low.

But, if you change the 220k to 100k then the V2b will see the parallel combination as 50k. Use 68k grid leak and then V2b will see only 40k load. Bottom line is the amp will sound different than it does now with the master volume set high because of that. But, you might like it better. Who am I to say.

As you turn down the master, the top portion - the part between the input and the wiper - gets put in series with the grid stopper, so the load resistance seen by V2b goes up to a maximum of almost 100k when turned all the way down.

Does any of this matter? Maybe not much if you end up happy with the sound.

[jjasilli]

Ok, we all agree on the theory.  Let's get back to the problem.  The amp gets loud too fast.  That's a problem with pot rotation, not some other part of the circuit.  I suspect the bypass resistor may be defeating the pot's audio taper. Troubleshoot logically:


1.  Make absolutely certain you actually have an audio taper pot. Don't trust what's stamped on the pot's body.  Temporarily disconnect everything & measure the pot.  If you don't know what to measure for, then read the Secret Life of Pots which someone posted above.


2.  Reconnect the pot alone, or a proper 470K audio taper pot -- no bypass resistor or cap. 


3.  Try out the amp.  If you're happy you can stop here.


4.  Replace the bypass cap.  You can try various values between 10pF - 470 or 500pF.  This should increase treble at low pot settings.


5.  If I'm reading the Secret Life of Pots correctly, a bypass resistor between lugs 2 & 3 may exaggerate the audio taper and give you more early pot rotation with very gradual vol increase at low pot settings.

[2deaf]

As you turn down the master, the top portion - the part between the input and the wiper - gets put in series with the grid stopper, so the load resistance seen by V2b goes up to a maximum of almost 100k when turned all the way down.

You lost me on that one.  You must be talking about V2a, but I get 242K for the stock MV when turned all the way down for a total load on V2a of 70.7K.

[2deaf]

If I'm reading the Secret Life of Pots correctly, a bypass resistor between lugs 2 & 3 may exaggerate the audio taper and give you more early pot rotation with very gradual vol increase at low pot settings.

That is what happens, but it is a little bit difficult to ascertain exactly what happens from the article.  At first "b" equals the taper resistance divided by the pot resistance.  But then in the graph, "b" equals the pot resistance divided by the taper resistance.  Then there are seven numbers for "b", but only six lines.  Furthermore, the only time "b" can equal zero is when the pot resistance is zero which would not produce the diagonal line seen in the graph.   

[Tony Bones]

1.  Make absolutely certain you actually have an audio taper pot. Don't trust what's stamped on the pot's body.  Temporarily disconnect everything & measure the pot.  If you don't know what to measure for, then read the Secret Life of Pots which someone posted above.

I must have forgotten that point. (I don't re-read the entire thread every time I open it.)

That Excel sheet assumes a linear pot. I can fix it to work with different log tapers, but for now we should understand that an audio taper pot will exaggerate the curve. I other words, with the parallel resistors, it rotating the pot will hardly do anything until you get almost full up, then all the change will come in a hurry.

[jjasilli]

@2deaf:  I think Tony Bones' explanation in Reply18 is OK.  For AC operation of V2b, lowering the load of its output effectively reduces the gain of V2b.  It's a different way of stating what I said in my Reply 19.


I reiterate that we should re-focus on fixing the early vol. issue.

[Tony Bones]

As you turn down the master, the top portion - the part between the input and the wiper - gets put in series with the grid stopper, so the load resistance seen by V2b goes up to a maximum of almost 100k when turned all the way down.

You lost me on that one.  You must be talking about V2a, but I get 242K for the stock MV when turned all the way down for a total load on V2a of 70.7K.

I'm thinking you have a 250k MV pot, not 500k as shown in post #12. With a 500k pot you should see just a little less than 500k (from one outside lug to the other) when turned all the way down. No matter, you have what you have and I'm not suggesting changing it. It's just that I was assuming 500k.

[2deaf]

@2deaf:  I think Tony Bones' explanation in Reply18 is OK.  For AC operation of V2b [sic], lowering the load of its output effectively reduces the gain of V2b [sic].  It's a different way of stating what I said in my Reply 19.

I thought that Tony's explanation was pretty good, also.  I thought it was a good idea to take the 500K out of the equation to simplify things and make it easier to see what happens.  But the one sentence I referenced makes no sense to me no matter how I look at it, so I asked.

[2deaf]

I'm thinking you have a 250k MV pot, not 500k as shown in post #12. With a 500k pot you should see just a little less than 500k (from one outside lug to the other) when turned all the way down. No matter, you have what you have and I'm not suggesting changing it. It's just that I was assuming 500k.

There would be no reason to think I have a 250K MV pot when I specifically cited the stock MV which would be a 500K pot with a 470K resistor from the hot side to the wiper and a 220K resistor from the wiper to ground.  The math also does not indicate a 250K pot.  With the volume turned all the way down, there is only the 500K pot in parallel with the 470K resistor, which is 242K.  242K in parallel with the 100K plate resistor is 70.7K.

Let's say that you are still leaving out the 500K pot when the volume is turned all the way down, even though your words indicate otherwise.  That would leave just the 470K resistor.  470K in parallel with the 100K plate resistor is 82.5K.  82.5K is not very close to a maximum of almost 100K.

What I really want to know is what the grid stopper has to do with it. 

[Tony Bones]

Yes, you're right about the 500k v. 250k pot. My mistake.

The grid stopper has nothing to do with it when the MV is turned all the way down. My point was that it does have something to do with it when turned all the way up, but NOT as you turn it down. I was not clear in my explanation. Sorry.

[jjasilli]

@2deaf:  "There would be no reason to think I have a 250K MV pot when I specifically cited the stock MV which would be a 500K pot with a 470K "
Good grief, you don't have a pot, Bike Mike does!  Big Mike didn't post the schematic, kagliostro did.  Until the actual pot is measured, it's only a presumption that it conforms to the schematic.

"What I really want to know is what the grid stopper has to do with it."
The grid stopper has nothing, per se, to do with the issue.  It's true that the upper resistance of the pot is in series with the grid stopper, but not materially relevant.  Mere "dicta", as we say in the law.  It would have been more clear if TB explicitly said, instead of merely implied, that the upper resistance of the pot forms a voltage divider with the 220k grid leak resistor.  But this is a Forum conversation, not a graduate school exegesis.  It is not helpful to dwell on non-issues which do not help Big Mike fix his early volume problem.

[Big_Mike]

Thanks everyone for all the input.  I did not have a chance to sit down and put the amp through the paces and do an A/B comparison of the amp with and without the 470K resistor between lugs 1 and 2.  I really want to compare the low volume response of the pot doing the A/B comparison.  I emailed the original designer of the amp to see what his thinking was when adding the resistor, but have heard nothing back from him. 

I assume that keeping the cap in place, even if I remove the resistor, will still preserve more of the treble at lower volume levels?  That is, is it the combination of the resistor and the cap that preserves treble, or is it the cap only? 

I am hoping to do the A/B comparison tonight and will report back to everyone.

[2deaf]

@2deaf:  "There would be no reason to think I have a 250K MV pot when I specifically cited the stock MV which would be a 500K pot with a 470K "
Good grief, you don't have a pot, Bike Mike does!

TB didn't mean that I physically had a 250K pot and neither did I.  We both meant that I was using a pot of some denomination in my example.

Big Mike didn't post the schematic, kagliostro did.

Dicta

The grid stopper has nothing, per se, to do with the issue.  It's true that the upper resistance of the pot is in series with the grid stopper, but not materially relevant.

For the sentence in question, TB sets the initial condition to "As you turn down the master".  That means that we are at a setting other than maximum and that we are in the process of moving further away from maximum.  He goes on to state that the part between the input and the wiper gets put in series with the grid stopper.  He immediately follows this with the word "so" which means that what he says next follows from what was said before.  The conclusion of the sentence states that the load resistance seen by V2b [sic] goes up.  So the series combination of the upper pot and the grid stopper in conjunction with a decrease in master volume setting is responsible for an increase in the load seen by V2b [sic].  The conclusion also states that this effect will max-out when the master volume is turned completely down at which point the load seen by V2b [sic] will be almost 100K.

Because we are talking about the conditions at which the OP wanted to know about, this is particularly relevant.  The OP could easily come away believing that the grid stopper is affecting what happens at low master volume settings.  A discussion on this point is absolutely warranted in order to clarify what actually happens.

It is not helpful to dwell on non-issues which do not help Big Mike fix his early volume problem.

You may have an ulterior motive for scolding me. 

[jjasilli]

@2deaf:  The conclusion of the sentence [TB Reply 18] states that the load resistance seen by V2b [sic] goes up. 
No.  You need to read TB's whole sentence in context:  "if you change the 220k to 100k.  .  . the load resistance seen by V2b goes up to a maximum of almost 100k . . ."


This is patently true.  There can be no genuine disagreement.

[2deaf]

@2deaf:  The conclusion of the sentence [TB Reply 18] states that the load resistance seen by V2b [sic] goes up. 
No.

No?  This is word for word what he said.  You even quote the exact words in the very next sentence.

You need to read TB's whole sentence in context

The whole sentence is a separate paragraph so it has its own context.

"if you change the 220k to 100k.  .  . the load resistance seen by V2b goes up to a maximum of almost 100k . . ."

Here you have taken words from a different paragraph and inserted them into the pertinent sentence.  68K is in that other paragraph, also, so maybe that is the context.  Maybe something from a completely different post is the context.

Let's say that we do have a 100K from the wiper to ground instead of a 220K.  How does that change the validity of the sentence in question?  It doesn't.  Whatever the resistor is, it is completely out of the circuit when the master volume is turned all the way down.

This is patently true.

No it isn't. It isn't even latently true.  In order for the the total load seen by V2a to be almost 100k, the resistance that is in parallel with the 100K plate resistor would have to be almost infinite.  The most you could get would be 500K and that would be with the 470K fixed resistor removed from the pot.

There can be no genuine disagreement.

I genuinely disagree.   

[Tony Bones]

Guys, this discussion seems to be getting a little hotter than it needs to be, at least for EL34World.

2deaf, you are correct that with the master at minimum setting the load seen by the driver would not be "almost 100k." My mistake is that I had forgotten about the 470k resistor. Were it not there I would have been correct: 500k // 100k = 83k. With is debatable almost 100k.  In fact Big_Mike brought the mistake to my attention when he showed that the resistance across the MV at min was something just under 250k, not something just under 500k as it would have been without the 470k resistor (as I had been thinking.)

We could go through the entirety of post #18 and I could annotate and explain each word in context, but honestly, it seems like a molehill not worth dying on.

As for the V2b vs. V2a issue, I was going by the schematic that Franco so generously provided in post #12. Thank you Franco! The layout in fact show the driver tube connected to pins 1, 2, and 3, which properly and correctly is the 'a' section by anyone's definition. If I apologize for that grave mistake, will you forgive me? In fact, I have become sorry about the whole thing. No kidding.

So let's all just chill. Maybe go jump in the pool with a beer.

[2deaf]

No need to apologize to me.  You make a lot of good points on this and other forums.  I just wanted a clarification on one sentence and you gave that to me.  There was a little bit of noise in between and that is why the general policy here is "don't talk to 2deaf."

I had already drawn out my own schematic from the layout before Franco posted one.  Now that I look closely at the schematic, the stage in question is numbered 1,2,3, but it is labeled V2b.

[Tony Bones]

It's all good. I need to be more careful. If I can take the time to write a post I should be able to make sure that it's not full of errors.

I've spent decades working hard to get rid of my short term memory. Many late nights spent working drinking hard. I don't recall regretting any of it, but that's no excuse for being sloppy. I'll try harder. I mean to not be sloppy.

[Big_Mike]

I was able to play the amp last night at low volume without the 470K resistor across the master volume lugs.  I liked the tone well with that setup.  The pot has better control at that level, however, I would like even greater control if possible, so I am still open to possible tweaks to get better low volume control. 

[jjasilli]

Good news, here are various suggestions for more early vol control:


*  try the 470K bypass resistor between jugs 2 & 3 of the vol pot.  This might exaggerate the pot's audio taper.
*  put a 470K fixed resistor from vol pot lug2 (output) > 220K grid leak resistor.  the 470K fixed resistor might be bypassed by a 470Pf cap.  This will reduce gain.
*  use a 100K or 250K audio taper MV pot.  Smaller value pots give more control over pot rotation; but at the expense of overall signal loss. If signal loss is a problem at high vol levels, signal can be recaptured by putting a fixed resistor under the MV pot.  That resistor can be shorted by a SW to take it in and out of the circuit. 

[Ed_Chambley]

I did not know we had a "Don't talk to List."  Debates are good and I learn from them and sometimes they get heated.  Adults can step back and reassess.


Great Thread IMO.  I hate pots ramping up too much to fast myself, so I followed along.  I learned something I did not know and it only happened when the need to be correct reared its head.  That is the change in the pot actually causing a faster ramp up. I have read the Secret life of pots many times, but until this thread I never considered this specific issue.

[2deaf]

I did not know we had a "Don't talk to List."

No?  My mistake, then.

[2deaf]

If you want to take extraordinary measures, you could use a dual-ganged pot maybe something like this:

[Big_Mike]

If you want to take extraordinary measures, you could use a dual-ganged pot maybe something like this:

Could you provide a layout on how that is wired up?  I am trying to visualize the layout.

[2deaf]

This one should have better brightness and more adjustment at low settings.  I don't know how to make those layout diagrams on the computer, so I did a hand drawn one.

[Big_Mike]

Thanks for the layout, much appreciated.  I have an extra PEC 250K log pot on hand from another project, could I substitute it for the 500K pot?  That would give me better control on the low end, but would I have to do any mods in addition to swapping the pot?