Treble Bleed Questions re: Tone Problem

[slejhamer]

Hi again.

Since I'll be doing surgery on my guitar soon, I was thinking of installing a treble bleed mod while I've got it opened up.

HOWEVER, I had previously done the simple version by running a .001 cap across the volume pot, and found that a small reduction in volume made the tone seem trebly and thin, and it got worse as I further cut volume. (Humbucker pickups, 500k pots.)

I was thinking of trying a smaller cap value - maybe 680pf.

Then I came across some info on the "Kinman mod", which incorporates a 130k resistor in series with a .0012 cap. Intuitively, this makes sense to me - first control how much of the signal goes to the cap, so the tone should be more consistent as volume declines. I searched the forum but didn't find anything on Kinman, so I thought I'd ask about it in a new thread.

As I was getting ready to post, I saw a reference to a Seymour Duncan version using a 100k resistor in parallel with a .002 cap. This one I don't understand, but "they" say it works.

What do you think? How can I retain some treble when reducing volume, without making the tone too trebly and thin?

Thanks!

[sumgai]

slej,

It's all in how you look at it.

Let me be the first to say, Kinmans are the primo pickups on this planet. They cost more than almost anyone else's product, but they do sound so much more desirable than anything else. As the friend who introduced me to them said: "once you try these things, you'll throw rocks at everything else!"

For the record, I don't add anything to the volume pot, I think that what I get out of a bit of roll-off sounds just fine. You might say that, to to my ears, it already sounds 'correct'. But I do understand that others don't feel the same way, they want something different. Well, lucky for them, we have options.

First, your analysis of a series R/C (resistor and capacitor) combo is fairly correct, you do limit how much signal gets into that cap. But the problem with that is, the resistor stays the same value. As you crank the pot downwards, the ratio between the resistor and the pot's upper section changes. In turn, this changes the total amount of signal being byhpassed by the capacitor. How much? Enough for you to hear it, for sure. Effecitvely, the further you turn down, the greater the percentage of total signal will flow through the pot's upper segment. Less signal flow through the capacitor means less effect, which in turn should translate to "less tinny" as you roll-off the control.

What about a parallel R/C combo? Glad you asked. Here, we're directly changing the shape of the pot's taper, or response curve, if you prefer. Consider, for a moment, the resistor in parallel with the pot's two "upper" terminals, no capacitor at all.

What you now see is the pot's upper segment of the total resistance being bypassed by the resistor to the wiper, give us a 'resistors in parallel' situation. As you might recall, the formula for that scenario is R1 x R2 divided by R1 + R2. If we do some example calculations, we'll get an idea of what that external resistor is doing to the response curve. For our purposes, the resistor is 100Kohms, and the pot is 500Kohms, with a linear taper.

Pot turned fully up:
(wiper shorts to terminal, resistor is fully bypassed)

Pot turned fully down:
(wiper shorts to ground, signal is null and void)

Pot turned 10% down:
10% of 500K = 50K

50K x 100K = 5,000K
50K + 100K = 150K
5,000K / 150K = 33.33K

Hmmm, interesting. Instead of there being 50K ohms present in the upper leg, we now observe only 33K ohms. So what's this do to the bypass capacitor? Well, without going into a bunch of wizardly exotic formulas, it should be easy to see that once again, more signal is flowing through that 33K resistance than would flow through a 50K resistance, right? And for the purposes of the capacitor, which is in parallel with the 33K, it will get less signal to work on (at it's appropriate frequencies). Once again, we see that the effect of the bypass capacitor has been reduced a bit, theoretically giving us a less tinny sound as the pot is turned downwards.

The net effect is almost the same for either a series or parallel scenario, given a judicious choice of resistor and cap values. While we could derive "magic" cap and resistor values scientifically, I find that no part of me wants to think like a scientist while I'm playing! It falls to the individual to choose what sounds best to him/her by trial and error. The values you see listed in circuits like these are merely good starting points, that's all. Your amp/speakers, your cables, the newness of your strings, all of these things and so many more will influence the effect of this mod.

My advice? If you didn't like the cap by itself, then go for the series R/C combo first. Try a few cap and resistor values, and see what plucks your string. Don't like it? Try a parallel setup.

"But wait", you exclaim, "that's a lot of work". Nonsense, watch this: Leave all the controls in your ax turned up. Cut off the bad end of a 'broken' cable, then strip back the insulation from the wires on the good end. Plug that gizmo into your guitar's output jack, and now you've got we we tecnical types call a 'test jig'. You can use alligator clips to hook up pots, resistors, capacitors, whatever you want. Just "wire" them up in a temporary fashion, and run another cable to your amp. If you're carefull (don't rock the house down!), you'll be able to use this test bed for finding that exact combo that stiffens your pencil.

When you're done, put the appropriate components in your rig, and then rock the house down!!

HTH


sumgai

[JohnH]

My Cort LP, up until the time that I gutted it for a rebuild, had 500k pots, with 0.001 caps, in parallel with 180k resistors on each volume control. that seemed to work well as a passive design - and better than no bleed components.

The 180k resitor made the volume control a bit more sensitive for mixing pups. It was also better at low volume. at low volume, the cap is then feeding more treble than you need, the resistor compensates with some more bass, and also loads the pup a bit more which trims the treble. It was a good compromise.

I think there is no perfect combination of passive bleed components - and listening tests are the best way to judge. The source of the problem is the capacitance of the chord to the amp. Use a shorter chord for less treble loss and less need for bleed components.

On my guitars, I fit a small active buffer circuit, which solves the problem at all volume levels.

John

[sumgai]

Come to think of it....

slej, refresh our memories here.... are you dealing with a SSS/1vol/xtone, or DH/2vol/2tone setup?

My diagram above shows the "classic" way of hooking up a volume pot, and the bypass R/C network when that's being used. However, in their infinite wisdome, Gibson reversed the incoming and outgoing connections when they used one volume pot per pickup - this was to keep each control independent of the other. If you wire two pickups and two controls together using the above diagram for each control, then turning down one control all the way will kill all sound output. Not a good plan.

Be advised that the diagram is reversible, should you find the need to do so. However, the arithmetic involved does not change, you'll still get the same results I showed above, no matter which way the pot is oriented.


sumgai

[sumgai]

John,

Have you been listening to the Moody Blues again? It seems as if you've gone on a walkabout, In Search Of The Lost Cord!


sumgai

[slejhamer]

Hey - wow! - thanks for all that info sumgai and JohnH!

I'm gonna have to read the responses a few times to make sure I understand it all (but I think I have an idea.)

Yes, setting up a test rig is a great idea.

sumgai said:

Come to think of it....

slej, refresh our memories here.... are you dealing with a SSS/1vol/xtone, or DH/2vol/2tone setup?

The guitar I'm working on has (or will have) two humbuckers, 1vol, 1 tone. Presently it's 2vol/1tone, but I'm simplifying.

Another one that I might add this to has two singles (P90s), 1vol, 1 tone.

Thanks for the info on Gibson - hopefully that won't apply to my new setup.

[sumgai]

slej,

Thanks for the info on Gibson - hopefully that won't apply to my new setup.

Nope, sure won't.

Note that in reading other topics on this forum, we tend to refer to my above diagram as "forward wiring" the volume pot. Gibson schema wires it in what we call the "backwards" way. This is all due to some arcane wizardry and electronics mumbo-jumbo, you needn't worry about it. I only point it out so that you will realize what we're talking about when you run across it as you peruse other topics here.

HTH


sumgai