The TBX tone control

[JohnH]

The TBX tone control

I've been leaving this tone control design until after a few of the simpler ones, in case they helped gain some further insight.

What is a TBX tone control?

To start off, I should say that the following is based on finding out what I can, analysing carefully, but not actually having used a TBX control myself.

The TBX tone control is a Fender design, intended to enhance the action of a tone knob. It's a fairly old design now, having been introduced in the 1980's. I'm not aware if it is supplied stock on any current models, but is is sold as an accessory option from Fender, including the pot plus a cap and resistor which is part of the intended design.

Fender TBX kit

They describe its action like this:

The Treble Bass Expander (TBX) is a detented, stacked 250K / 1Meg control that enhances your tonal palette. It functions as a standard tone control from 1 to 5; after 5 it decreases resistance, letting more bass, treble, presence and output flow to your amp

'kay......

What it really seems to do is to provide somewhat of a normal tone control range, compressed into the lower settings, and at higher settings it transitions to a much higher resistance, moving through most of the extra range that a no-load pot can provide, but gradually instead of as a step. It doesn't add anything, but just takes away less at high settings.

The pot is special, and unusual. Physically, it is a dual-gang pot with a detent at the centre (at 5 if your knob reads 0-10, or 5.5 if you have a Fender knob that goes 1-10) . One half of the pot acts like a 250k log pot, but compressed into the range 0 to 5. So at 2.5 it is about 25k to ground, rising up to about 250k just before midway. At 5, it becomes no-load with infinite resistance, which is how it is from 5 up to maximum. At the detent, it is at the no-load position.

The other half of the pot has a very low resistance (<1k) from 0 to 5, then a linear rise up to about 1M at maximum.
Here is a very nice hand-drawn description of how it is wired, (with an alternative option):

TBX mods and details

It is designed to be connected like this, with a 0.022uF cap, and another 82k fixed resistor:

The whole combined circuit is wired to the same two points as a standard tone circuit, between ground and hot, usually before the volume control. 

Analysing the TBX response

I took some care with this, starting by watching a video by Phostenix, who measured resistances step by step on an actual pot. I've used values that he measured, so the 250k is really 230k and the 1M gets up to 950k maximum. I was also able to draw a graph to see that the 250k pot is really at about 10% at the 1/4 point of the pot turn.

So here is a set of calculated responses, wired as intended. It shows results together with a 250k volume pot and a single coil pickup, because this is a Fender gadget and that is where most likely it will be used.

There's a few things that come to mind when looking at the schematic and the responses (including, what were they smoking? and where can I get some?)

• With the 82k resistor in place, the centre detent position is not doing what one might expect, i.e. some kind of standard normal setting, but rather it is one that is significantly muted, more like a normal tone pot on about 6. So it doesn't seem like a place where one would want to have a detent. Instead, a more usual 'normal' position is not at the detent but at about 6 on the TBX pot.
  
  
• As you approach halfway, there is a step in the response, from say 4.9 to 5, as the smaller pot cuts out (see the dashed lines on the graph). It is not a huge step, but any tones within this range are not available.

• The 82k, wired past the cap, does actually reduce bass response by a tad at minimum, by about 1/2 db. You get this back at higher settings. Maybe this is supposed to be the 'B' part of the 'Treble and Bass Expander'?

Modifying the TBX

The most common mod that I have seen referenced is to increase that 82k resistor, so that the centre detent is a more desirable normal position. 220k works quite well. If wired in the same position to replace the 82k, the result is this:

Maybe better, but now the step in response on arriving at the detent is larger.

(Note the Blueguitar pdf link above shows a slight variation on this mod, in which the cap is not bypassed by the resistor. The response is similar but a bit higher, just under 1db greater at low frequencies and low knob settings.)

Alternatives to TBX

Let's assume that the intent is to give a smooth control range, extending up towards the no-load range. A simple alternative would be a no-load pot, maybe a higher value such as 500k, so the step at max is quite small. Here is the trace (posted before), with a 500k no-load pot, used with a single- coil pickup and a 250k volume pot.

The taper from 0 to 2.5 is actually about the same as the TBX, and the overall range is the same, but no odd mid-range steps. And half the cost!

Or, why was it necessary to use a dual-gang pot? why not just one pot that goes 0 to 25k at the quarter point, up to 250k midway and 1M at max? Would that be too hard to make?

What else can you do with a TBX pot?

So having found this gadget, and gotten bored or annoyed with it, what else would it be good for?

It turns out that it may be very good for a combined treble control and bass cut, with a neutral no-load, no-cut condition at the centre detent. There are a number of such schemes on the net, here is Phostenix's diagram, and more good info on the TBX, (including the video that I watched):

Phostenix' TBX mods and info

I have already posted on treble and bass cut circuits, and this arrangement should fit right in there.

Conclusion

Well, overall the TBX design seems maybe useful but a bit quirky and rather complicated and there seem to be easier ways to achieve what it does. I'd be happy to hear from anyone who has one though, and whether or not they like it.

Using it as a bass and treble control however, seems like a cool idea!.

cheers

John

[col]

Feb 28, 2014 22:32:53 GMT -5 JohnH said:

Or, why was it necessary to use a dual-gang pot? why not just one pot that goes 0 to 25k at the quarter point, up to 250k midway and 1M at max? Would that be too hard to make?

I don't see why it would be any more complicated than manufacturing a standard log pot, which is usually just three linear tacks in series, only approximating logarithmic scale.

Feb 28, 2014 22:32:53 GMT -5 JohnH said:

Using it as a bass and treble control however, seems like a cool idea!.

I think I would just as well use a dual concentric pot, and have the option of controlling both treble and bass, independently, or in varying combination.

[0wnyourtone]

Mar 1, 2014 16:21:57 GMT -5 col said:

Feb 28, 2014 22:32:53 GMT -5 JohnH said:

Or, why was it necessary to use a dual-gang pot? why not just one pot that goes 0 to 25k at the quarter point, up to 250k midway and 1M at max? Would that be too hard to make?

I don't see why it would be any more complicated than manufacturing a standard log pot, which is usually just three linear tacks in series, only approximating logarithmic scale.

Feb 28, 2014 22:32:53 GMT -5 JohnH said:

Using it as a bass and treble control however, seems like a cool idea!.

I think I would just as well use a dual concentric pot, and have the option of controlling both treble and bass, independently, or in varying combination.

Agreed,
I've been tempted to do a bass-or-treble cut with a TBX, but you would miss one of the key benefits of the bass cut IMO - the interactivity with a treble cut!

- Also, John is really finding his tone as a writer with this series. I see best-seller in the near future.

[davidofchatham]

I'm very interested in using the TBX control as a bass/treble control in my one-volume, one-tone Les Paul Special–ish guitar. I found this forum searching for bass cut/contour circuits, but found this after I got here. I know that many who use those say that it's more powerful to have separate treble and bass controls, including 0wnyourtone above, but if I were going to add a second tone knob to my guitar, I'd rather have separate tone controls per pickup. (I'm not a fan of the look and feel of the concentric knobs I've seen so far, by the way.) Is anyone else actually using it as suggested by Phostenix?

Also, I can't find any physical measurements for it; does anyone know the maximum body depth (from surface to cavity) that it would be compatible with?

[JohnH]

Hi David, thanks for your message. I don't actually have a TBX myself, but from he photos in this thread, it looks to be made from an adaptation of a standard dual gang pot, which i think looks to be a full sized ie 24mm one. So dimensions for that would be a reasonable starting guess unless better info is found.

[davidofchatham]

Sept 1, 2021 16:34:23 GMT -5 JohnH said:

Hi David, thanks for your message. I don't actually have a TBX myself, but from he photos in this thread, it looks to be made from an adaptation of a standard dual gang pot, which i think looks to be a full sized ie 24mm one. So dimensions for that would be a reasonable starting guess unless better info is found.

JohnH, thanks for replying! Your posts here have been highly informative.

• Do you concur with Phostenix's values for the modified TBX bass cut circuit, .001uF on the bass side and .022uF on the treble side?
• And would it pair best with a 250k volume pot (with treble bleed), like the regular bass cut circuit? This will be used with humbuckers and I want to keep them on the brighter side.

By the way, I humbly suggest that you update the link in the original post to the GN2 copy of Phostenix's TBX page, which still has its images.

Thanks!

[JohnH]

Sept 2, 2021 10:31:13 GMT -5 davidofchatham said:

Sept 1, 2021 16:34:23 GMT -5 JohnH said:

Hi David, thanks for your message. I don't actually have a TBX myself, but from he photos in this thread, it looks to be made from an adaptation of a standard dual gang pot, which i think looks to be a full sized ie 24mm one. So dimensions for that would be a reasonable starting guess unless better info is found.

JohnH, thanks for replying! Your posts here have been highly informative.

• Do you concur with Phostenix's values for the modified TBX bass cut circuit, .001uF on the bass side and .022uF on the treble side?
• And would it pair best with a 250k volume pot (with treble bleed), like the regular bass cut circuit? This will be used with humbuckers and I want to keep them on the brighter side.

By the way, I humbly suggest that you update the link in the original post to the GN2 copy of Phostenix's TBX page, which still has its images.

Thanks!

Thanks for the link suggestion, I've made the change. In 2020 we were very happy to have the opportunity to mirror the Phostenix archive here on GN2, together with all the diagrams.

On the values, given that with this TBX mod you have a no-load setting available, then I would say that for most humbuckers, a 250k volume pot will be great. The pickup loading and tone with a 250k and a no-load at max is identical to a normal wiring of 500k volume and tone. i have a no-load and 250k on my HSS guitar (humbucker is 8.4k) and it sounds great.

If you do that, then on the bass cut, you might like to try a bigger cap such as 1.5nF or 2.2nF to compare to 1nF, totally a personal choice and its worth setting up a test before deciding. The 250k volume pot is part of making the bass cut work. The smaller the bass cap, the higher the frequency the bass starts to roll off (see the PTB thread in this sub forum)

But if you prefer to stick with a 500k volume pot, then that may suggest the smaller bass cap value, since a larger volume pot value will reduce the bass cut. The 0.022 treble cap is a pretty common value, and its a free choice to suit your preference.

[dylanhunt]

Thanks so much for the analysis JohnH.  It was very helpful in my figuring out what is going on with the TBX mods and understanding how the variations impact the sound. [For anyone who saw my confused and since deleted post, I'm going to take another swing at this.]

After wiring up two TBX kits, one according to the Fender diagram, as a repair job to get a Strat Plus back to original specs, and one with a couple of changes to Phostenix's ingenious mod, I decided I better understand the mod better.  The two changes to Phostenix's mod, in a guitar with a 500kOhm volume pot, a single tone pot (i.e., the TBX pot with modified wiring), and strat-style single-coil pickups, were:

1) use of a 200kOhm resistor, which is (roughly) what blueguitar uses (with different wiring), rather than Phostenix's 1MOhm resistor--in parallel with the main capacitor (0.022uF)
2) use of a 0.0022uF secondary cap on the 1MOhm gang, instead of a 0.001uF secondary cap.

Notice that the difference between the main cap and the secondary cap is an order of magnitude.

Having implemented that mod, I didn't get what I expected.  The first half of the sweep seemed to do very little, and then after the detent, things got a little strange: There was an abrupt change to the next range of the sweep and then, just a minute turn later, lower and mid frequencies dropped out completely; so much so that I thought the volume dropped to zero (I have neighbors and the amp was on low volume). It turned out that only high frequencies were preserved and that there was still output, but there wasn't much difference from 6.5 to 10, suggesting that my curve for tone values 6-10 would look like JohnH's plot for tone values of 8-10 on the stock (as intended) wiring.  In the space between the detent (5.5?) and 6.5, the mids and lows dropped off rapidly.

[I wondered what was going on--Was it because of my mods to Phostenix's mod?-- Was it because I had a 500kOhm volume pot instead of a 250kOhm?--Did I perhaps wire things backwards with respect to the bottom and top gangs?  Probably no one else who used Phostenix's mod wondered if they got the diagram backwards, but if you did... I have since decided that Phostenix's statement should have been phrased "The modified TBX control still has the 250K no load pot attached to the output of the pickup selector switch via a jumper from the 1MOhm gang to the 250kOhm gang", because his diagram states, "1M ~Log on Top / 250K Audio on Bottom", while his (and Fender's stock) diagram has the pickup selector switch output going to the 1M gang. If his original statement serves as the key to interpreting the diagram, there's likely to be some confusion about what he means by the "bottom" (does he mean the lugs farthest from the diagram's foreground, which is where the pickup selector actually connects, or does he mean conventional bottom row adjacent to the metal casing?): A or B in the diagram.]

From what I can understand from the schematic, the diagram looks accurate. So, I tinkered a bit more.

a) I used a variable resistor to add 800kOhms back to the parallel connection with the wired main cap and 200kOhm resistor--eliminating that difference between my mod and Phostenix's. [Unfortunately, I "added" 800k resistance back in by means of a parallel connection, which means (as frets pointed out) the resistance is not 1000k, but 1/x, where x= 1/800 + 1/200; so, the resistance is 160k and nowhere close to Phostenix's impedance value.]
b) I replaced the 0.0022uF secondary cap with a 0.001uF cap, as in Phostenix's design.

I got more or less the same results: very little, if any, response from the 1-5 sweep range and very little response, if any, from 6.5 to 10.  Keeping my 500kOhm volume pot and TBX specialty dual gang tone pot constant, I tried the following combinations with little difference, compared to my mod of Phostenix's mod:

I)    0.022uF main cap in parallel with 1M 160 Ohm resistor, and 0.0022 secondary cap
II)   0.022uF main cap in parallel with 1M 160k Ohm resistor, and 0.0010 secondary cap
III)  0.022uF main cap in parallel with 200k Ohm resistor, and 0.0010 secondary cap
IV)   0.032uF main cap in parallel with 200k Ohm resistor, and 0.0010 secondary cap
V)    0.044uF main cap in parallel with 200k Ohm resistor, and 0.0010 secondary cap
VI)   0.069uF main cap in parallel with 200k Ohm resistor, and 0.0010 secondary cap
VII)  0.032uF main cap in parallel with 1M 160k Ohm resistor, and 0.0010 secondary cap
VIII) 0.044uF main cap in parallel with 1M 160k Ohm resistor, and 0.0010 secondary cap
IX)   0.069uF main cap in parallel with 1M 160k Ohm resistor, and 0.0010 secondary cap

Changing the main cap for one with a larger value (IV through IX) helped the 6.5 to 10 sweep, but only for the neck pickup.  I was starting to get the sense that I would be better off with a three way switch instead of a modded TBX pot since that was roughly what you got with it: three distinctly different settings.

It wasn't until I changed out the secondary cap for larger values (0.033uF, then 0.047uF [I'm not sure what else as I seemed to misplace a decimal places]) that I started to see a major difference in both the 1-5 sweep and the 6.5 to 10 sweep. The best combination with a 500kOhm volume pot seemed to be:

X) 0.047uF main cap in parallel with a resistor in the range of 500-750kOhm, and a 0.047uF secondary cap. (Notice again how the difference between caps is a neat order of magnitude, while the resistance is half what Phostenix suggested for a 250kOhm volume pot.)

With that combination, I got a pretty even change across the 1-5 sweep range, a smooth transition from the detent to the 6-10 range, and an even change across the 6-10 sweep range.  My guess is that this difference has something to with the value of my volume pot, but I'd have to use a 250k or 1MOhm volume pot to find out.

[dylanhunt]

[Full disclosure: I'm not working in a laboratory environment and also, apparently, I only use 2% of my brain.  I posted before peer review; well, before any review and I'd be laughing if I could be sure no one had seen the previous version of this post.  After my talk about orders of magnitude in my post above, while doing the tests with the 1M volume pot, I accidentally mixed caps of different orders of magnitude for the bass filter without realizing they were 10x higher than I thought.]

I've replaced the 500k volume pot with a 1MOhm volume pot and I think I can suggest some relationships now:
1) The higher the resistance of the volume pot, the higher the value of the secondary (high-pass/bass filter?) cap should be.  Getting this right seems to manage the transition from the detent to the upper sweep (too low and you get an abrupt transition).
2)  The higher the resistance of the volume pot, the higher the value of the parallel resistor should be.  Getting this right seems mostly to manage how abruptly the lows and mids were filtered on the top sweep range, but it seemed to limit how much treble was filtered on the bottom sweep range too. [Retracting this due to using the calculation for series, as frets kindly pointed out, rather than the parallel calculation that was appropriate for the connection I was using. Instead of comparing 1.2M Ohms (much higher) to 200k, I was comparing 160k Ohms (a little lower) to 200k.] 
3)  The value of the main (low-pass/treble filter?) cap didn't seem to be that finicky: whether I had a 500k volume pot or a 1M volume pot, the 0.022uF cap seemed best (0.047uF seemed muddy to me, and 0.01uF was probably not effective enough).  It does seem best for the low-pass/treble filter cap to be a little smaller in value than the high-pass/bass filter cap.  Who knows what is going on with this or why it sound so smooth (transition-wise) and good.

So, the question remains: If using capacitors of values 0.022uF to 0.047uF as usual for the low-pass filter, what value for the resistor in parallel with that cap, and what value for the high-pass/bass filter cap?  By (happy?) accident, I was using caps for the high-pass/bass filter that were 10X more capacitative than I thought.  So much for relationships (well, expecting to find the wrong kind of patterns is why people thought for so long that the sun had to orbit the earth)!  The strange thing is that the higher-value capacitors smoothed out the transition from the detent to the upper sweep range.  I'm not sure if there is a con to that pro.  What did I lose by gaining smoothness?

The resistors for the low-pass/treble filter should be roughly 1:1 with the volume pot value (250k:250k, 500k:500k, 1M:1M).

For a 1M volume pot, I would go with the following:
Either this--
A1) 0.022uF cap (#223[0.022]) in parallel with
A2) 200k Ohms resistor
B)   0.057uF cap (#473[0.047]+#103[0.01])

or this--
C1) 0.047uF cap (#473[0.047]) in parallel with
C2) 200k Ohms resistor
D) 0.068uF cap (#683[0.068uF])

The following are not suggestions because I no longer have my 500k pot in the guitar, so they're more hypotheses--
For a 500k volume pot,:
E1) 0.022uF cap in parallel with
E2) 150kOhms resistance
F)  0.047uF cap

For a 250k volume pot (though I tend not to use them):
G1) 0.015uF cap in parallel with
G2) 100kOhms resistance
H) 0.022uF cap

The test guitar has only two pickups, one in the neck position and one in the middle (until I put the 1M volume pot in, I thought the bridge position was unplayable--too few harmonics: more hammer on iron plate than hammer tapping on a bell, but I think I will move the middle pickup back to the bridge position and go back to having the middle occupied with a dummy pickup, which is currently sitting in the bridge position).

Note: I like lots of treble in my pickups, period, and in my guitar output signal as well since I can reduce treble with EQ on the other side of jack or using the amp controls. With a regular tone circuit, I tend to keep my tone at the three most treble increments of the sweep (10, 9, 8, 7) for my neck pickup, and near the bottom of the sweep for my bridge pickup (5, 4, 3).

With the configuration described above, I can get usable settings on the tone knob from 1 to 10.  That's far better than a regular tone circuit, even if I used twice the sweep on a standard tone pot that I do for any given pickup.

[Yogi B]

Note: this was began before the latest reply, and as such responds solely to the message preceding that.

I'm finding this all a little odd...

Oct 11, 2022 14:10:21 GMT -5 dylanhunt said:

After wiring up two TBX kits, one according to the Fender diagram, as a repair job to get a Strat Plus back to original specs, and one with a couple of changes to Phostenix's ingenious mod, I decided I better understand the mod better.  The two changes to Phostenix's mod, in a guitar with a 500kOhm volume pot, a single tone pot (i.e., the TBX pot with modified wiring), and strat-style single-coil pickups, were:

1) use of a 200kOhm resistor, which is (roughly) what blueguitar uses (with different wiring), rather than Phostenix's 1MOhm resistor--in parallel with the main capacitor (0.022uF)
2) use of a 0.0022uF secondary cap on the 1MOhm gang, instead of a 0.001uF secondary cap.

Having implemented that mod, I didn't get what I expected.  The first half of the sweep seemed to do very little

Assuming "the first half" is 1 to 5.5, this is weird.

At the detent: the bottom gang of the pot (that which you labelled B) should be completely open with no continuity between any of the terminals; the top gang should have low-ish resistance (at most a few kiloohms) in parallel with the "secondary" (high-pass) cap, and have about 1 megaohm between the middle terminal and the unused third terminal. In other words, the modded TBX is effectively bypassed and should have virtually no effect on the signal.

Going from that at the detent down to having the signal grounded via the "primary" 22nF cap (and parallel resistor) when at "1" is similar to the range of a standard tone control and should be very noticeable.

then after the detent, things got a little strange: There was an abrupt change to the next range of the sweep and then, just a minute turn later, lower and mid frequencies dropped out completely; so much so that I thought the volume dropped to zero (I have neighbors and the amp was on low volume). It turned out that only high frequencies were preserved and that there was still output, but there wasn't much difference from 6.5 to 10

This makes some sense, the upper gang is similar to the M or N (I forget which is which) taper of a ("true") blend pot: between lugs 1 & 2 it stays close to zero resistance from "1" to the centre detent, past that point it increases pretty linearly up to its maximum of around 1Meg. Arguably this makes the increase in resistance too steep at the beginning (at "6" or "7") and either a log taper or lower overall resistance would be better suited for the purposes of the mod.

Since you describe too much mids being cut, the remedy for that would be a larger "secondary" cap (which is ultimately the solution you arrived at). But whilst there will always be some volume reduction with a bass cut (even some reduction of treble), the combination of the bass-cut dropping so much volume and a 22nF treble-cut cap doing seemingly little makes me think you have a particularly dark sounding guitar.

I)    0.022uF main cap in parallel with 1MOhm resistor, and 0.0022 secondary cap
II)   0.022uF main cap in parallel with 1MOhm resistor, and 0.0010 secondary cap (Phostenix's mod)
III)  0.022uF main cap in parallel with 200kOhm resistor, and 0.0010 secondary cap
IV)   0.032uF main cap in parallel with 200kOhm resistor, and 0.0010 secondary cap
V)    0.044uF main cap in parallel with 200kOhm resistor, and 0.0010 secondary cap
VI)   0.069uF main cap in parallel with 200kOhm resistor, and 0.0010 secondary cap
VII)  0.032uF main cap in parallel with 1MOhm resistor, and 0.0010 secondary cap
VIII) 0.044uF main cap in parallel with 1MOhm resistor, and 0.0010 secondary cap
IX)   0.069uF main cap in parallel with 1MOhm resistor, and 0.0010 secondary cap

Changing the main cap for one with a larger value (IV through IX) helped the 6.5 to 10 sweep, but only for the neck pickup.

This does not make any sense. The cap you're changing (the "main" cap, treble-cut cap) should be completely disconnected from the circuit within the range you specify — at the detent the 250k gang should go open, and remain open up to "10" — thus changes to its value should have no effect within this range.

My guess is that this difference has something to with the value of my volume pot, but I'd have to use a 250k or 1MOhm volume pot to find out.

The volume pot value will predominantly affect the bass-cut. In isolation, looking at just the bass-cut cap & volume pot resistance, upping the resistance lowers the cutoff frequency (a lower selection of bass frequencies are cut) — so in order to maintain the same cutoff frequency the cap value should be lowered.

Related to this is a further tweak you can make to a bass-cut (provided it is wired a certain way, which this happens to be) — add a resistor between the unused lug (closest to your "A" label) and ground, this is 'blended' in parallel with the volume control as the bass cut is increased. This allows you to use a larger capacitor than usual, in order to get less cut near the 'flat' setting (in this case at the detent) but more cut towards the extreme — somewhat helping with the unideal linear taper. For starting values I'd recommend something in the range of 100k—200k and experiment from there.

---

Finally a quick note about the resistor, in the BlueGuitar Modified TBX Wiring the 220k resistor acts to prevent the centre detent from being no-load, somewhat smoothing the transition and ensuring that the upper half of the pot has a purpose. With it: the resistance in series with the tone cap increases with a somewhat logarithmic taper from 0k at "1" up to about 120k at about "5", then jumps to 220k at the centre detent, and increases linearly to 1.22Meg from "6" upwards. Without: the resistance would be around 250k at "5", jump to no-load at the detent and remain at no-load all the way to "10".

Whereas in the wiring of Phostenix's mod the resistor placement does not allow for such smoothing, its purpose is simply as pop reduction — giving a path by which any charge stored across the "main" capacitor can equalize once it is disconnected from the rest of the circuit at (or above) the detent. Since the resistor is directly in parallel with the cap, decreasing its value should do little aside from reduce the overall volume when the TBX is turned all the way down to "1".

[dylanhunt]

At this point, I can't be sure I didn't make a bunch of errors along the way.  My original build didn't include the jumper wire between gangs, but when I added it, I didn't notice a difference.  I think if anyone is willing to look into my work and explain the results, those results, rather than the unsound process by which I got there, would seem to be the only thing worth discussing at this point.  I could have been a little more systematic about my experimenting, and a lot more careful.  So, I don't feel confident about hardly anything I wrote in that first post (#7).  The second post seems solid.

It sounds heavenly, with a great treble cut and a great bass cut.  With the Seymour Duncan SSL-52 neck pickup, the APS-2 pickup, the 1M volume pot, and the modded TBX circuit, I don't think anyone could guess from that sound that it's a $300 (+$200) Squier Paranormal Cyclone and not an American Professional Stratocaster at three times the cost.  It also has reverse-phase switching and series/parallel for two pickups built into the three Jaguar-like switches.

And here's the final post-pickup/post-switch segment of the circuit:

[dylanhunt]

Oct 11, 2022 22:19:25 GMT -5 Yogi B said:

Since you describe too much mids being cut, the remedy for that would be a larger "secondary" cap (which is ultimately the solution you arrived at). But whilst there will always be some volume reduction with a bass cut (even some reduction of treble), the combination of the bass-cut dropping so much volume and a 22nF treble-cut cap doing seemingly little makes me think you have a particularly dark sounding guitar.

Thanks for taking the time to read that first post, which probably doesn't deserve more attention at this point, and thanks for your thoughful comments.  I'm not sure I want to get into a discussion of the psychoacoustics (because I am of the opinion that what goes out of the output jack is 90% due to the pickups and tone circuit rather than other aspects of the guitar construction).  I also don't mean to highjack JohnH's thread and change the topic, but I did want to say that there might be something to your idea that it is a particularly dark guitar.  The sound of the stock pickups were certainly such as to make me think they might have high capacitance.  The replacement pickups were much better, but the sound nonetheless made me feel like the strings were too tight (they were new, but I think there was more to it than that).  Moving from a 250k colume pot to a 500k pot and then to a 1M pot, however, made all the difference.  Therefore,  I'm not sure how to evaluate the concept of "a dark guitar" at this point, except in terms of what value of volume pot is required to make good strat pickups (like the SSL-52) sound clear and stratty (and other than a difference in scale between a Strat and the Paranormal Cyclone).  Of course, that would be question-begging of me: Something physical must cause the conditions for which I (or someone with similar aesthetic senses) would judge a 250k, 500k, or 1M to be sufficient for the guitar.

I suppose the relevant point that falls out is whether my results with modding the TBX will turn out to be representative or only useful for people with particularly dark guitars like mine.  The other TBX I installed, according to Fender's instructions, was on a Strat Plus with Lace Sensors.  Different scale certainly.  Different pickups.  Who knows?

[frets]

Dylan,

Interesting analyses. In the diagram you have 1M in parallel with 200k. That would equal 170k not 1.2M.

[dylanhunt]

You're so right, frets... and too kind.  Well, the analyses might be interesting, but not in the kind way you intended.  That looks like a small error: use the series calculation for resistors in parallel calculation.  In this case, the result is that I'm not sure any of the "analyses" have value.  I guess that's what a forum is for, but maybe GN2 would be better served if John H moved Posts #7 onwards to a new thread.  Maybe I'll try again, more carefully, as I have another of the same model of guitar that I'm installing a TBX on.

[frets]

Dylan,
Your analyses and results still have much merit as you are reporting on what “sounds” good. And as far as I’m concerned, “sounding right” is “what matters most.” So having a 170k in parallel with the .047 is the best combination to you and could be for many others. Everybody on the forum is open to various analysis. You ought to see the stuff I’ve put up. So please don’t come away from this thinking what you did was wrong. It was right because it sounded right to you.

And, because there are many on this forum with far more electrical engineering/sound engineering experience, they may point out that I am wrong given the resistance introduced by the pot itself or something I failed to consider. And, one has to include the pickup dynamics as a crucial variable in what makes a particular tone circuit sound good.

So know your circuit has just as much merit as anyone else’s😸😸😸😸

[stevewf]

By the way, here are some photos that might help answer the question about physical dimensions of the TBX:



...and also of a more standard CTS concentric dual-ganger, since I'm moving photos:


The background grid is in inches. I tried to line up the bottom of the bushing with a gridline (where the inner face of a pickguard would be).

I'm seeing that the TBX needs about 22mm of cavity depth. In SAE, that's very close to 7/8".

As an aside ("aside", like in the photo; see what I did there? ), the concentric dual-ganger's depth is the same.

Concurrence: I put that TBX on the multimeter, and it did behave as documented.

Finally, and perhaps more germane: While there are codes stamped into the metal bodies of each, there's no "human-language" identifier to tell what's what:

Maybe those codes tell us what's what, but I've decided to put labels on 'em (hope I don't forget, and melt the labels when soldering).

[JohnH]

Oct 12, 2022 14:18:40 GMT -5 frets said:

Dylan,
Your analyses and results still have much merit as you are reporting on what “sounds” good. And as far as I’m concerned, “sounding right” is “what matters most.” So having a 170k in parallel with the .047 is the best combination to you and could be for many others. Everybody on the forum is open to various analysis. You ought to see the stuff I’ve put up. So please don’t come away from this thinking what you did was wrong. It was right because it sounded right to you.

And, because there are many on this forum with far more electrical engineering/sound engineering experience, they may point out that I am wrong given the resistance introduced by the pot itself or something I failed to consider. And, one has to include the pickup dynamics as a crucial variable in what makes a particular tone circuit sound good.

So know your circuit has just as much merit as anyone else’s😸😸😸😸

Absolutely agree, what sounds right is right, and if it sounds wrong then it is!

Some (like me) like to analyze stuff, because like, that's what I do. When it comes to music and audio circuits, I see that such analysis can help us in a few ways, which are all secondary to the real test which is by playing and listening:

1. it helps us understand how something good works, so maybe it can be made even better, or simpler, or adapted to a new situation
2. If listening reveals some issue that may imply that a change could be desirable, the analysis can help us home-in on that and figure out things to try
3. It might help us explore the parameters and the range over which they work, and what quirks may occur so that we can listen for them
4. If we cant test something in real life, the numbers may help predict the outcome, or help select from a wide range of options to test just a few.

[dylanhunt]

I appreciate the support, frets and JohnH.  I can be an idiot on one or two occasions and not feel too bad about it (of course, if I continue to make the same idiotic mistake, then I'm more likely to feel bad about it).  I'm mainly worried about the guitar modder who comes to JohnH's nicely set up thread and finds it in decline after Post#7.  Here's where I'm at currently in this object lesson of a tragicomedy(?) of errors:

With three components (treble-filter cap, treble filter resistor, and bass-filter cap), each having at least three potentially significant variations, there's a lot to keep in your head as you make changes.  If you're being methodical about it, there are 27 physical states to set up experimentally and to which each of your data log records (e.g., rows on a spreadsheet) need to be tracked (i.e., put the recording for Cap A in one of the entries for Cap A).  27 test states is a lot, compared to hearing how a 250k Ohm volume pot sounds relative to a 500k Ohm volume pot (which is only two test states).  Even setting up the spreadsheet alone for 27 states consumes a fair bit of time.  I just want to know the final results, finish settting up the guitar the best way possible (for me), and go play it. . . but I also want to know that I learned something true in the process (and not something resulting from fallacious reasoning).

Given the number of test states involved, my first attempt was opportunistic: I tried to look at only what I was interested in and tried to figure out what the test states I was interested in implied about the TBX tone circuit without being systematic.  The value of theoretical relationships like "As the treble-filter cap increases in capacitance value, the filtering of treble frequencies increases" is that we can pay less attention to individual sampling points in the treble-filter range (but only after those relationships are demonstrated) and move on to another dimension.  As you start to understand the relationships for one dimension (and between components), you can move on and ignore the previous data points that were not important.  That was the up-side, or at least the idea of the up-side of my rough-and-ready approach.  The down-side of not being systematic in setting up the experiment is that it's easier to make mistakes in your method and what you write down as data.

The opportunistic approach was a disaster: I didn't always check closely what value of component I was using and therefore didn't record the right value of the component, and also because I was adding resistors to an already complete tone circuit, and adding them in parallel while calculating the series resistor values.  Having systematic test states laid out would have ensured the method and recording were systematic too.

With 27 test states, if you want to listen to (and log) each of the 10 marked positions on the tone knob, you're going to have 270 combinations, but I'm looking at only three or four sample points/ranges per test state.  These sample ranges are:

A) high-filter extreme (tone knob value at 1.5 or 2)
B) transition from just before the detent to the detent (tone knob value range of 4.5 to 5.5),
C) transition from the detent to just after the detent (tone knob value range of 5.0 to 6.5), and
D) low-filter extreme (where mids/lows are mostly removed).

That brings the number down from 270 to 108. However, because I was leaping to conclusions during the first attempt, I feel I need to (dis)confirm some of the inferences I made.  This means that there are 4 resistor values of interest and 5 bass-cap values of interest.  So, with 4 tone-knob ranges, that means I'm up to 240 possible results (i.e., rows on the spreadsheet).  One good thing is that patterns should emerge quickly if I'm recording things accurately (now that there is a clear structure to the test scenarios).  For the components, here are the variations of interest--

Treble-filter cap values (1nF=0.001uF): 22nF, 33nF, and 47nf.
Treble-filter resistor values (in Ohms): 82k, 170k, 220k, and 1M.
Bass-filter cap values  (1nF=0.001uF): 2.2nF, 10nF, 22nF, 47nF, and 68nF.

[Yogi B]

Oct 15, 2022 16:06:39 GMT -5 dylanhunt said:

With three components ... each having at least three potentially significant variations ... there are 27 physical states to set up experimentally

If you're using the Phostenix layout (not necessarily component values, but layout) — there isn't 27, there's only 12.

With Phostenix's wiring, the gangs of the TBX pot are almost entirely independent. At and above the centre detent the 250k pot in the treble cut side is open, so the treble filter cap & resistor are completely removed from the circuit during experiments with the other (bass cut) side. The reverse is also essentially true, at and below the centre detent the 1Meg pot is at very low resistance (in Phostenix's measurements, less than or equal to 1.2k) — with Phostenix's 1nF suggested bass cut cap, standard loading (470pF cable capacitance & 1Meg amp input impedance), plus your 500k volume (at maximum) — the difference between in response between the parallel 1.2k & 1nF and the ideal scenario of zero ohms is a less than a third of a decibel reduction in the resonant peak, in other words inaudible. This difference is even less with larger caps: as low as a tenth of a decibel with your largest bass cut cap (68n).

Combining this with your four test ranges: C & D are only affected by the choice of the bass cut capacitor; whereas A & B are only (non-negligibly) affected by the choice of treble cut cap & resistor.

[dylanhunt]

Thanks, Yogi.  As always, your thoughts are (reliably) exceptionally helpful--especially in moving from the physical facts to the intuitions about sound.  Unfortunately, I've broken my intuitions about tone circuits, and started with very little knowledge to begin with.  In this case, I am moving from a point where I no longer trust my ears and no longer have any sense of how resistance and capacitance are working to produce a given sound.  So, I really need to see the relationships myself... but I probably will start with the 48 cases your comment implied, rather than the 108 I calculated.  The goal now is to leave my ears outside the picture and to visualize it in something like the plots I've seen on this forum.  Unfortunately, nothing I've tried with my USB scope yields anything like those plots, either measuring a single pickup, or a pot or two, or the signal from the output jack.  I don't think the scope is broken, but it isn't giving me anything like I see here. 

JohnH (or Antigua or anyone else who's been posting bode plots): I'm missing a little bit about how you went from Phostenix's video to the bode plot, though I understand the bottom (x-)axis is frequency and the side (y-)axis is decibels.  I don't really understand what is going on with the +/- dB/octave talk, but maybe that is a subject for a different thread.  I'm using a Digilent Analog Discover 2.  I've read antigua's posting about measuring pickups and antigua's short review of the AD2.  I think most of you posting plots are using a Velleman unit, but how much of what is going behind the scenes involves spreadsheets formulas, importing .csvs, or other steps that aren't mentioned?  Is there a procedure for getting the interface to produce something like the plot at the beginning of this post, or does the tool do that automatically?  How did you go from data you gleaned from Phostenix's video to the plots?  If it's a lot more work to describe than I am anticipating, please feel free tell me to sod off.

[JohnH]

Hi Dylan

Just on the question about plots: The plots that I do are calculated, mostly using a spreadsheet or sometimes using analysis software. The maths for doing this when there are capacitors and coils etc, is well established though it takes a bit of work to get one's head around it. So for working things out about this TBX pot for example, Phostenix' video showed how the real pot resistances varied at each setting (which is very unusual on a TBX), and that allowed me to code that into to my spreadsheet in order to estimate what it will do in various circuit configurations.

But we also have really excellent physical testing plots on GN2, such as by Antigua and others on pickups. Such techniques can also be applied to the tone circuits such as TBX, starting by rigging up the physical circuit, with a pickup too. These tests can look similar in shape to the theoretical plots, and if they do, then the tests and analyses support each other.

[dylanhunt]

Thanks JohnH.  It looks like this involves a skillset I didn't anticipate needing.  I shelled out $200-300 for a USB scope expecting to be able to plug a pickup or tone circuit into it, much like you would measure a pot's or pickup's resistance with a meter.  It has turned out to be not exactly a rabbit hole as much as a newly seeded lawn dotted with gopher holes.  

You once commented that I must like puzzles.  You might have a different opinion now.  I was content to work on the problem of how to build a switching system for going between parallel and serial circuits for three transducers for months and months on my own until I got it.  Somehow the challenge of getting a USB scope to "work properly" is more demoralizing.  Perhaps I have been spoiled by all the help GN2 has provided me with wiring problems, but it would be a great service if GN2 folks could lay out the methods of measurement so that others could emulate them.  I realize that might be a lot of work, but those instructions could be made available at a charge that compensated for their time if they were willing.  I have been laughing for a year or more at the kind of responses that, say, antigua's graphs on squier-talk engender among that community.  If I feel like watching Idiocracy but cannot find it on a streaming service I subscribe to, I know the comments antigua's analyses produced on squier-talk by those who rely on their ears and reject what their eyes tell them are freely available for a laugh.  At this point, I have to confess to myself that I can't even claim to know what those plots are supposed to represent if they're not a result of direct measurement by a scope (measuring, say, how output voltage varies with the frequency of a signal).  At least, I can't seem to figure out how to get my USB scope to measure resonant peaks directly.  All I get is a slowly rising slope that keeps rising off the right-end of the graph (even beyond 50kHz), rather than a Matterhorn-like rise and fall. 

[nuke]

I thought I'd chip in this bit of info. This the TBX tone control from a 1984 Fender Esprit Elite. The instrument was part of Fender's "Master Series" which were the Esprit, the Flame and the D'Aquisto jazz guitar. The Esprit and Flame are chambered guitars with alder bodies and plain maple tops. The pickups are Schaller humbuckers, guitars came in 3 trim levels, Standard, Elite and Ultra. The Esprit and Flame standard had conventional tone controls, the Elite and Ultra had TBX, pickup coil splitting, and more ornate trim. The Esprit has symmetrical cutaways and is slightly larger than the Flame. 

All three models were made in Japan, at a time when CBS was divesting Fender and selling it into private ownership. Despite the MIJ origin, these were very high end instruments and pretty unique for Fender, more Gibson-esque but with a lot of unique features of their own. The Esprit was a favorite of Robben Ford, and the model was re-designated as the "Robben Ford Signature". 

This is I think, the earliest example of TBX from Fender in a production model. 

The pots were made in Japan by Alpha, very good quality. They differ a bit electrically from the current CTS pots used by Fender. The 1M lower pot is not connected at terminal 3 at all. Also, from CCW-to-Detent, the 1M pot is purely conductive, meaning, 0-ohms. It only becomes resistive after the detent is passed going CW direction. The 250k upper pot tracks from full CCW to detent at 250k, so at the detent, it is at the full 250k resistance. Only after passing the detent, does it go to "no-load". 

The other circuit difference is the factory-original 4.7M-ohm resistor across the .022uf capacitor. 

Other note of interest is the 1M volume pots, where 500k would be more conventional with humbuckers. 

I find this TBX+1M works well together with the Schaller humbuckers, (8k neck, 12k bridge). It's perhaps, a little dark at the detent, but good and in single coil-mode, just about perfect. Turning the tone down has the same effect as typical tone control, but turning it up past the detent is more subtle and it has a "more everything" kind of sound and the presence (upper treble) comes up more towards the end of rotation. All of the circuit operation above the detent is due to resistive loading of the pickups, since the capacitor is effectively out of circuit and thus, no reactive element. 

Being a young man in 1984, I lusted after these guitars back then. Fender didn't make many, and they're hard to find today, especially in original form, people absolutely modified them heavily back in the day. My buddy had one that I played all the time, and he had some idiot route it for a Kahler tremolo and messed it up totally.

I also have a USA Masterbuilt Fender Custom Shop Robben Ford Ultra. Spruce top, Seymour Duncan 59 and JB pickups, conventional tone controls with 10pf bleed on the volume. It's gorgeous, and awesome, but they're very different guitars than the original Esprit. I like the TBX in the Esprit.