Brian May Series/parallel switching SSS (3x 2P2T + 1x 4P2T)

[newey]

Oct 15, 2021 1:55:40 GMT -5 dylanhunt said:

Or is it more relevant that the ground we're talking about is a "ground" in a guitar circuit?  Or does it matter which concept is accurate: a total instantaneous, completed circuit where the circuit that has the least resistance is the only/most important one, versus envisioning movement along a path that occurs over time rather than instantaneously?  My intuitions are not what I would call informed on any of these topics.  My meager understanding lets me imagine that the current is flipping back and forth so many times each second and that the decision of which wire is the high potential ("hot") and which is the low potential wire becomes an arbitrary decision (on a RWRP pickup, it only matters that the hot wire is opposite relative to the other pickup(s), and it doesn't matter what we call the matching lead on the RWRP pickup as long as it is the opposite of the matching lead on the other pickups).  I realize this isn't an engineering class, so no one should feel any obligation to fix my misunderstandings.  These are just questions I am wrestling with.

All of the above, AC versus DC, inductors, RWRP is irrelevant. In any electrical situation, we must have a completed circuit. In a guitar, the circuit is completed inside the amplifier, so we are only concerned that the "hot" signal ends up at the tip of the jack and the "not hot" (in parentheses since, as you point out, this is AC) ends up at the jack sleeve.

The problem that JohnH has identified is what we call a "short circuit", Mark's diagram is pretty stylized, it doesn't show the jack and doesn't show all the connections to the pots, that may be why you are having trouble "seeing" it. Mark said his guitar works as it should, so I suspect his diagram doesn't reflect the actual wiring (or, alternatively, he hasn't checked all the positions/switch combinations).

[dylanhunt]

Since his schematic is essentially the same as mine, here's how I would complete it for this single state of the system at a single instant.  I left out the encircled T's he had in the original (what are those supposed to represent?). This is my best guess as to what the objection is (with additional points to fill in the details):

1) There is no circuit without a signal generator
2) The high potential of the signal from the inductors is carried along the wires marked in green
3) Contrary to mark3796's report, it is impossible for the signal to be carried through the pots and to the tip of the hotput jack
4) The explanation for (3) is that the signal is carried along the path of least/no resistance marked by the five-pointed stars.

[newey]

Oct 17, 2021 19:32:30 GMT -5 dylanhunt said:

I left out the encircled T's he had in the original (what are those supposed to represent?)

Those are the Neck and middle pickup tone pots (highly stylized, with the connections assumed.)

1) There is no circuit without a signal generator

Well, there is certainly no guitar signal without one. The electrical circuit that is powering the LED desk lamp, the light from which I am using to type this, doesn't have a signal generator, but it is most definitely a circuit.

2) The high potential of the signal from the inductors is carried along the wires marked in green

By which it reaches the Volume pot hot lug, at which point it is short-circuited to ground via the purple wires. You're over-thinking this, it's just a short.

3) Contrary to mark3796's report, it is impossible for the signal to be carried through the pots and to the tip of the hotput jack

Yes.

4) The explanation for (3) is that the signal is carried along the path of least/no resistance marked by the five-pointed stars.

With the added fact that the path of least resistance is then connected to ground. If the purple wire were connected instead to the jack tip, the signal would make a complete circuit via the cable and amp. The signal would still be taking the path of least resistance; we'd just be bypassing the volume pot. The path isn't the problem, it's that the path is shorted to ground.

[dylanhunt]

Oct 17, 2021 21:17:22 GMT -5 newey said:

Oct 17, 2021 19:32:30 GMT -5 dylanhunt said:

I left out the encircled T's he had in the original (what are those supposed to represent?)

Those are the Neck and middle pickup tone pots (highly stylized, with the connections assumed.)
...
The path isn't the problem, it's that the path is shorted to ground.

Hmm, I don't think I've ever seen tone pots connected that way.  Maybe they are doing something that is making the behind-the-scenes connection you alluded to, newey--somehow avoiding the short.

Sounds like a path that shorts to ground is a problem.  If you don't examine the path, you won't know how to fix it.  Thanks for helping me spell out JohnH's and your thoughts, newey.  I can see it now.  Hoping markxyza comes back and offers a better representation of what his working version actually looks like. 

[newey]

Oct 17, 2021 22:25:52 GMT -5 dylanhunt said:

Hmm, I don't think I've ever seen tone pots connected that way.  Maybe they are doing something that is making the behind-the-scenes connection you alluded to, newey--somehow avoiding the short.

Again, he's not showing the connections, just that there are 2 tone pots, one for the middle and one for the neck pickup, as per usual on a Strat-type guitar. Each one is wired "across" its respective pickup, before the phase switch.

No matter how one wires those tone pots, the output is still shorted to ground.

[reTrEaD]

Oct 17, 2021 22:25:52 GMT -5 dylanhunt said:

Hoping markxyza comes back and offers a better representation of what his working version actually looks like. 

Yeah, that would be good if it happens.

I can say one thing for certain ... the DPDT on-on-on configuration associated with each coil will work in the series mode but is destined to fail in parallel mode. The problem here is that in series mode, it's necessary to replace any unused coil in the chain with a connection that completes the circuit. In parallel mode, any unused coil must be replaced with an open circuit. I can't see any way of being able to make that change, using DPDT on-on-on switches for the local switching for each coil.

Sept 13, 2021 12:44:07 GMT -5 dylanhunt said:

I did wonder at some point why the son of an engineer who could and did build his family's television wouldn't have simply used one 2P2T switch with a center-off position for each pickup instead of two on/off 2P2T switches.  I concluded that the component simply must not have been available to Brian May's father in 1960's Britain.  Now, that seems silly: If the components of a television and other household appliances were available, a center-off DPDT probably would be too.

A DPDT with a center-off position wouldn't do the job necessary for the SERIES circuit configuration Brian wanted. However the double-pole, three-position slide switch that Jaga mentioned (the Fender Mustang slide switch) was available at that time. I can see a way to use those for each of the pickup switches that would work is a most wonderful way in a series configuration.

However ...

Even if Mr Harold May was aware of those switches and had the vision to see how they could be employed, Brian may have found them to be undesirable. Two-position switches are extremely easy to operate. You slap them to one end or the other. Three-position slide switches are a bit fiddly to place in the center position.

[dylanhunt]

After being inspired by John Hewitt's series blender...
guitarnuts2.proboards.com/thread/7117/strat-ssm3-series-parallel-switch

...and mainly trying and failing to translate Fender 5-way switch wiring into independent pickup switching--decided it was a topology problem that I was ill-equipped to tackle (anyone know if there's an already assembled table for moving from individual pickup switches to 5-way switching and back again?)...

...I found some new material that relates to this thread (and other BM+ ambitious wiring threads on GN2):

Thomas Fredrick Wnorowski's patent for a "Method for switching electric guitar pickups" uses six DPDT switches (3 are on-off-on for pickups), with no 4P2T switch.
patents.google.com/patent/US6998529B2/en?oq=6998529

[My reason for posting this is to demonstrate that an arrangement like Brian May's and related to part of the concept that this thread (and others at GN2) is devoted to has been developed.  In pointing out that one particular version is patented, I'm taking it as obvious that making, using, selling, importing, etc. of this particular design could be patent infringement and, therefore, you should NOT do it.]  

On top of the usual parallel only and the series only settings, as well as normal (in phase) and reverse (oop) polarity, Wnorowski's design gives you N+(MxB), M+(NxB), and B+(NxM).


"SW-4 is used to electrically connect the bridge pickup and the middle pickup in series.  SW-5 is used to electrically connect the bridge pickup and the neck pickup in Series.  SW-6 is used to electrically connect the middle pickup and the neck pickup in series. Both SW-4 and SW-6 together put all three pickups in series (SW-5 has no effect in this circumstance)."

Here's a product (Awesome Guitars) based on the patent and it looks like their order page might still work:
www.awesome-guitars.com/1-products-010.php

and some photos:
www.awesome-guitars.com/0-photos.php

There's a discussion of this patent and similar designs in Donald Baker's paper, "On the Topologies of Guitar Pickup Circuits".
www.researchgate.net/profile/Donald-Baker-5/publication/323390784_On_the_Topologies_of_Guitar_Pickup_Circuits/links/5a9330030f7e9ba4296f4c05/On-the-Topologies-of-Guitar-Pickup-Circuits.pdf

[dylanhunt]

The new project is to mod a Squier VM Jaguar.  I'm scrapping the bass cut and using the three treble-side slots for independent pickup switches.  Was thinking of replacing the rhythm circuit with JohnH's serial blender, which would make good use of the tone roller, and the on/off switch.  I'm looking at what happens in the initial stage of switching (SW1, SW2, and SW3).  I don't want three additional switches for serial/parallel switching.  I want just one.  Now, I'm wondering if Wnorowski's design could be modified to exclude one of his switches: What his SW4 and SW6 do could be done by a single 4PDT switch (NxM, off, MxB).  I'd use the tone and volume rollers for when any serial switches are on, and use the master tone pot otherwise.


[UPDATED 10/21/2021]

[Yogi B]

Oct 20, 2021 23:28:30 GMT -5 dylanhunt said:

Now, I'm wondering if Wnorowski's design could be modified to exclude one of his switches...

Kind of, you can always permanently wire the remaining switches as if it was set a certain way, as you do in your modified diagram (wired as though Sw5 was set to parallel). However, this does not come without cost, you'd lose the ability to select the bridge and neck pickups together as a series group, namely the following six combinations:

• B × N, and the equivalent −B × −N
• B × −N, and the equivalent −B × N
• (B × N) + M, and the equivalent (−B × −N) + −M
• (−B × −N) + M, and the equivalent (B × N) + −M
• (N × −B) + M, and the equivalent (−N × B) + −M
• (−N × B) + M, and the equivalent (N × −B) + −M

Alternatively, if one were to wire the scheme as though Sw5 were in the series position you then would lose the ability to select the bridge OR neck as part of any parallel combination (including, selecting those pickups individually, i.e. in parallel with nothing else) resulting in many more 'missing' settings, twenty of them to be exact — leaving only 15 remaining.

[dylanhunt]

Great!  Thanks so much for taking a look at it, Yogi B.  I am already indebted to you for your work on Andrew Roberts' design, which I implemented and posted about on this thread.  Donald Baker's paper comments on the fact that Wnorowski (and other inventors of similar wiring schemes) do not always document all the useful possibilities.  You've pointed out some that were left out of the patent application.

I realize that I could probably swap out one or both of the pots on the Jaguar's rhythm circuit to make room for a total of 6 switches and still not have to drill any holes or add a push-pull, but I'm confident I would not miss the N x B series combinations or their added parallel variations.  More fundamentally, I decided that three pickups in series combination makes for a sound that is too wooly for my tastes, which is what motivated this strategy (to eliminate one of the three "connection" switches).

Here are the actual positions available with Wnoroswki's design if you do eliminate SW-5 (represented using his numbering scheme from his Figure 7) and the associated settings that come with it ("x" = series; "+" = parallel; "-" = reverse phase/polarity):

1)   Bridge (B)
2)   Middle (M)
3)   Neck (N)
4)   M + B
5)   N + B
6)   N + M
7)   N + M + B
8)   M + (-B)
9)   N + (-B)
10) N + (-M)
11) N + M + (-B)
12) N + (-M) + B
13) (-N) + M + B
14) M x B
15) n/a
16) N x M
17) n/a
18) M x (-B)
19) n/a
20) N x (-M)
21) n/a
22) n/a
23) n/a
24) N + (M x B)
25) n/a
26) (N x M) + B
27) (-N) + M x B
28) n/a
29) (N x M) + (-B)

Taking a page out of your book, Yogi B, it looks like there would also be:
N + (M x (-B))
N + ((-M) x B)
(N x (-M)) + B
((-N) x M) + B

So, 25 useful tones.

[dylanhunt]

I believe this schematic should work:


[Updated 10/23/2021 (Thanks JohnH for the push!)]

[newey]

dylanhunt-

It's unclear from the link you provided whether the Wnorowski patent is still active or not, but if it is, you do realize you could well end up owing him royalties for using his design?

Although that would only be true if his patent is legally valid (although it might take a lengthy and expensive lawsuit to establish that it is invalid). But I would not want to be the lawyer relying on that patent; series/parallel switching and OOP wiring were all done years before and are trivial, not novel, concepts. This not that much different than Brian May's father's wiring, circa 1963. I'm no expert on patent law, but I doubt this thing has a leg to stand on in court.

[Jaga]

Oct 21, 2021 17:49:24 GMT -5 newey said:

for using his design

What about personal use?

[dylanhunt]

Thanks for the heads-up, newey, and for the prodding to be a little less reckless about posting possibly patented designs (and posting designs I might want to patent myself).  I was actually wondering whether my change merited a big enough difference to warrant filing an application myself.   In this case, my innovation (use of a 4PDT) alone would make it less likely that anyone could mount a case of patent infringement here.  From the first of three claims in that patent application: "A guitar pickup Switching System for a three-pickup guitar, comprising. . . a switching means consisting of three double-pole, double-throw (center off) switches and three double pole, double throw (no center off) switches for selecting outputs of said pickups to provide all of the following combinations".  It is probably that sort of specificity in his application that made it even remotely possible for him to get the patent.  His other two claims are based on the first, and I'm proposing something with features different from the distinguishing features of his (only?) independent claim.  I'm certain that if I tried to make money off the design, I would have good reason to look over my shoulder.  But the spirit of your point is well-taken: it's not necessarily the outcome of litigation that anyone worries about, so much as the threat and the costs of litigation.

I think this gives the status of Wnorowski's patent (from the patents.google site): "2018-04-03 FP Expired due to failure to pay maintenance fee; Effective date: 2018 02 14"

But you raise an interesting general point, especially since this website seems to be devoted to inventions, some of which are potentially patented: Does "use" of a patented method, in the case of wires and switches work the same way as use of a patented method for, say, a pharmaceutical compound and are there any sort of (or more) exemptions/privileges that would apply in the former (especially to hobbyists) compared to the latter.  After all, the primary justification of patent laws in various countries and internationally is to protect commercial capital invested in using proprietary methods.  Apparently there is an "experimental use privilege" (in the USA) for research and development, which applies to uses of patents (by whom?) for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry, but I'm no lawyer and I don't know that this protection applies to hobbyists.  If Disney's IP litigation practices are anything to go on, it wouldn't matter that Brian May is Brian May: If someone in the right jurisdiction successfully applied for a patent that described his guitar, Brian May would be liable.

I think some part of the Wnorowski patent is truly original.  I did think of, in a very general way, the seeding strategy for the serial side that Wnorowski used (as I'm sure others have), but I stopped pursuing it when the earlier switching stage seemed to be flawed.  I failed to exhaust the possibilities for wiring an on/off/on DPDT for independent pickup switching.  Wnorowski's originality at the initial stage of pickup switching is obvious when you consider how many people on this website have suggested a successful method of using a single DPDT for the normal/off/reverse phase switching for each pickup, all together, in combination with serial/parallel switching--a method that would not block serial switching down the signal chain, and also would not ground out in a parallel circuit context.  The answer is none.

[newey]

Oct 21, 2021 19:19:25 GMT -5 dylanhunt said:

It is probably that sort of specificity in his application that made it even remotely possible for him to get the patent.  His other two claims are based on the first, and I'm proposing something with features different from the distinguishing features of his (only?) independent claim.  I'm certain that if I tried to make money off the design, I would have good reason to look over my shoulder.  But the spirit of your point is well-taken: it's not necessarily the outcome of litigation that anyone worries about, so much as the threat and the costs of litigation.

I was only about 33.33% serious on that . If you don't profit from it, the patent holder has no damages for which he or she could sue. Worse case, you get a "cease and desist" letter.

You are laboring under a very common misconception as to the manner in which the Patent Office grants a patent application. No one at the Patent and Trademark Office (at least here in the USA) examines each patent to see if the idea is novel or patentable. The Patent examiner reviews the application to see that all of the applicable requirements of an application are met, the claims are properly cited and so forth. Patents which are obviously not novel (you attempt to patent "The automobile" or "the electric guitar", for example) would not pass muster with the examiner, but no one there is going to get into the intricacies of guitar wiring to see whether a particular scheme has been done before. That is all left up to the courts.

Getting a patent doesn't get one anything but a right to sue usurpers for infringement, with the patent holder enjoying a legal presumption of originality. But the alleged usurper can defend if they can overcome that presumption, with evidence that the idea is not new or not sufficiently distinctive from the work of others to be patentable.

Also, BTW, it looks like the fee was paid, since the last line indicates that the patent was reinstated through some date in 2023.

[JohnH]

That patented wiring has some very quirky settings:



The phase switches for each pickup are on-off-on, and so the switching-off of each pickup is done by setting the respective phase switch to the middle setting, so total disconnection.

The other switches each either send a pickup to the output in parallel, or make a series connection of it to another pickup.

Operating in parallel is OK. But in series, if you want to change which pickups ae involved, you have to operate not only its on-off/phase switch but also one of the series switches. eg, in a series setting of all three, you switch one off. It breaks the series chain and all sound is dead (except possibly for a slight open-circuit hum) until you also switch off its series switch.

That, plus the fiddle of setting three position switches, and the quirky logic, would drive me Nutz!

Also, i can't see the patent being enforceable, when there were earlier, better, simpler versions of using dpdt toggles to get the claimed settings, particularly this SuperStrat one by Dan Armstrong (add phase switches as wanted) which I think was published in the 1990's:

[Yogi B]

Oct 22, 2021 14:42:44 GMT -5 JohnH said:

That, plus the fiddle of setting three position switches, and the quirky logic, would drive me Nutz!

It's also vastly inefficient from a standpoint of the ratio of unique tones to switch positions. The complement of three 3-position switches plus three 2-position switches gives a total of 216 possible switch states, yet there are only 36 unique selections (including those present in the wiring but not claimed by the patent, plus also counting 'off').

Operating in parallel is OK. But in series, if you want to change which pickups are involved, you have to operate not only its on-off/phase switch but also one of the series switches. eg, in a series setting of all three, you switch one off. It breaks the series chain and all sound is dead (except possibly for a slight open-circuit hum) until you also switch off its series switch.

That pickup's not "off", it's just "in series with the air"!

[dylanhunt]

Here's a link to more on the Dan Armstrong wiring: ashbass.com/AshBassGuitar/

Thanks for the Dan Armstrong connection, John H, and to you and Yogi for the food for thought.  I think it partly comes down to whether the beholder is someone who prefers a 5-way or similar switch (Swiss Army blade switch) rather than being compulsively focused on independent switches for each pickup (especially where reverse phase is of interest).  For someone of the latter bent of mind, adding three or two more switches is fairly minor.  If I can figure out the external wiring to accommodate the internal wiring of a standard 4PDT or if I can find a 4PDT on-on-on switch that will accommodate my little mind (center position: up/up/down/down), then you're only talking about a single additional switch.  Of course, with my solution, you lose any NxB combo, which means several Wnorowski combinations, as well as two combinations Yogi identified that are missing from Wnoroski's simple switch figure: ((−N × B) + M) and (N × −B) + M.  However, on the positive side of the ledger, you get 25 useful combinations in total.

I think the preference might depend on whether the main thing in the luthier-guitarist's mind is the interface outcome (which pickups, normal vs. reverse phase, series vs. parallel) or the engineering logic (switch positions/states or wiring logic).  I can see both perspectives, but am INTERFACE dominant (form informs function) when designing, rather than ENGINEERING dominant (form follows function).  The idea that there's a "dead spot" simply because there's no sound is an idea that looks wrongheaded to someone who is focused on the interface.  This came up with Andrew Roberts' design too.  If you are looking for a setting that includes a middle pickup and you get a dead spot when the middle pickup switch is off, then it's not a dead spot: It's a mental disconnect in the guitarist, who simply needs to turn the switch on.  There's also a third perspective, which usually is on display when someone says, "That's not how it works," but that's not the perspective of a designer.  It's the vantage of a developer who has already finished designing and debugging and is now receiving complaints.

Newey, I resent your remark: I'm too lazy to labor under a misconception (about patents or anything else).  I just muddle right on through. 

[newey]

Oct 23, 2021 12:26:02 GMT -5 dylanhunt said:

However, on the positive side of the ledger, you get 25 useful combinations in total.

Well, "useful" is the operative word there. A number of those 25 may not sound distinctively different. For example, I can't hear any difference between (N X M) + B and N + (M X B), in a guitar with 3 identical pickups.

As for the issue of dead spots, with time one gets used to the intricacies of pretty much any switching, but where this breaks down is if the guitar is used for playing live, where switching "on the fly" is a consideration. Yes, you may know consciously how to avoid the dead spots, but most gigging guitarists don't want there to be any chance of losing sound in mid-song if they quickly hit a toggle switch.

[dylanhunt]

Oct 23, 2021 12:39:47 GMT -5 newey said:

Oct 23, 2021 12:26:02 GMT -5 dylanhunt said:

However, on the positive side of the ledger, you get 25 useful combinations in total.

...I can't hear any difference between (N X M) + B and N + (M X B), in a guitar with 3 identical pickups...

Very good points (I'm not a gigger, so that explains a lot)!  I'm curious--because I haven't wired or even heard {(A x B) + C} or {(A + B) x C}--can you, or any other above-average listener you know, tell the difference between two pickups in series with a third in parallel from two pickups in parallel with a third in series?

[newey]

That, I do hear a difference, but it's pretty subtle. Of course, if the pickups are dissimilar, then these may vary more one from another.

[aroberts8089]

Sept 15, 2021 12:19:34 GMT -5 dylanhunt said:

I've been a little worried about not seeing any real-world testing of the design.  Looks like AR's brother used AR's build, but I don't think AR reported back on whether there were any problems with the switching related to the imperfections Yogi B found.  That had to do with AR's hybrid/Easter-egg position on the series/parallel switch ({BxM}+N).  My impression is that nobody has been able to eliminate that issue without going higher than a 4PDT,

Ah good to see this old nugget is still being chewed over. I will mention that my brother did wire up my design and it did indeed have the dead spot - which I had actually confirmed with some detailed flow tracing but interesting to see it in use.

That one spot has been a thorn in my side for sure as it's quite disappointing and I have been trying to solve the issue in several ways. As far as I can tell you would have to go to 2 switches to make it work. I do want a N x M + B option to simulate a broadcaster with a Humbucker setting.

I was also contemplating Hermetico's design where the phase switching is separate and the series/para switching is the pup switch, but then one is back to 5 switches.

I might go down the road of using a 4 pole 5 position switch, but then feel like I have a bunch of positions I won't find useful. But the hoarder in me can't let it go down to a 4-pole 3-position switch either. I think I might prefer a rotary switch as well. Not sure if I can get the 3 Phase switches with a standard blade switch into a Tele plate.