Interlocking switches: how to represent them in schematics?

[stevewf]

I ran across "interlock pushbutton switches" like these on a worldwide retailer web site:


As I understand, they're a set; when you press one button down, it locks in the Down position while the other(s) go into the Up position. Only one button remains in the Down position at any time (barring any notch-style hacking, of course).

Does there already exist a standard pictograph to be used in schematics for such a set of switches? Does anyone have a proposal on an easily understandable schematic pictograph (... icon... thing)?

-Steve

[JohnH]

Ive never seen such a symbolic representation to show how pushing one in, makes the others pop out mechanically. It's not an unusual arrangement though, on every car radio at least until the late 70's!

Unless one of our highly respected EE's knows better, I'd suggest you just draw all the switches, with line-work or referencing (eg sw1a sw1b etc) to show how the contact pairs within each dpdt switch work ganged together, and then add a note or diagram to show the extra push-in / pop-out action between switches.

[stevewf]

May 18, 2024 23:35:50 GMT -5 JohnH said:

It's not an unusual arrangement though, on every car radio at least until the late 70's!

While I was dreaming of ways to use the above type of switch, I also reminisced about those old radio buttons. When you pressed one button in, you could feel that there were mechanical gizmos making the other buttons slide out. Then I was dreaming about a guitar with those very same buttons as controls. Like, pulled from a '72 Pinto. Instant nuvo-retro!

[JohnH]

Actually Im remembering the radio on my '77 Fiat. When you pressed a channel button, somehow it span a wheel and moved the pulley that shifted the dial from one station that Id set, to another. And made the other buttons pop out.

[stevewf]

Yup. And you could hear the stations that the tuner ran through on the way between presets.

[stevewf]

So how about something like this, using the OP photo as an example of a 4-station DPDT interlocking pushbutton switch:

...and then somehow showing them all ganged together.

The number of physical stations determines both a) the number of columns in the set above and b) the number of virtual positions for each pole - in this example four;
The number of poles in each physical station determines the number of rows in each column  - here, two rows since each station is a 2-pole switch; there could be differing numbers of rows in each station;
The number of throws determines whether the green intra-connected pins exist or not - here they do exist, since there are 2 throws for each pole.

[Edit: I should go on to say that within the above framework, pressing one of the stations puts all of switches into that station's virtual position - for example, pressing station #3 throws all the switches into position #3.

Also, of course, schematic diagrams using this type of switch do not have to retain the row/column layout; the switches do need to remain ganged/synched, though.]

It can get very busy. I've seen 10-station 6P2T interlocking pushbutton switches online. Represented in the above style, there'd be a 10 x 6 grid of poles, with 10 positions, and intra-connected little green wires on nine of those positions for each of the 60 poles.

[stevewf]

Or another proposal, basically the same as above, but more compact by consolidation of some of the throws:

[JohnH]

Seems logical. Looking at the graphic in the last post, I get it and can see what it means, more so than the previous one which TBH was not sinking in for me.

Also, the notation and graphic is looking no more complex than that needed for a group of seperate switches, once the 1,2,3,4 annotation is followed.

What does sumgai think? I'd expect he'd probably have had items with such mechanically-interlocked button groups on the bench in years gone by. Is there an accepted way of drawing them on a schematic?

[newey]

I seem to remember there was a '60's Japanese guitar, maybe a Teisco (?) that used a similar set of switches for the pickup switching. Don't recall exactly how it worked but I recall seeing something like that.

The second set of schematic symbols seems to convey the idea well. The type of switches are, I believe, called "latching switches" but I'm not sure how a latching switch is desgnated as such on a schematic.

[stevewf]

Far as I've figured out, a latching switch is one that's not momentary. Even Les Paul style toggle switches would therefore fall under the category of latching switches. This is not the same as a locking switch, which can be latching (or not) but requires a special action to allow it to move from at least one of its positions (like turn key before throwing, or lift before throwing, or twist before throwing).

I think these are latching switches. The thing that distinguishes them from a single push-push is that that latching of one switch depends on action performed on another switch (that's in the interlocking set).

There's at least one other case to be drawn: an interlocking set that allows multiple stations to be "down" simultaneously. This sometimes (always?) includes a "clear" button along with the rest of the set.

[Yogi B]

May 19, 2024 20:16:07 GMT -5 stevewf said:

There's at least one other case to be drawn: an interlocking set that allows multiple stations to be "down" simultaneously.

In those situations, when reading a schematic all we care about is which position a specific DPDT is in, so numbering each possible state is pretty much irrelevant and drawing them as separate DPDTs makes most sense. (That's what I did in a schematic for a Teisco Spectrum 5, which may be the model of guitar newey is remembering.) Since the number of states grows exponentially with the number of switches that can be depressed (and latched) simultaneously, the number of "virtual" positions required by your first representation method basically precludes its use for all but the simplest cases.

This sometimes (always?) includes a "clear" button along with the rest of the set.

It can also be possible that partially depressing a switch can be enough to release the others, but without becoming latched itself (thereby giving the same all-off state as a clear button).

---

The IEEE-315 / ANSI-Y32.2 standard (or at least the version I can find online), doesn't have much to say. Section 14.1 deals with mechanical connections: item 14.1.1 specifying the usual dashed lines; item 14.1.2 indicating a caret placed along such a line is a fulcrum (which I can see potentially being used for an interlocked pair of switches, but that would breakdown for three or more); item 14.1.3, "Mechanical interlock, other" simply reads "INDICATE BY A NOTE".

[stevewf]

Thanks for help and feedback. I'm considering using a set of interlocking pushbuttons for a lab strat that for checking out 3-coil hum canceling combos. There's been some discussion on this forum, much by JohnH, so I'm gratified to see him in this thread.

Interlocked multi-pole pushbuttons seem like a compact way to throw poles at a problem. They also seem to lend themselves to my idea for the above-mentioned lab strat.

I've got the pickups for a four-coil SHS layout (middle pickup is a dual rail HB). I'd like to avail combos of single, double and triple coils, with the double and triple combos being hum-canceling. These radio buttons seem like an intuitive way to be able to control that.

Well, for controlling part of it. Since hum canceling trios can be in Ax(B+D) and A+(BxD) form, I want to be able to switch between those. With a single switch, of course.

I'm also exploring out-of-phase trios in the form of Ax(B+-C) and A+(Bx-C). Swap in an OoP counterpart with a single switch, yes?

And the four-coil combos (AxC)+(BxD) and (A+C)x(B+D). Add-a-coil, up to four. All of that is why I have some muti-station 4PDT switches on the way.

I've got some starter schematics drawn up, but I was stumbling on how to draw the interlock switches. So, thanks, people. Hopefully, I'll be able to report about the guitar in a new thread when the time comes.

[kitwn]

May 19, 2024 17:53:40 GMT -5 newey said:

I seem to remember there was a '60's Japanese guitar, maybe a Teisco (?) that used a similar set of switches for the pickup switching. Don't recall exactly how it worked but I recall seeing something like that.

The second set of schematic symbols seems to convey the idea well. The type of switches are, I believe, called "latching switches" but I'm not sure how a latching switch is desgnated as such on a schematic.

I have a fascinating book called Legendary Guitars, An Illustrated Guide by Tony Bacon. It's crammed full of colour photographs of every electric guitar you've never heard of from the 50's and 60's. Ridiculously low price too.

There is a 1966 Teisco Del Ray Spectrum 5 in there with huge, rainbow coloured switches but I'm not sure they latch. There are also several Italian designs that have rows of switches that look more like they might be latching.

Such designs obviously didn't catch on. Mechanically unreliable? Too complex and expensive to make? Too "Star Trek' in appearance for the discerning musician? Too hard to find the right button in the middle of a gig compared to the ever-popular 5-way we know and love? All of the above?

Kit

[stevewf]

Some potential problems that have me concerned, while I await the eBay ordered units' arrival:

• Mechanical noise when they're unlatched and pop back into "up" position. If you're sustaining the strings, it might be noticeable. Even if damping, some vibration might get passed into the pickguard and onto the pickups. I remind myself: any resulting noise is not necessarily a bad thing.
• They require a straight line; the more stations, the longer that line. No stylish curves for you!
• They require a flat surface. Pretty much spells pickguard.
• Tiny lugs. Well, the ones I've ordered have PCB-mount pins on one face and solder lugs on the other. I've noticed many that present only PCB pins.
• Depth limit. I haven't found anything greater than 4P2T that'll fit in a Strat-depth cavity. And guaranteed that if they exist, they'll have closely-spaced PCB pins only.
• Mounting holes. For mounting, they have one hole in each end of the switch complex. The pair of extra screws on top of the pickguard doesn't appeal to me. So I'm looking into making a bracket that uses the other panel mounted controls, probably the pots, to hols it in place. By the photos, it looks like the units would be better-positioned a little deeper under the pickguard anyway.
• Buttons. I've had difficulty even building search terms to look for them.

[sumgai]

May 19, 2024 15:58:41 GMT -5 JohnH said:

What does sumgai think?

I think I need to see photos of the actual items that Steve receives, preferably from several angles.

Today's history lesson on these switches has been postponed to a later date TBD. Sorry 'bout that.





sumgai

[stevewf]

May 20, 2024 23:57:31 GMT -5 sumgai said:

May 19, 2024 15:58:41 GMT -5 JohnH said:

What does sumgai think?

I think I need to see photos of the actual items that Steve receives, preferably from several angles.

Today's history lesson on these switches has been postponed to a later date TBD. Sorry 'bout that.

sumgai

Thanks for standing by, sumgai. ETA 6 June (and then add some time for photos).

[col]

May 19, 2024 17:53:40 GMT -5 newey said:

I seem to remember there was a '60's Japanese guitar, maybe a Teisco (?) that used a similar set of switches for the pickup switching. Don't recall exactly how it worked but I recall seeing something like that.

The second set of schematic symbols seems to convey the idea well. The type of switches are, I believe, called "latching switches" but I'm not sure how a latching switch is desgnated as such on a schematic.

I believe they are referred to as 'radio buttons'. The phrase is even borrowed in CSS, for the forced selection of a single option from a list.

[newey]

I'm envisioning a 3 pickup Strat-ish guitar with one of these for the pickup selector. 5 buttons, giving the 5 standard Strat selections but instead of having to move a lever through 4 positions to get from neck to bridge, it's one button.

[stevewf]

May 21, 2024 16:52:19 GMT -5 col said:

I believe they are referred to as 'radio buttons'. The phrase is even borrowed in CSS, for the forced selection of a single option from a list.

Yes, and calling them radio buttons in a computer user interface even predates the notion of cascading style sheets!

[stevewf]

May 21, 2024 20:39:23 GMT -5 newey said:

I'm envisioning a 3 pickup Strat-ish guitar with one of these for the pickup selector. 5 buttons, giving the 5 standard Strat selections but instead of having to move a lever through 4 positions to get from neck to bridge, it's one button.

And what about removing all knobs and levers and offering only pushbutton controls. Like set volume levels (or even mixes) and a varitone sort of tone control.

And then on the player level, analogous to flipping the LP toggle back & forth rapidly, you could cycle through several settings also rapidly. Almost accordion-like button controls   

[stevewf]

May 21, 2024 20:39:23 GMT -5 newey said:

I'm envisioning a 3 pickup Strat-ish guitar with one of these for the pickup selector. 5 buttons, giving the 5 standard Strat selections but instead of having to move a lever through 4 positions to get from neck to bridge, it's one button.

Actually, sorta more seriously, I'm looking into something similar. Here goes. It's long, so I bolded text that seem to me to matter most, in the hopes of getting feedback.

It's a stratish guitar, but with a single-size dual-rail HB for the middle pickup (four wires plus shield). The fourth coil is added for getting lots of hum-cancelling combos. In fact, this guitar is about just that - hum canceling, including three-coil combos discussed in this forum (below, I stick to the coil-labeling from that thread). And then, to add fun, it should also do series/parallel swapping (a little bit of detail is offered below). And maybe even reach into phase-swapping, which should be interesting especially in the 3-coil combos... and perhaps a turn-off in 2- and 4-coil modes.

So the player's view of the controls would be like this:
- the three-way blade chooses which pickup is the main pickup - N M or B. In 3-coil setups, this would be the outer member.
- 4 radio buttons would choose whether the coil is to be solo, part of 2, 3 or 4-coil sets. Choosing 2 starts with the closest neighbor (having opposing magnet); 3 adds the other opposing magnet, and 4 adds the last coil.
- Another button for choosing ser/pal. Gonna need at least 6 poles here, I reckon, so that's why I've chosen another button.
- Master Vol. Master Tone
- Series blend in the last knob... and maybe parallel blend too, using a C-curve 2nd gang. Turns down one of the inner coils, specifically the one that's also used in 2-coil combos, to facilitate fading in 2-coil mode also)

Pickups: I got a set of Fleor pickups that show similar DCR among them, though the middle twins each measure slightly lower than the neck single and of course the bridge single is higher. (sorry, DCR is the only measurement that I feel confident is taking). The middle coil having a dip in "power" is ok with me; I've found it tends to offer stronger quack in the right combos, and I do want this guitar to be able to quack.

As for detail concerning series/parallel: The three-coil combos that are hum-canceling (or hum-reducing) each use both series and parallel su-combos in their layout, eg A x (B + D) and A + (B x D). For convention, I propose to use the outermost relationship for overall naming: A x (B + D) would be called a "serial" setup, since the outermost is serial, and I anticipate that the outermost will be the stronger factor in making the overall sound character.

In any case, because the three-coil combos have a bit of each (ser/pal), to swap one would require swapping the other. That means more than a DPDT for s/p switching. That's a secondary reason for looking into the interlocking switch sets - more poles! (the primary reason was for jockeying which coil goes in to which slot in the overall layout - outer member vs inner).

I found I was focusing strongly on three-coil combos and leaving out the four-coilers. Until I thought: if someone were to hand me the guitar (having 4 coils) and told that I get can use up to 3 at a time but not 4? I'd wanna open it up and heat up the soldering iron. So I also wanna include two of the four-coil hum-cancelling layouts. These ones: (A + C) x (B + D) and (A x C) + (B x D) which have like-magnet coils together in pairs (sharing parentheses). I know this excludes the usual HB setup of (A x B) + (C x D) which has opposite-magnet coils paired; it's a concession for simplifying the switching for three- and two-coil options. I'll post what I've come up with when I've got it digitally, and maybe a Nut can see past my problem... later.

And then, phase swapping. I'm wavering here on whether to include phase-swapping, as I've found some OoP strat combos tend to be not very good, like B + ~M etc. (neighboring strat coils in parallel, out of phase). Choking one of the coils via a cap in series helps, but adds complexity. There'll already be plenty of wires under the pickguard, and plenty of buttons. Academically, though, it seems like just a few poles could make it happen: swap ~C in where B was, and ~B in where C was, for example. And since we're filling the cavity with poles, what's a few more among dozens? 

In the tables below, the magnets assumed are: Neck=north, Middle=south and Middle=north, Bridge=south. That's in physical order on the pickguard. I write "Mn" and Ms" to differentiate the two coils of the middle pickup.

Parallel mode:

3-way | radio  >  VV   | buttons>	1 coil	2 coils	3 coils	4 coils
1	Nn	Nn + Ms	Nn + (Ms x Bs)	(Nn x Mn) + (Ms x Bs)
2	Ms	Ms + Mn	Ms + (Mn x Nn)	(Ms x Bs) + (Mn x Nn)
3	Bs	Bs + Mn	Bs + (Mn x Nn)	(Bs x Ms) + (Ms x Nn)

Serial mode:

3-way | radio  >  VV   | buttons>	1 coil	2 coils	3 coils	4 coils
1	Nn	Nn x Ms	Nn x (Ms + Bs)	(Nn + Mn) x (Ms + Bs)
2	Ms	Ms x Mn	Ms x (Mn + Nn)	(Ms + Bs) x (Mn + Nn)
3	Bs	Bs x Mn	Bs x (Mn + Nn)	(Bs + Ms) x (Ms + Nn)

The coil to have the fader on it is underlined.

I haven't even yet seen the set of buttons so this whole thing might go out the window in a puff of dreamy-eyed wind. Assuming otherwise, here are some
decisions that're still up in the air:
- How to and whether to include serial/parallel swapping in 4 coil mode; or, for that matter, how to avoid including it.
- Whether to invite the monster of phase swapping into the room. How to work it in 4-coil mode?
- Do I really want that parallel fader just for the 2-coil/parallel mode? Or, for multi-coil sounds, if I allow the parallel fader to the affect whole bracketed pair, wouldn't that let hum in, especially in the 4-coil sounds?

Waiting for the "interlocking 4P2T pushbutton" aka "radio button" switches to arrive; meanwhile, any further thoughts are welcome.

Edit: oh, yeah, I forgot: I plan on using a 3-position blade switch with 8 poles.

[kitwn]

May 21, 2024 20:39:23 GMT -5 newey said:

I'm envisioning a 3 pickup Strat-ish guitar with one of these for the pickup selector. 5 buttons, giving the 5 standard Strat selections but instead of having to move a lever through 4 positions to get from neck to bridge, it's one button.

Yes, but it has to be the right 'one button'. Our beloved 5-way is still easier to operate by feel if you're on stage with all the spots aimed at the singer and you're half blind whilst looking cool in your RayBans 🤣

Kit 😎

PS   The modern version of the radio buttons would be a row of capacitive touch sensors feeding a microcontroller operating solid state switches programmed to give whatever set of 4/5/6/etc. pickup combinations and phases you had previously programmed into it from an ap on your phone. The irony is that that would probably be cheaper to mass produce than a complex mechanical switch.

[newey]

May 23, 2024 1:20:38 GMT -5 kitwn said:

The modern version of the radio buttons would be a row of capacitive touch sensors feeding a microcontroller operating solid state switches programmed to give whatever set of 4/5/6/etc. pickup combinations and phases you had previously programmed into it from an ap on your phone.

We've had that discussion in the past, and it certainly can be done. I seem to recall Fender having a phone app to control one of their modelling amps, IIRC.

The objection has alwaqys been the same one as with adding any active electronics into one's guitar- the need for battery (or other power supply). Here, that issue is doubly compounded in that you could lose the 9V battery mid-gig, but would also have to keep your phone charged to control the thing.

At some point, this becomes "tech for tech's sake". Those of us who lived through the 1980's will remember how automakers, especially Chrysler, started installing all digital LED/LCD instrumentation and putting in voice chips once those became available. This was all cool and hi-tech stuff for the time, but it didn't really add anything useful. People quickly discovered that a quick glance at an analog dial was better than having to read digits, and a simple warning light was preferable to some computer voice telling you "The door is ajar" over and over again.

And, stevewf-

- 4 radio buttons would choose whether the coil is to be solo, part of 2, 3 or 4-coil sets. Choosing 2 starts with the closest neighbor (having opposing magnet); 3 adds the other opposing magnet, and 4 adds the last coil.

You said these were "4P2T" switches, but I'm only seeing six solder pins on each one, which would suggest a 2-pole switch to my eyes. Even if they were 4 poles, I'm not envisioning how you could make all those choices happen, particularly given that only one of the 4 can be "active" at a time. I think you should work out that switching for feasibility before going much further down this road.

[sumgai]

May 23, 2024 1:20:38 GMT -5 kitwn said:

The irony is that that would probably be cheaper to mass produce than a complex mechanical switch.

Arduino to the rescue!

There are other single chip systems out there, to be sure, but the Arduino community is probably bigger than Facebook. Like this site, help of any sort will not be lacking.

Long timers here will recall that both ChrisK and I have long advocated for a digital switching system, but the knock-out blow is the fact that any such setup will require a battery. And who wants to be Mojo-ing all over the stage, only to have the battery go Gefooey®, and all sound ceases to be audible.  The answer is, of course, to have a 10 year battery permanently installed into the chip in the first place, just like fire/smoke alarms that are now required just about everywhere you might go.

Or, as in my "glory" days of using a MIDI-driven synth-enabled axe, I did indeed have the luxury of being able to digitally switch between any number of possible combinations.  Therein, I had a remote power source using two of the 13 available wires in the cable.  That works, but in going to an RF cable, I had to give that up.  A Behringer FC-1010 foot controller took over those duties, but didn't look as cool, I admit.

Sigh.  Those were the days.





sumgai

[stevewf]

May 23, 2024 5:59:38 GMT -5 newey said:

And, stevewf-

- 4 radio buttons would choose whether the coil is to be solo, part of 2, 3 or 4-coil sets. Choosing 2 starts with the closest neighbor (having opposing magnet); 3 adds the other opposing magnet, and 4 adds the last coil.

You said these were "4P2T" switches, but I'm only seeing six solder pins on each one, which would suggest a 2-pole switch to my eyes. Even if they were 4 poles, I'm not envisioning how you could make all those choices happen, particularly given that only one of the 4 can be "active" at a time. I think you should work out that switching for feasibility before going much further down this road.

That photo was an example of a row of 2P2T switches, but I've ordered 4P2T switches. I stopped short of the 6 pole offerings because they were reportedly 56mm in length - too much for the cavity.

I''ve kinda worked out how to do it in principle, but after getting halfway through a schematic, I decided that there were just too many damn connections already (and I wasn't even done yet). If it's this much of a drag to design it, then actually building it will be a nightmare.
Here's the halfway-done schematic:

SPOILER: Click to show

So I've abandoned the idea of switching in 1,2,3 or 4 coils at will. I'll take a step back to the "up to 3 coils" and see if it seems worthwhile.

[kitwn]

May 23, 2024 5:59:38 GMT -5 newey said:

May 23, 2024 1:20:38 GMT -5 kitwn said:

The modern version of the radio buttons would be a row of capacitive touch sensors feeding a microcontroller operating solid state switches programmed to give whatever set of 4/5/6/etc. pickup combinations and phases you had previously programmed into it from an ap on your phone.

We've had that discussion in the past, and it certainly can be done. I seem to recall Fender having a phone app to control one of their modelling amps, IIRC.

The objection has alwaqys been the same one as with adding any active electronics into one's guitar- the need for battery (or other power supply). Here, that issue is doubly compounded in that you could lose the 9V battery mid-gig, but would also have to keep your phone charged to control the thing.

At some point, this becomes "tech for tech's sake". Those of us who lived through the 1980's will remember how automakers, especially Chrysler, started installing all digital LED/LCD instrumentation and putting in voice chips once those became available. This was all cool and hi-tech stuff for the time, but it didn't really add anything useful. People quickly discovered that a quick glance at an analog dial was better than having to read digits, and a simple warning light was preferable to some computer voice telling you "The door is ajar" over and over again.

And, stevewf-

- 4 radio buttons would choose whether the coil is to be solo, part of 2, 3 or 4-coil sets. Choosing 2 starts with the closest neighbor (having opposing magnet); 3 adds the other opposing magnet, and 4 adds the last coil.

You said these were "4P2T" switches, but I'm only seeing six solder pins on each one, which would suggest a 2-pole switch to my eyes. Even if they were 4 poles, I'm not envisioning how you could make all those choices happen, particularly given that only one of the 4 can be "active" at a time. I think you should work out that switching for feasibility before going much further down this road.

I'm well aware of the battery issue, it's a matter of personal choice and, if you go with it, proper management before the gig. Tommy Emmanuel copes OK, I think his Matons are all battery powered.

One of the problems with mechanically switching a complex set of options is the need to remember exactly how to set umpteen switches correctly in between songs on stage. Putting an Arduino in your guitar allows the complicated setups required for each song to be made in advance and easily recalled from the machine's memory by appropriately simple switching on what appears to be a conventional guitar. I have the cunning block diagram scribbled on a bit of paper somewhere, though I'm not sure I'll ever take time out from playing my existing guitars to actually try and build it.

Kit

[stevewf]

For those interested, here's a sketch of how I was thinking of doing Add-a-Coil buttons:


Sorry about the smudginess. There are three colors:
-Black ink that represents coil groupings (the squares and circles), and GND and OUT. They're the same for all 8 modes drawn.
-Purple highlighter that represents the circuit when in each mode (eg 2-coil parallel). They change depending on mode.
-Light pencil the shows the circuit that's missing (open) in each mode, but closed in other modes; maybe you  And there was a separate light pencil that did a serial fader, overarching the diagram in each mode. This is to help me envision the switching needed.

Those squares and circles don't represent any given coil; they represent a pair of sections of the 3-way switch. The pins seen on the squares and circles represent the poles of the sections; the throw lugs will go to two or three different coils, a hot and a cold for each coil. The squares and circles just gloss over which coils go in which position for this high-level diagram, so I could design the switching on a zoomed-out level, ignoring the precise location of each coil at this stage. Square versus circle means that a coil in that position should be of the right magnetic polarity for hum-canceling. 

Reminder: 3-coil hum-reducing combos have one of two formats: A x (B + C)   or  A + (B x C), with the two members inside the parentheses bing of a common magnetic polarity, which is opposite of the outer member's. Example: Neck-north x ( Middle-south + Bridge-south ).

My approach: using the black ink layout, designate the circle that's hanging on the output as the Main coil. It designates which coil will go directly to OUT. Since we don't like hanging coils on hot, it'll have to be the coil that's used all the modes for that 3-way position. Example, let's plan to put the Neck (north) coil onto the OUT when the 3-way is in position 3. So now when the 3-way is in position #3, all the modes (1- 2- 3- and 4-coil modes) will feature the neck coil attached to OUT.
-Starting with 1-coil, the neck coil will be solo (so, by the way, we'll have to make sure its cold goes to GND when in Pos#/1-coil).
-Then in 2-coil, we'll introduce one of the south coils (square) into the mix: Middle south. That's represented by the square that's tied to GND.
-For 3-coil, we'll add an inner member with our second coil - now we're at Nn & (Ms w/ Bs).
-And for 4-coil, the other circle gets added, becoming an inner peer with our "main" Neck, giving (Nn with Mn) and (Ms with Bs).

Remember, the whole thing is also subject to serial/parallel switching, which is why I used "with" and "and" above; I needed to generically mention the relationships, since they can swap independently of the controls mentioned so far. And on that note, when the ser/pal mode is swapped, then all the relationships being used get swapped (both inner and outer relationships flip to the opposite).

Designing became a bit of a mess when any 2 modes (like 1- and 2- coil modes) use the same network, but not the other two modes (eg 3- and 4-coil modes). Then, two poles from the pushbuttons become necessary for that one connection - an "or" with two switches in parallel or an "and" where they're in series).

This all spells many poles. Which also spells many wires and lots of soldering crammed into a Strat, and maybe it won't even sound good! That's why I'm stepping back to "3 coils maximum", as it was shaping up a lot simpler. That endeavor, I think will go in a separate thread when I get further. This thread was mostly about how to draw the pushbutton switches in a schematic, after all.

[Yogi B]

May 20, 2024 17:07:53 GMT -5 stevewf said:

Some potential problems that have me concerned, while I await the eBay ordered units' arrival:
...

• Buttons. I've had difficulty even building search terms to look for them.

Is that the correct link? Your mention of "buttons" (which I take to mean the switch caps), which the linked switches have and therefore makes me think not. And the following from the item description (with emphasis added), makes me hope not:

The white switches are interlocking: they lock when pressed and release when one of the other white switches is pressed. The blue switch is independent.

---

May 23, 2024 16:20:04 GMT -5 stevewf said:

I've kinda worked out how to do it in principle, but after getting halfway through a schematic, I decided that there were just too many damn connections already (and I wasn't even done yet). If it's this much of a drag to design it, then actually building it will be a nightmare.

To me it looks like you're over thinking this. You have enough switches/poles to keep things modular, and restrict special interlinking between switch states to a minimum.

For example: you have an 8P3T switch and 8 coil terminations, so you can use that switch to independently swap around the coils — enabling you to design the rest of the switching for just one of the 3-way's positions.

The series/parallel switch is just three 'regular' (DPDT) series/parallel switches stacked together (with the function of the centre one the reverse of the other two), then all that's left for the radio buttons to do is to make the appropriate connections to that series/parallel switch for use with one, two, three, or four coils. The below diagram should do just that.

Note that there's only one place (Sw3-Sb) that the disengaged throw of the radio button switches comes into play. This is only to prevent shunting the (unused) 'second' coil (e.g. Mn, with the 3-way in position "1") in single coil series mode, so could be omitted (instead tying node "G" permanently to ground). That way the radio buttons would need only be 4PST switches (if such things exist), saving on unnecessary terminals & cavity depth required.

[stevewf]

Gee whiz, Yogi B!

OK, let me remove the question of which switch(es) are ordered, and why they need caps. I've actually ordered a few different types. One of them has appeared in this thread (only via link). The photo that appears at the top of this thread is not what I've ordered; it' just one I hastily grabbed for the original purpose of this thread: to discuss in principle how to diagram them.
What I have ordered:
- 4-station 4P2T, 3 interlocking, one independent (and hopefully self-latching). Includes 4 caps.
- 3-station 4P2T all interlocking. No caps included.
- 4-station 4P2T all interlocking. No caps included.
- 5-station 4P2T all interlocking. No caps included.
- 6-station 4P2T all interlocking. No caps included.

The ones with 5 & 6 stations were ordered on the off-chance that I could omit one or more station (remove it from the rig, and not make a hole in the pickguard for it) and replace one station so that it's not in the interlocking set, so that the unit could be used for both Add-a-Coil and ser/pal switching while keeping them visibly separated. And of the off-chance that someday I'll find use for them regardless. I mean, the units are cheaper than the shipping, so why not pile them on in the order?

Other options I've seen, but not ordered:
- DPDT versions. More poles! (and with zero change in footprint, why not get more poles?)
- Interlocking sets with a "clear" station. They let you add stations in the Down position, and you pop them all back up at the same time with one designated station. The seller's text did not say whether the "clear" station is momentary or latching.
- Switches with more poles. As mentioned, too tall.
- Switches with only PCB pins (no solder lugs). I think the ones without solder lugs could be smaller, and so maybe a 6P version could fit, but I'm not thrilled with the idea of soldering so many wires onto tiny, crowded pins.
- Caps designed for tactile switches. There are many available, but they have round holes; the pushbutton switches need caps with 3.3mm x 3.3mm square holes, according to the sellers.

---

Back to the schematic: and again, Gee whiz! And an extra Golly, for having figured out a good answer to the OP question (how to diagram the switch set). JohnH did mention an option of using separate text describing the interlock.

Yes, I'm overthinking in the wrong place. Always! 

In both of our drawings, the idea was to wire coils directly to pairs of the 3-way's throw terminals, and fix the 3-way's sections to er, roles (positions in the coil selection topology), and then use other switches to activate/deactivate the sections, and yet other switches to do serial/parallel switching (and the now-scrapped idea of adding phase switching would have necessitated yet another set of switches).

Yogi's use of little labels to show connections without having to draw lines (like the little hexagon labeled "A"; what's the official word for those?) keeps the drawing much tidier.

Aside: I really have to take a step back and retrain myself to use Yogi's method for designing (abstracting the components, considering them only as their terminals, and then, still abstracted, "attaching" them to named networks, attaching other components to those named networks and then working out which switches join those same nets... and then drawing it up. I've been too daunted to try, partly because I'm picking projects that are kinda complicated. I'd have to, as I say, take a step back, look at simpler (already done) projects and shove the principles into my stiff brain. Like a guitar player who's learned bad habits but can still play ok, I'm reluctant to become short-term slower in order to eventually become long-term better. Call it short-sighted... or human ... or an elected official.

For now, let me ingest Yogi's gratefully-accepted offering. Looks brilliant at a glance. I think that perhaps further developments could be moved to a new thread, since it's no longer about the principle of drawing the switches, but rather about a particular implementation of them.

BTW, I think that using single-throw switches wouldn't make each station shorter in height, since the little slider inside still would need the space to move up & down. The mfr would have to change the way the switch works overall to gain from removing a throw terminal. In any case, the present (-ly ordered) switches have the 2nd throw, so avoiding the shunt is an option that should be taken. 

[stevewf]

Update: I received some of the switches.
- Switch caps: They all came with caps (red, not my first choice, but oh well). I won on a gamble by ordering some from Mouser (black and chrome) and waddaya know, they fit.
- They seem to work as advertised. The one set that has three interlocking plus one independent switch - that single switch is confirmed self-latching.
- The spacing is not the same for all sets. Again, it's the one with the independent single - it has less space between stations.
- The interlock can be "hacked" by the player. Multiple stations can be locked Down simultaneously; also all stations can be Up simultaneously. All combos are possible (eg my three-station set can do 8 combinations of buttons being Down - including none Down).
- They're confirmed shallow enough to fit in my Monoprice ST-style cavity, but the caps might stick out further than I'd like.
- I still haven't found a bracket-mounted 6P2T self-latching solo (non-interlocking) unit that will fit, depth-wise. Anyone wiser?

Lacking a 6P2T switch, I'm looking at using a pair of interlocking 4P2T pushbuttons (more!?!) to do the job of serial/parallel switching. It can be done, right?