|
Post by sumgai on Apr 19, 2006 16:49:35 GMT -5
Gang,
Project abandoned due to inability to realize the goal.
It's back on, sorry for the false alarm. See Reply #18 (on the next page) where I get back on track. Yes, the goal has been realized.
sumgai
|
|
|
Post by ChrisK on Apr 19, 2006 18:08:17 GMT -5
Edited for human consumption 20060420. Yep, Inheritance is a bear. Positional inheritance, that is. If two pickups (A and B) are in parallel, and the third (C) is in series, you want (A+B)*C. If two pickups (A and B) are in series, and the third (C) is in parallel, you want (A*B)+C. Assuming equal inductance, the former is 1.5 * nominal, while the latter is 0.67 * nominal. The resulting tone is different. It's further interesting in that the resulting inductances are the reciprocals of each other. In my ToggleCaster post, I added a mode switch just to allow selection of the desired resulting mode. This design was not intended to be the final solution, but rather an exercise in what one could realize with toggle switches, one per pickup for mode, and one per (pickup-1) for phasing. I did do a reductive exercise for the final solution a few years back (March 2003). I discarded the M*(B+N) config since I never thought much about it. Of course, Fender surprised my by including it within the Am Dlx SSS Strat (of which I'm fond (as fond as I get on Fender factory gear) since I apparently have two of them (the one w/ the rosewood board is in pre-OP for a set of Rio Dual Calibrated pickups)). Anyway, to realize all (but as above) SSS modes (inclusive of phasing) required an 8 pole (I think, inclusive of phase) 5 position switch for the bridge and neck (-S, -P, OFF, P, S) and a 10 pole 3 position switch for the middle (P, OFF, S). I'd called the design 2x7p3t_10P3T, apparently I'd seen another pole optimization beyond the obvious one for the bridge and neck. One might note that I tally only 16 unique positions (sans the (b+n)*m one, and ALL OFF) exclusive of phasing. NOTE: We actually need two 8P5T and one 8P3T switches. One could use just three 8P6T Grayhill's and position pin away. (Note to the demented, the Schaller fabled 8P5T would be "peachy" for all three switches (if'n we can figure out how to limit one to three positions, I haven't at all sussed out the use of a 5 throw for the middle pickup, but it may well be straightforward.) There are 3 combinations of 1 of 3 pickups; There are 3 combinations of 2 of 3 pickups, in parallel; There are 3 combinations of 2 of 3 pickups, in series; There is 1 combination of 3 pickups, in parallel; There is 1 combination of 3 pickups, in series; There are 3 combinations of 3 pickups in SSP; There are 2 combinations of 3 pickups in SPP (as sans above); For a total of 16 (excluding phasing and NONE). I did check my work on this SCHEMA (not schematic) and eliminated the obvious pole redundancy. Here's a SCHEMATIC using Grayhill rotary switches. I do find the use of symmetrical knob legends to be fast (enough for me) to change while playing, but how many different set up does one need? (I think that's why we need many different guitars.) ;D There are a total of 27 switch combinations (sans phasing) inclusive of OFF on the schema. On the schematic, there are a total of 75 switch combinations inclusive of phasing and OFF on the schematic. I'm too tired to reductively suss out the different ones tonight. I know that I can do this mathematically, but when I close my eyelids, I still can see the screen changing as I type. I never built it, but did source the series 71 rotary switches (in all of their maximum possible extrapolations) from Grayhill. I decided that I can fit a series 71 w/ 4 decks within a Strat or Tele depth body. This allows an 8 pole 6 position (the series 71 mechanism is pin insert settable for any subset number of positions), a 12 pole 4 position, a 16 pole 3 position, and a 24 pole 2 position maximum possible configuration. Note: The 4P6T switch is $20. We're looking at about $80 for the three switches (but I gots mine already - engineering samples rule![sometimes]). I'd actually done this exercise for my MIM Nash Power Tele (sans power). A volume knob, three rotary switches, and a tone knob. It will fit on a tele plate! Such designs (phasing-based) actually work much better on guitars w/ dissimilar pickups (such as the MIM Nash, Tex-Mex Tele bridge, Tex-Mex Strat middle, and a Tex-Mex Tele neck) IMHO. I guess that I have to build it now ........And fabricate the plate. Isn't CAD wonderful sumgai? One can bind (via layers) the selected modes to the switch wipers, and easily check one's (mis)wiring.
|
|
|
Post by UnklMickey on Apr 19, 2006 18:20:30 GMT -5
Sumgai,
i hate to burst your bubble, but...
look what happens when none of the selectors are in the 4 or 5 position.
i did some work on this same concept, months ago.
it is possible, without those dead spots.
unk
BTW:
"What we have here is what is properly called a Series-Parallel switching scheme. In short, after about 4 hours, I am forced to conclude that doing a Parallel-Series scheme just isn't possible with regular Superswitches. Can anything else be done?"
yes, parallel-series can also be done.
|
|
|
Post by UnklMickey on Apr 19, 2006 23:08:54 GMT -5
"unk,
What happens when none of the switches is in Position 4 or 5? Why, we get a single, dual or triple combination of parallel pickups, all of them in phase if Position 1 or 2 is chosen exclusively. If a mix of Positions 1 and 2 is chosen, then we'll get something out of phase. Check it out." i did check it out. if none of them is in 4 or 5 you get a short. i.e. no sound. like i said before, i worked on this a while back. so i'm familiar with the pitfalls. i'll be back tomorrow if you need some help. unk
|
|
|
Post by ChrisK on Apr 20, 2006 1:02:09 GMT -5
I edited my original response for human consumption.
And I agree, a system that can actually be switched and played (as in real time) is important. the long and short of it is, I have some trouble needing anything more than what Mike Richardson came up with.
Ooooh, kimschee (I 'aven't a clue as to how to spell it, but I like (most of) it). Ate way too much when in Korea the many times while in a past corporate life. The good stuff is buried for months to "set up" properly. Korea does have a high incidence of stomach cancer, hmmmm.
(However, when my hosts took me out to show me what spicy squid would do to a gaijin (spelling again?) (a foreigner, or round eye), I prevailed mightily. Ate two large serving bowls, and wiped the second out w/ bread. The restaurant declined to bill for this cultural "(re)training session".)
|
|
|
Post by CheshireCat on Apr 20, 2006 3:28:12 GMT -5
look what happens when none of the selectors are in the 4 or 5 position. If I'm not mistaken, the whole thing shorts out.
|
|
|
Post by JohnH on Apr 20, 2006 7:45:50 GMT -5
Whaaa...what were you guys trying to do here? John
|
|
|
Post by ChrisK on Apr 20, 2006 14:20:16 GMT -5
sumgai,
Once the 8P5T Schaller lever is available, one can do my full schema. You could do a subset thereof. Limiting one to 3 positions for the middle is moot, just wire reverse phase positions the same as the normal phase, since proofing of phase for the middle PU hasn't been done in my schema, and redundancy is better than embarrassing silence.
Of course, if there is any market for this schema, I could make PCB's for the Schaller switch. The bridge and neck would be identical, and the middle different. I do need to find out what plating they are using, probably silver or hopefully Rhodium. We'd also need Molex-style interconnect cables.
|
|
|
Post by CheshireCat on Apr 20, 2006 17:34:16 GMT -5
Gang, Project abandoned due to inability to realize the goal.sumgai DUDE!?! Did you need to obliterate the whole thing!?! I was kinda looking forward to seeing this develop. Granted, it may not have panned out, but it could have offered some really excellent opportunities for exploration. Oh well. Still, fascinating idea. Chesh
|
|
|
Post by sumgai on Apr 20, 2006 22:49:04 GMT -5
Chesh, Yes, there was no point in the project as envisioned. Regardless of how I feel about 'minimal effort' when switching, I temporarily lost sight of the fact that if you want all the pickups in series, then you're not gonna get anything in parallel at all, no matter how many poles or throws one might add. Believe me, that lesson is etched in my pea-sized brain much more deeply than anything else I've learned this year. sumgai p.s. Yes, there is a way to make it all work - just add a resistor in each leg (1 for par, 1 for ser), and add active electronics to compensate. If there's further interest, I'll re-publish the schematic.
|
|
|
Post by sumgai on Apr 20, 2006 22:56:54 GMT -5
John, Sorry, but you weren't quick enough to see my brain-fart, whereupon unklmickey (correctly) made mincemeat of my alleged masterpiece.
Besides, I somehow mislaid my usualy courteous self, and said some things about your LP Maximiser circuit that were not called for. Hopefully that won't happen again, but since they were the bread and butter of what I was attempting to accomplish, it was better to just drop the whole thing.
sumgai
|
|
|
Post by ChrisK on Apr 20, 2006 23:27:52 GMT -5
Uh, I ain't buying it
'cuz I do it. Just neet the Schaller 8P5T lever!
And I'm getting real warm about using levers 'cuz they're "touch (position feel) sensitive".
What's the cost on the Schaller 8P5T. Does anyone know where to order it in Germany? I'll buy some if anybody knows.
|
|
|
Post by CheshireCat on Apr 21, 2006 0:18:15 GMT -5
Sorry, but you weren't quick enough to see my brain-fart, whereupon unklmickey (correctly) made mincemeat of my alleged masterpiece. You're kidding, right? Show me one masterpiece that didn't have a flaw in it along the way which eventually got worked out. There's no telling how many flaws Michaelangelo's "David" had going along. Fortunately for Mickey A, and the Art world in general, he probably had a lot of marble to work with at first, and only when the final buffing was done did we realize what a magnificent masterpiece it was. For instance, for all we know, David's thighs might have been bigger. Or perhaps for all we know David's hands may have been envisioned as anatomically correct, and would have been, were it not for perhaps Mickey A discovering at the last minute that if David's hands were rendered as anatomically correct, they wouldn't have been weighted properly, optically speaking, and look minuature and dwarfish when David was finally set on his rooftop perch. A lot of people didn't know that David was meant to be on a rooftop, and wonder why his hands were oversized, not realizing that they were weighted to be optically correct from that distance. Also, consider, a perfectly flown airline flight is offcourse some 90% of the time. Yet, it still manages to reach it's destination without event and without the passengers being any the wiser. It's all par for the course, dude. If only you knew how many times I've rewired my Utah, not to mention a host of other things . . . Chesh
|
|
|
Post by sumgai on Apr 21, 2006 3:43:42 GMT -5
Chesh, You got it right in your first post. The insurmountable problem here is, if at least one pickup isn't in series mode, then the "bypass" for each and every pickup is engaged. That's why The Unkster didn't have to look any further - a dead short is still a dead short, no matter how many switch poles you put in the path. The other side of that same coin is that if you don't provide the bypass function in the first place, then whichever pickup is 'bypass-less' will always be in the signal path. To fix that pretty much requires that at least one possible series combination be left out of the picture. Again, that fails to meet the goal of 'no compromises'. But seeing as how I like you, I'll tell you what I'm gonna do. Here are my previous images, but I modified them to show the short circuit with red lines over the top of the original black lines. The schematic: .... and the layout diagram. This one should make it crystal clear why unklmickey was right: sumgai
|
|
|
Post by sumgai on Apr 21, 2006 3:55:43 GMT -5
Chesh, Umm, that wasn't the issue. Believe me, I've made even bone-headeder mistakes (and lived to tell about it), and my ego just isn't up to the task of feeling like nobody loves me. I've got my head on more square than that, my fine feathered friend. The deal is, if the envisioned project can't meet the goals set out for it, then it's a waste of people's time to even consider looking at it. I know that I get more upset when I spend 20 minutes reading what I'm sure is drivel, only to have it confirmed, than I am at someone's lack of proof-reading skills (to be PC about it). That's where I differ from most folks - I can honestly self-assess the value of my output, and if it's lacking in the essentials, then I can and will remove it from the landscape. I don't wanna be on the receiving end of a mental assault that says "that idiot wasted my time". Did I make myself clear here? (I'm not up to unk's level of vagueness yet, but I'm working on it!) sumgai
|
|
|
Post by jhng on Apr 21, 2006 7:31:30 GMT -5
Out of curiosity, what was the design goal here and what controls were you planning on using?
I'm guessing a switching scheme that does all 17 SSS non-phased combinations (including all six hybrid series/parallel ones). It would be lovely to get at 16 of them (probably abandoning the awful [N+B]*M) using two four position switches...
With another four position switch to do the phasing...
(I can foresee a weekend surgically attached to my pencil and paper...)
Hastings
|
|
|
Post by UnklMickey on Apr 21, 2006 9:33:35 GMT -5
...What's the cost on the Schaller 8P5T. Does anyone know where to order it in Germany? I'll buy some if anybody knows. i'm hoping to have much information for you early next week. when i do get the info, i'll update on the megaswitch thread. unk
|
|
|
Post by CheshireCat on Apr 21, 2006 13:10:04 GMT -5
The deal is, if the envisioned project can't meet the goals set out for it, then it's a waste of people's time to even consider looking at it. I know that I get more upset when I spend 20 minutes reading what I'm sure is drivel, only to have it confirmed, than I am at someone's lack of proof-reading skills (to be PC about it). That's where I differ from most folks - I can honestly self-assess the value of my output, and if it's lacking in the essentials, then I can and will remove it from the landscape. I don't wanna be on the receiving end of a mental assault that says "that idiot wasted my time" Run the parallel lead and the series lead each thru a preamp. That should keep the signal from running backwards and shorting out. You can get them cheap these days.
|
|
|
Post by sumgai on Apr 21, 2006 19:56:33 GMT -5
Chesh, What you say is true, but that requires two small amps, not just one. I had earlier suggested using a summing amplifier at the junction of two resistors, one each for the Series leg and the Parallel leg. Lower cost, less battery drain. And certainly less headache than the design exercise I just went through!! However, I've noodled around, and have to admit, if there is a will, then I wanna be in it! ;D No, seriously, the solution may have fallen into my lap, but before I say "Eureka!", I'm letting everyone vet the thing first. And even before that, Hastings has asked that I review the goals. Fair enough, seeing as I pulled the trigger on the original post pretty quick. 1. Implement as many possible combinations from as few switches as possible. Don't leave something out because it doesn't sound good to someone else. 2. Use standard Strat-style blade switches, yielding the best tactile feedback and requiring the least amount of coordination while on stage. It's a matter of ergonomics. 3. One switch per pickup gives both parallel and series possibilities, and with phase reversal directly available on that same switch. Again speaking to ergonomics, the associations one makes mentally with "one pickup, one switch" allows for nearly mindless grabbing of a desired tonal combination. And yes, as a matter of fact, I do need all the help I can get like this when I'm in the thick of things! ;D 4. Cost of the switch, for this project, is of no concern. Given all that, I had started out with three normal Superswitches, and thought I had it. Unklmickey did his usual proofing job, and lo and behold, I had forgotten the most basic tenet, you can't short all three series pickups if you want anything to come out on the parallel side. So I thought it over this morning, and I reasoned that I only needed to break the series chain while selecting one or more pickups for parallel. Of course, I'd also have to restore that broken connection when I selected two pickups for series, and the remaining one for parallel. At that point in my self-delusion, the 8 Pole Megaswitch popped into my mind. With a fevered mouse (in AutoCAD), I quickly added the new sections, and started connecting them. It turned out to be much easier than I had originally lamented. In fact, I spent more time making the ancilliary drawings for the purpose of following each path with colors to lead the eye, for each possible switch combination. That got old in a hurry, let me tell you! So I've posted the main layout diagram and three others, which I'll explain as we go along. Ready? This is the overall layout drawing. (No schematics this time, I don't need any complaints about eye doctors and such! ;D) Note that as in the previous incarnation, the middle position (Pos 3) is off for each pickup. That means that a short exists in the series side of things, between Poles 2 and 4. But where I had merely joined (shunted, for you Britishers) the terminals for Pos 1/2 to Pos 3, I now run them through a complicated switching scheme that, in essence, breaks the series chain if certain conditions are not met. More details on that as we go along. Next, let's look at the signal path for the Neck pickup being switched on in Parallel mode. Yes, the thick red lines are hard to follow, that's why I put in those little arrows, to give you a clue. (Hey, I needed them, and if that were the case, then why should I cut you out of the fun, eh?) In summary, if the Neck switch is set to Pos 1/2, Poles 2 and 4 carry the series connection. However, if only the Middle switch or only the Bridge switch is set to Series, then the series chain breaks at Pole 6 of the respective switch. If both Middle and Bridge switches are set to series, then the series chain is intact via Pole 6 of those two switches. Let's go through the drill for the Middle pickup: If the Middle switch is set to Pos 1/2, Poles 2 and 4 carry the series connection. Howver, if only the Neck switch or the only the Bridge switch is set to Series, then the series chain breaks at Pole 6 of the Neck switch, or at Pole 8 of the Bridge switch. If both Neck and Bridge switches are set to series, then the series chain is intact via Pole 6 of the Neck switch and Pole 8 of the Bridge switch. And last but not least, just so you can collect the whole set of 'em: ;D If the Bridge switch is set to Pos 1/2, Poles 2 and 4 carry the series connection. However, if only the Neck switch or only the Middle switch is set to Series, then the series chain breaks at Pole 8 of the respective switch. If both Neck and Middle switches are set to series, then the series chain is intact via Pole 8 of those two switches. Yes, I'm still missing the 12 possible (1 + 2) x 3 combinations (including phase reversal). Maybe that will come about some day, but not just yet. But I"ve still got 36 valid tones to deal with here. Will I use them all? Not likely. But one never knows when something might come up, or even that serendipituous moment when you hit an unintended combo, and out comes just what the doctor ordered, even if you didn't. Such are my thoughts at the moment, anyway. And finally, I've not checked every combo yet, but I know that I've deleted at least 12 duplicates. If you switch on any two pickups in parallel, you can't "accidentally" add the remaining pickup by selecting series. It used to be that you'd get all three in parallel (the shorting chain was broken), but now, you don't get any difference in tone - the series chain is still broken, due to the two parallel selections. Enough, I gotta go look like I've been busy before the Boss gets home! sumgai ( Edit: pointed the drawing links to a new server)
|
|
|
Post by Mini-Strat_Maine on Apr 21, 2006 21:24:59 GMT -5
Ohmigosh, the Unknown Diagrammer's been here.
|
|
|
Post by JohnH on Apr 21, 2006 21:46:36 GMT -5
Sumgai - thanks for putting the diagrams up. Great ideas but I suspect Im not following something here. Those paths through the circuit via the red and blue all seem to show a direct connection from output to ground = short circuit?
cheers
John
|
|
|
Post by UnklMickey on Apr 21, 2006 22:09:29 GMT -5
Sumgai,
thanks for not giving up.
you are taking a different path at solving the problems with the shorting, than i did.
i think you may be onto something there, so i won't bias you with my ideas yet.
mine still has one minor unresolved problem.
this is turning into what might be an interesting exercise,
of how 2 different approaches could be used to solve a problem.
i'll wait 'til next week to upload mine, and we'll compare notes then.
maybe i might be able to solve an unresolved issue in yours.
maybe you can solve the final piece to mine.
if nothing else, you're creating a tutorial on the way a design problem is attacked.
it will be a great learning experience for the intermediate designers who are watching.
off for the weekend.
see you next week!
unk
|
|
|
Post by sumgai on Apr 21, 2006 22:30:00 GMT -5
John, It's the curse of too much versus too little information. My intent was to show the paths of several combos with one drawing doing double or triple duty, if you will. I think, after looking at it some more, that maybe it would be more clear for folks if I said: "These are basic diagrams. The first one starts by assuming that you put the Neck switch into the Parallel position (1 or 2). From there, you can follow the arrows in the series chain until you get to a blue section. If the terminals covered by that blue section are not selected by the switch being in series mode, then the series chain stops dead right there - the rest of the red line is null and void. Given that any pickup is put into parallel, the requirement is, both other pickups must be selected for series, or else the chain is broken. From this, you can trace out the effects of selecting any two pickups in parallel mode, and see that the remaining pickup can not be selected in series mode - the series chain remains broken." I could have posted 36 drawings, one for each and every case. (But I would have had to make 'em!) I'm sure that once you see the "stops at blue if....", you'll see how one drawing does the work of many. (I hope, I hope, I hope! ) Did that help, partner? sumgai
|
|
|
Post by JohnH on Apr 21, 2006 22:41:53 GMT -5
Sumgai - OK - Im with you now, thankyou! John
|
|
|
Post by sumgai on Apr 21, 2006 22:42:53 GMT -5
unk, Pretty rare, that. I sometimes "mis-prioritize" my time, if you get my drift. The wife thinks I do it all the time. JohnH: So does my cat! Yeppers, that's the ticket! Tell a man how to fish, and he'll eat for the night. Show him how to fish, and he'll eat for a lifetime. But teach him how to use the innerweb, and he'll leave you alone for weeks afterwards! ;D No, seriously, if we can get some of these guys a leg up on how to make ideas into reality, then who knows what will come popping out of the woodwork? Call it my 'hidden agenda', if you like. sumgai
|
|
|
Post by jhng on Apr 22, 2006 11:11:14 GMT -5
Sumgai,
Glad you resurrected the project. Very interesting!
I have a little voice in my head saying that it might be possible to do it using "ordinary" (for guitarnutz!) four-pole-five-way switches.
I can't see the whole thing yet, but the starting point would be connect the hot and cold of each pup to two POLES on each five-way. Then those two poles can be used to put the pup in and out of phase and "put" it into the required position in the wiring set up. The two left over poles from each five-way can be used to fix up all the rest.
Hastings
|
|
|
Post by jhng on Apr 22, 2006 11:19:06 GMT -5
Sorry. Accidentally duplicated my last message.
|
|
|
Post by sumgai on Apr 22, 2006 13:06:02 GMT -5
Hastings, (emphasis added) The cost of those 8P switches (you need three of 'em) just might be the straw that breaks the bank, tipping most folks towards the 'resistors and summing amplifier' path of enlightenment. I am getting less and less "resistant" to going this route myself. Good luck on finding that shortcut, my friend. I've spend quite a few hours with this thing, and I'm not done yet! I started out as you stated, using the wiper (or common) terminals of each Pole to hold the pickup's + and - wires, then connected the other terminals as needed. When the rat's nest had finally overflowed onto the next piece of paper, I swapped things around a bit, and what you see is not only a whole lot neater, it solved several logistic problems, too. The lesson I've taken away from all this is: In all things series versus parallel, there are actually two entirely separate paths from ground to output (or 'hot'). One has to keep that in mind when devising switching circuitry, otherwise one runs the risk of doing just what I did at the outset, and that is, make a stupid mistake for unklmickey to spot. ;D sumgai
|
|
|
Post by sumgai on Apr 23, 2006 0:34:04 GMT -5
OK, time to up the ante! You see that this project is meant for a SSS style guitar, right? But I'm sure you've also noticed that real-meal-deal here is that I got what I was after, and that was a "one coil, one switch" affair. Each switch, all by itself, selects either parallel or series, and in or out of phase. Got that so far? Good. Now let us all remember that this forum isn't focused solely on SSS users, there are beaucoup 2Hb users here too. Are you starting to get a tingle yet? Tha's right, boys and girls, you can increase your fun time by a whopping 33% if you apply this circuit to the four coils found in a 2Hb guitar. Of course, you'll need to add another switch, but the overall theory is still the same - one coil, one switch. And yes, there are enough poles to make this work. Now what are ya gonna do with your so-called "spare time"? <laughs like The Shadow, because only he knows, hehehehe> sumgai
|
|
|
Post by CheshireCat on Apr 23, 2006 0:39:03 GMT -5
OK, time to up the ante! Tha's right, boys and girls, you can increase your fun time by a whopping 33% if you apply this circuit to the four coils found in a 2Hb guitar. Of course, you'll need to add another switch, but the overall theory is still the same - one coil, one switch. And yes, there are enough poles to make this work. No brainer. Just add an extra switch to the extra coil. (Incidentally, if one SuperSwitch worked as a subset of the other one with each humbucker, then you could do some very fascinating innovations indeed . . . sort of an adapted Double Barrel Switching mod for each humbucker.)
|
|