|
Post by frets on Jul 24, 2021 15:43:48 GMT -5
Hi Guys, Happy Saturday!!😺😺😺 I got my hands on an Ibanez Artcore AFS75T. It’s a model I had a few years ago that was destroyed when my Mother’s China Cabinet fell over, smushing it (quite the noise). I’m toying with the idea of doing the illustration below. And it made me think about how many guys (and gals) think that a Coil Split on an HH is really only useful on the Neck pickup? I know it’s a subjective thing; but on this guitar, I really reflected upon the usefulness of a Coil Split to the Bridge. Especially on a big hollow body. I decided to eliminate it in favor of individual Series to each pickup. And I eliminated phase altogether.
|
|
|
Post by JohnH on Jul 25, 2021 16:43:35 GMT -5
Hi Frets
I got to the same views. I built a few guitars with phase switches, and never ever used them at all except to to check ' yep, the phase switch works! ' ,and also to enjoy working out the associated wiring puzzles.
I also agree about the splitting of bridge HB's and although I have those, I never use them, and i do like a neck split. But to build in a bridge split, I prefer a 47nF cap across one coil instead of a full split, which is what i have on my LP.
I guess the choice is about the purpose of the guitar. When its for yourself or for those who know what they like, then i reckon less can be more. But id expect that a new client in your shop, still exploring guitar and never had a custom wiring, would be very excited to try all the options.
On that hollow-body Artcore, you are the only person i know who will be able to thread all that through to install it!
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jul 26, 2021 4:45:34 GMT -5
Not true to the Numbers you want Hexadeciaml Switch with 10M Resister going to each of the 1,2,3,8 Legs (ABCD) E being the Input would be less Resisters, Less Capacitors and more of a Range could trim it down to a 25nF than 28nF and get about a 3n3F differences between them all from 12 to 16 positions
|
|
|
Post by frets on Jul 26, 2021 14:12:15 GMT -5
Angel, You know me well, you’d know I’d try it if I knew how to do it. You know what I really want to use.😺
|
|
|
Post by Yogi B on Jul 26, 2021 14:22:23 GMT -5
Not true to the Numbers you want ... more of a Range "More of a range" is true, both in that (50n7 - 3n9 = 46n8) > (47n - 3n3 = 43n7) and that yours has a greater number of positions — however, "not the numbers you want" perhaps needs more emphasis than you realised. Note where 15n is on both of yours and frets' switches: On frets' switch 15n is the middle cap value, which is about correct if we want the variation in notch frequencies to be distributed over a roughly equal spacing of octaves/semitones/decades/whatever-log-scaled-measure-of-frequency-you-choose, i.e. the geometric mean of 3n3 and 47n is \sqrt{3.3\mathrm{n} \times 47\mathrm{n}} \approx 12.5\mathrm{n} ; whereas the capacitance needed for the mid frequency of your switch is \sqrt{3.9\mathrm{n} \times 50.7\mathrm{n}} \approx 14.1\mathrm{n} , roughly only a third of the way along the resulting values. In the ideal case we want to switch the capacitance values in a geometric progression, where the next value is a fixed multiple of the previous value. Whereas switching capacitors in parallel using a binary coded switch gives (approximately, depending on the values used) an arithmetic sequence, where the next value is the sum of a fixed value and the previous value.
The standard values of components generally follow a geometric progression for each tolerance band. For example, the E12 series (10% tolerance) has values [1.0, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2, etc.] each of which is approximately \sqrt[12]{10} \approx 1.21153 times the previous value. For unknown historical reasons (perhaps something to do with colour codes??) the values used slightly differ from those of the exact sequence rounded to one decimal place, which are: [1.0, 1.2, 1.5, 1.8, 2.2, 2.6, 3.2, 3.8, 4.6, 5.6, 6.8, 8.3, etc.]
|
|
|
Post by frets on Jul 26, 2021 14:34:51 GMT -5
John! I got one push pull stuck. In the bottom corner. I was testing the ability to pull the push pull pots through the enlarged controls and they all fit, including the Varitone. Then in my test, pulling the bottom corner pot back out it got wedged kitty corner. I having spent much time trying to get it out, I will. I’ll get it out. I’m using Bourns pots. But it is frustrating.
I had considered (and still am) using PIO caps on the 6-Way by attaching the caps on long leads, collecting them through the bridge cavity, hot glue them together, and then glue the row of PIO’s behind the Bridge by about two inches. Using superglue gel and holding the group in place until they adhere to the inside front. If I do this, I will take some photos because it’s so crazy.
But, with the test, I am assured all pots will indeed fit.
On the 47nF coil split, I’ve seen a dry switch configured like that. However on the current Spin-A-Split, would it be possible to emulate the effect you’re getting with the 47nF? Whereby the split would see the 47nF but the pot could still pull to its effect? I don’t see how I could do that; I.e., have full HB but then be able spin to a 47nF split?
I’ve seen 4Real’s Spin-A-Split mod with cap but I think it’s pull is a coil select. I probably am wrong.
But I’m glad to know someone else feels as I do about phase. Guys are constantly asking for it in guitar harnesses I build and I don’t get it. I’m not sure they do. I sometimes feel like the read about it on a forum or guitar mag and then just want it.
Oh, and a tip for all of the newbies out there, when you want to enlarge an import control hole to SAE Imperial size, use a round file. Don’t go the drill route.
|
|
|
Post by frets on Jul 26, 2021 14:37:43 GMT -5
Yogi,
That is interesting. The only ones I know that could actually make a switch like that are you and Angel😺. I’d love to have it.
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jul 27, 2021 1:57:49 GMT -5
The big problem with Summing Capacitors is the 10% , so 28nF could be 25n2F and just need few high and a few low and then you can have over lapping numbers
i think Hex has more of a Fault with the 10% but you get that as well with the Capacitor switching . do think you wouldnt even notice the difference of 10 @10% to 3.9+6.8=10.7@10%
Could we do a SOUND off, between the Two types of Systems HEX Vs 1P12T just the but just the Par settings
Rereading this .. never seen MATHS like this SQR (AxB) in any maths , Electronics or Electricals
Old English saying "Im hitting all the right Notes just not in the Right order "
|
|
|
Post by Yogi B on Jul 27, 2021 5:57:00 GMT -5
The only ones I know that could actually make a switch like that are you and Angel😺. I’d love to have it. It's more of a physics/maths problem with combining impedances, than a switching problem. Combining multiple components of the same value can easily give: arithmetic sequences, e.g. [1n, 2n, 3n, 4n, ...] by combining capacitors in parallel; or harmonic sequences, e.g. [1n 1n/2, 1n/3, 1n/4, ...] by combining capacitors in series. However the easiest way to get a geometric progression is by using separate caps of differing value. Rereading this .. never seen MATHS like this SQR (AxB) in any maths, Electronics or Electricals Along with the arithmetic mean and the harmonic mean it forms the triplet of Pythagorean means known about since antiquity. However since nowadays "mean" is generally is synonymous "arithmetic mean", even when the others do come up in maths people don't usually think "oh, I'm finding the such-and-such mean". Oddly the best example I can think of involving the harmonic mean is from the world of electronics: if we have a number of differing resistors in parallel, and wanted to replace them with an equal number of parallel resistors but all with the same value — this required value is the harmonic mean of the resistances. (In other words, the value of the total parallel resistance is equal to the harmonic mean divided by the number of resistors.) Whereas the geometric mean comes up a little more often, usually in the contexts of geometry (unsurprisingly) and music: the position of a guitar fret (in equal temperament, and discounting intonation offsets) can be calculated as the geometric mean of two other frets — i.e. the position of the 6th fret (as measured from the bridge) is the geometric mean of the distances measured from the bridge to the 5th fret & to the 7th fret (or 4th fret & 8th, or 12th fret & nut, etc.). Due to the laws of logarithms, the geometric mean also comes up when we want to find the arithmetic mean of log-scaled values without needing to convert back and forth between the linear & log scales. That is, in general: {1 \over n} \sum_{1 \leqslant i \leqslant n}{\log(x_i)} = {1 \over n} \log{\left(\textstyle\prod\limits_{1 \leqslant i \leqslant n}{x_i}\right)} = \log{\left(\left(\textstyle\prod\limits_{1 \leqslant i \leqslant n}{x_i}\right)^{1 \over n}\right)} = \log{\left(\sqrt[n]{\textstyle\prod\limits_{1 \leqslant i \leqslant n}{x_i}}\right)} (i.e. where valid, the arithmetic mean of the logarithms of a set of numbers is equal to the logarithm of the geometric mean of the numbers.) Or in the specific case with only two values: {\log(a) + \log(b) \over 2} = {\log(ab) \over 2} = \log{\left((ab)^½\right)} = \log(\sqrt{ab})\\ Wise words, but hang on weren't they Morecambe's?
|
|
|
Post by gckelloch on Jul 27, 2021 6:14:03 GMT -5
Yes, pickups wired out of phase sound very thin. That's why someone (Bil Lawrence?) came up with HOoP (Half Out of Phase) option. Just use a cap going to ground on one coil, as you planned, and wire the coils out of phase. It will only cancel the range between where the cap rolls off the low and high end of each coil. In my experience, it sounds more like a SC than when wired in phase, but possibly sweeter than just using one coil, depending on the range it cancels. I have it set up as a blend on one guitar to cancel the most sensitive 3-3.5kHz hearing range. It's a subtle effect, but very nice. A 47nF cap would cancel down in the ~700Hz range for a ~4.5H HB pickup -- the opposite effect of the classic 600-800Hz tone knob peak for a horn-like cleanish tone. You could instead aim for a lower cancelation dip to reduce muddiness from 250-400Hz but you'll get some fundamental note loss in that range, or go higher for a mid-dip in the nasally 1.2kHz range that normally pushes 2nd or 3rd harmonics in the harsh range. A ~2kHz range dip sounds nice too for a less metallic/more vocal quality. Those are the ranges I'd aim for.
|
|
col
format tables
Posts: 464
Likes: 24
|
Post by col on Jul 27, 2021 8:31:37 GMT -5
Wise words, but hang on weren't they Morecambe's? Ah - that's right! I knew the phrase (of course), but did not recall its origins. I believe the phrase is more correctly, 'I'm hitting all the right notes, but not necessarily in the right order'.* * Actually - nope. I checked - my correction above is only partial. Eric Morecambe to Andre Previn: "I'm playing all the right notes. But not necessarily in the right order."
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jul 27, 2021 14:26:15 GMT -5
Sqr((R^2)-(Xc^2)
Never seen sqr (a x b)
Im perfect word. 1) 3.3 2) 6.6 4) 13.2 8) 26.4
Why i need number near but just far away If I wanted it pure then i could use 15 capacitors all summing together of the same value.
As dealing with cost capacitor values that change and the human ear that sucks This hex switch would be a bother
So if we can sort it a Switch Face off. hex vs 1P12T and just do the 12 stages that is is near
|
|
|
Post by unreg on Aug 31, 2021 18:13:27 GMT -5
kind note: Regardless, his math at the end there is correct… (log (a) + log (b)) / 2 (log (a*b)) / 2 log ( √(a*b)) are all equal. 🙂
|
|