Variable Interstice Pickup Prototype

[KhzDonut]

This project is in some ways the offspring of my previous (failed/abandoned) project:

Multi-Coil Prototype - Wal-Style Pickup For Guitar

In that design, I had intended to have the coils be small enough that they could be squeezed together on the bass side, and spread apart on the treble side.

Example, though here I'm holding it rotated 180-degrees like a derp.




The idea is that I want to get a more "focused" sound on the bass side, for tighter lows, a shift towards upper mids by moving the North coil closer to the bridge, and hopefully attributes that would align with high-gain applications in general. The hope for the treble side is that it would either retain a typical humbucker sound, by using narrower, taller coils, it would provide a more "vocal" or throaty sound with smoother highs.

Given the dimensional space that I have to work within, the standard humbucker routes probably don't afford enough space for this design to get a significant effect from the physical design, but hopefully the accumulated subtle differences in design would still culminate in a perceptible change to the tone.

I haven't known what to refer to this design as, I've thought of calling it a "variable aperture" design, but I see people refer to the coil itself as having an aperture, so I found some old-timey word that also describes the space between two things and used that instead. So, I give you, the "Variable Interstice Pickup"

I'd trademark it, but that would be like... Effort.

My multi-coil design suffered from the fact that I wanted to connect everything in parallel, and, not understanding the physics at the time, would have needed a preamp to achieve my desired goals. So I scrapped the design, having not enough expertise to create a preamp.

I decided to just do a typical 2-coil design, but with taller, narrower coils like a typical Strat pickup, and a "Variable Interstice" by way of angling the North Coil so that the bass side was closer to the bridge, reversing the angle of a typical Strat.

I had the option of using Alnico pole pieces or steel slugs with little Neodymium button magnets on the back, because obviously a single bar magnet wouldn't work for this. I decided to go with Jumbo Steel Slugs and Neodymium magnets.

Started by modeling the bobbins on a typical 7-String Passive Humbucker, as that was the mounting bracket my Squier has routed into its pickguard at the moment, for reasons that are numerous and mostly having to do with horribly failed mods






Cut some Forbon, I think it was .063"? I don't remember. The pole pieces are just 1018 steel in .25" and I cut it all by hand with a Hacksaw. Just marked every .75" or so and cut it. I drilled little 1/16" holes and then used a 1/4" Forstner bit to drill them out bigger.




You can see some of the crummy drilling templates I made out of MDF, which didn't work all that well, but were better than eyeballing it.




Closeup of complete Bobbin. It just applied glue and kind of eyeballed the placement. It's bad, but it technically worked.




Visual comparison of my bobbins compared to a standard Passive 7-String Humbucker.




I've made much more progress at this point, but I have to go drink beer and eat fried chicken so I'll have to post about winding bobbins tomorrow

[Antigua]

You could wire the individual coils in series, which is what Zexcoil does IIRC, but regardless, one thing that seems to be true based on recent testing is that you get the most voltage when the coil area is wider, while a smaller coils around the pole pieces would result in smaller coil areas.

Where the two coils are farther apart, you'd get more comb filtering. You can model the filtering with this too Guitar Pickup Response Demonstration , but set the aperture width value to something, approximately the size of the pole piece.

[KhzDonut]

Thanks for the info and resources!

I'm not sure how to really utilize the pickup response demo you linked to, other than that it is neat to punch in the numbers and see how subtle or dramatic the results are. I know all the frequency curves change a lot based on all the other factors that go into building, so I know I have a lot of maths and learning to do before I can really utilize it as a theoretical tool. I'm really surprised by how deep the cuts can be on the comb filtering, it just doesn't sound that severe when you're listening to someone play a strat or something. I just don't expect huge cuts in random places to be the way that sound is created, but I guess how else could it change it that much from position to position if it wasn't fairly dramatic?

The more you know, I guess

.

As for the Zexcoils, I wanted to be able to really tune the individual coils, so just wiring them in series wouldn't have allowed me to, say, put a LPF on the high strings to cut harshness, etc... I really wanted to have control over every string's response. Someday, when I learn how summing preamps work, I'll figure that out

[Antigua]

Thanks for the info and resources!

I'm not sure how to really utilize the pickup response demo you linked to, other than that it is neat to punch in the numbers and see how subtle or dramatic the results are. I know all the frequency curves change a lot based on all the other factors that go into building, so I know I have a lot of maths and learning to do before I can really utilize it as a theoretical tool. I'm really surprised by how deep the cuts can be on the comb filtering, it just doesn't sound that severe when you're listening to someone play a strat or something. I just don't expect huge cuts in random places to be the way that sound is created, but I guess how else could it change it that much from position to position if it wasn't fairly dramatic?

The maximum operating frequency is around 4kHz, so if you input two pickups, with a width of about 0.25 inches, and space them as far apart as your coils are, you can see if the resulting comb filtering cuts down below 4kHz. For example, here's a model with the coils 0.8" apart https://i.imgur.com/kzgUnjV.png versus 0.2" apart https://i.imgur.com/Z93V5AY.png , you can see that as the farther apart the coils, the more scooping you get in the higher frequencies, which is the difference you would hear between the strings with the coils close together, versus farther apart. Incidentally, there are ways to achieve similar "scoops" electronically by splitting two series coils with a lower value capacitor.

As for the Zexcoils, I wanted to be able to really tune the individual coils, so just wiring them in series wouldn't have allowed me to, say, put a LPF on the high strings to cut harshness, etc... I really wanted to have control over every string's response. Someday, when I learn how summing preamps work, I'll figure that out

That's true, although even with parallel wiring, the filter would have to be in series with the parallel coil, which would practically limit you to high pass filtering for a given coil. Another approach to attenuating the high end for a specific string would be to try inducing eddy currents by placing copper in between the pickup and the strings, but testing has shown that eddy currents will attenuate the response of the entire coil, so that could be a case where two coils are better than one, if the eddy current effects can be isolated to a specific coil. That's worth testing out.

[LtKojak]

As for the Zexcoils, I wanted to be able to really tune the individual coils, so just wiring them in series wouldn't have allowed me to, say, put a LPF on the high strings to cut harshness, etc... I really wanted to have control over every string's response. Someday, when I learn how summing preamps work, I'll figure that out

Well, this is not entirely true. The tech behind the zexcoil products allow for a certain "fine tuning" for each coil, although it's proprietary tech so I don't know how much Dr. Lawing will be willing to share if asked... but I'm pretty sure he'll share enough knowledge at a conceptual level, to "inspire" your own research if asked nicely and politely.

Sign up and follow his blog. A whole lot of very useful p'up info is there. Heartly recommended!

[KhzDonut]

The maximum operating frequency is around 4kHz... (more useful info) ...if the eddy current effects can be isolated to a specific coil. That's worth testing out.

Thanks for all the info, I really appreciate it, I have to admit that this is well into territory that is over my head. Without further digressing into the multi-coil discussion, I can say that the flexibility I really want seems to only be achievable through an active preamp that would give me total control over individual strings, so I'll come back to it when I get to that point. This info definitely gets me in the headspace of wanting to go back to it sooner than later, though

The tech behind the zexcoil... (more useful infoo) ...Sign up and follow his blog. A whole lot of very useful p'up info is there. Heartly recommended!

Thanks for the recommendations, I really appreciate it. I'll check out his blog, that sounds awesome, and I'll return to the multi-coil designs when I can get a preamp designed that will give me the expanded flexibility and control I'm wanting to get, but I really appreciate the help with a passive-version design. It will certainly give me some prototyping material as I work through the process in the future

[KhzDonut]

Variable Interstice Pickup Prototype Part 2: Electric Boogaloo

Was trying to get to this yesterday, but life gets in the way, as it does.

First, thanks to everyone that's already come in to help me with this, I'm always surprised at how instantly supportive this forum is to people like me that have no clue what they're doing from an electrical engineering standpoint.

I feel really comfortable with building things, not so much with math

I wound each coil with 8,000 winds of 44awg wire, which is a lot for a humbucker. Part of the reason I'm making these pickups is to try and alleviate certain problems that exist for people who tune low, use high output pickups, and high-gain amplifiers. Making mega-tight lows makes for thin and shrill highs, and warm, full highs make for muddy lows. When using 7, 8, or 9 string guitars, the frequency spread becomes extremely hard to deal with in fast and complex compositions.

I'm hoping the angled bobbins help to more appropriately tune the response on the low and high strings for said application.




As you can see from the picture, I didn't actually anchor the start and end of the coil wires to anything. I probably don't have to tell anybody how big a problem that quickly became. Live and learn, I guess.

I checked the specs on a Duncan Distortion 7-string (which has been my passive of choice for awhile, and also favored by such guitarists as Ola Englund, whose tone is very much akin to the type of thing I'm designing this for) and they clock in at 19.5k DC Resistance, and are reportedly wrapped with 44awg wire. As I have found out, 8,000 winds of 44awg on a 7-string bobbin is a lot more than 19.5k, but we'll get to that later




So, some of the random fumbling and mind-changing I did during this project was figuring out how to fit all the pieces together, and this prototype did not go smoothly in that regard. In my last post, the bobbin is actually upside down; the protruding pole pieces were designed to fit through the baseplate so that I could easily change out the Neodymium disc magnets and see what size was actually appropriate.

The whole process went super unsmooth because I did so much of the drilling without proper templates, and most of my bits wobble really bad.

After some handwork with a round file I got things to line up though.




And probably the most cobbled-together thing of the project (which is saying a lot) is my wax potting rig.

I'm using 100% beeswax because I'm a filthy hippie and avoid petroleum products where possible. I understand that is a less common choice than paraffin or a 50/50 blend, but if I can support beekeepers and avoid non-biodegradable products I'm into that




The pump itself is called a "MityVac" and is for brake-line maintenance. It goes to about 25hg vacuum, which seems to do the job well enough. Better than nothing. I just make it create bubbles and leave it for 10 minutes or so. It never really stops putting out bubbles, but they eventually slow down a lot and I figure if people like Lollar are just dipping the pickups in wax for, like, 30 seconds and calling it good, than I'm probably over-doing it, if anything. Granted, he's not generally talking about high-gain applications, but, again, if it's not perfect, it's surely better than nothing.

So, the obvious question is, after wax potting, how did it all turn out?

And the answer is that it didn't. I didn't get DC reads from one coil, and the other was difficult to get to report anything and wasn't consistent when it did. Now they're both dead and I'm assuming I just broke or shorted the leads somewhere.

Whatever the case is, I got to work on the second iteration pretty quick, and this time I shaped the lower bobbins so they could actually have the wires anchored to eyelets like a typical vintage Strat bobbin.




This iteration went much smoother, but was still lousy with irritations and screw-ups. Not to ruin the reveal, but it does in fact work, so you won't be stuck looking at another dead pickup when this is over

And I realized later that it was dumb to refer to the North and South coils as "Top" and "Bottom" because that caused some confusion later.

Live and learn.

[Antigua]

If you have an LCR meter on hand, it would be cool to know the inductance values you ended up with. Also if you end up thinking the pickup is too dark or too bright, and you know what the inductance is, it might be possible to better guesstimate how many turns need to be added or removed.

[KhzDonut]

If you have an LCR meter on hand...

Unfortunately I do not, but I'd also like to know the inductance, especially when I start screwing with metal base plates. Not something I can swing right at the moment, but soon

[KhzDonut]

Variable Interstice Pickup Prototype Part 3: Redesign

So after things kind of went belly up on the last iteration, I decided to try and make things a little bit easier on myself with this one. I didn't succeed in making many things much easier, but a few things here and there went much more smoothly.

I made the bottom of the bobbin a trapezoidal shape, similar to a traditional Strat pickup but shaped to fit in a slant. That'll give me more room for anchoring wire.






I made several rounds of templates off of previous templates with slightly improved accuracy each time, so things lined up more accurately than last time. It's amazing how hard it is to get super accurate holes drilled into something. You get really used to having a CNC machine around :P






I have definitely found acrylic to be superior to using MDF or another piece of Forbon for making templates, but this was scrap from a pickguard and is too thin to be very durable against the drill bits, so I need to get some thicker acrylic.




And after yet another round of making pole pieces with a hacksaw (at least one arm is getting pretty buff at this point) and using the 1/4" Forstner to widen the bobbin holes after making little index holes (it's marginally accurate doing it that way...) I have what I need to assemble the bobbins.




I made these MDF spacers so gluing the bobbins in place would be easier and more accurate. It worked ok, but as expected, I end up with a little bit of cleanup where the MDF glues to the bobbin. I'll solve that in future iterations.




It's supposed to make for a .500" tall coil, but I think this ended up at like .47 or something like that, because MDF and sanding and clamping and etc...

[epizootics]

Keep going! Prototyping is a lot of work, especially when cutting odd-shaped bits of forbon by hand. Think of it as a necessary pain...And how nice it will be to get those parts laser-cut once you get your design down to a T.

[Jason Rodgers]

This is the type of rabbit hole I'd go down! Thanks for documenting your trials.

This latest flatwork shape is very clever, the way it's one shape but you flip it over to get the angled coil. That'll make the laser or water jet cutting much easier because there's only one part.

[KhzDonut]

Variable Interstice Pickup Prototype Part 4: Better Bobbins

I got a much better result putting the bobbin together this time because of the spacers, though it is still not perfectly square. I didn't have trouble with winding, but part of the reason for that is obsessively beveling the edges of the flatwork to avoid snags. I have so many variables always going on that I really hope to avoid major incident with the winding itself, as I have little experience and correcting mistakes as they happen is still a pretty daunting task.








This time I still had the pole pieces punch out of the back of the baseplate flatwork, but I was able to get a bit better result this time. Hacksawing pole pieces isn't an exact science, but after awhile I guess you get better at it






Comparing the first prototype to the second, I got the size of the coils dialed in a little better. They're just under .5" tall coils, and I was able to get the center of the left-most north-coil's pole-piece just about perfectly centered on the baseplate, so it's a slightly tighter grouping of the coils. 8,000 winds of 43awg on the north coil was almost too much for the bobbin.

I also refined some of the sanding/finishing techniques to get a better overall look.

Obviously, they're still just a weird looking pickup and not really aesthetically desirable, but for prototyping purposes they get the job done

It tested out at 21.5k DC Resistance with 8,000 winds per coil, 43awg North and 44awg South.




Here's the guitar it's going into, which is a pretty mangled Squier Jazzmaster. It's got 5 strings on it because after I rolled the edges of the somewhat uncomfortable square-ish fingerboard, I was getting strings rolling off the sides. My mistake, I just really wasn't being cautious and I honestly wasn't that concerned about it. I had planned on getting a Warmoth neck but haven't gotten around to it yet. With 5 strings, they line up on a 7-string pickup's inner strings so it's useful for testing out passive-route 7-string pickups. (Well, adequate if not good)




Turned out I still didn't have a big enough route in the bridge position because I was getting the pickup hitting the sidewall and angling weird, so... It required a very skillful solution with finesse and accuracy, which is the hallmark of this project *cough*






Lot's of drilling and then lots of smoothing it out with a Dremel. Not a lot of fun, but it worked.

Next time: The results!

[KhzDonut]

Variable Interstice Pickup Prototype Part 5: Conclusion

Kinda forgot I was working on this thread; been editing video working on the next prototype


These are the final specs on the pickup (I misreported the DC Resistance in an earlier post, but can no longer edit it)


20.6k DC Resistance
8,000 winds per coil
43awg on the North Coil
44awg on the South coil
6.35mm/.125" (1/4") 1018 Steel Pole Pieces
12.7mm/ (1/2") Coil Height


My Thoughts:

It's kind of cool, but also kind of garbage because of the high output. It's just too much.

I tried several different sized magnets, and the weakest magnet to supply adequate signal was still very very loud. Below that size, notes just cut out or barely rang. With magnets of adequate size, the output was consistent and it felt a lot like a regular pickup, but I did notice that the pull was significantly stronger than my Seymour Duncan Distortions, and it was the first time I've ever really experienced anything I'd liken to a magnet choking a string, which mostly happened if I was playing in the upper register; the note would die more quickly than I expected, and after a certain threshold, just cut out.

The output was very high, almost clipping the input on my Helix, which the Duncan's don't do at all. The sound wasn't really bad, kind of muddy and lacking highs, but I expected a bit of that. There's just so much.

One thing I've got to take into consideration is that in my mind "scatter wound" meant that making lots of quick criss-crosses with the wire would be a good idea, and I've seen a few posts on the forum talking about getting 50-100 turns per layer, and I was probably get 10-30 most of the time, so I'm sure as I adapt to a more standardized TPL I will see different results in general.

The thing I like about the pickup is that it is very very punchy, and it pushes the low-mid chugging sound that I like. I really prefer darker tones for my high-gain sounds, and my solution to avoiding muddiness comes from low-cuts and generally avoiding playing overly fast on the lowest notes on extended range instruments. High output, but somewhat darker pickups, like the Distortion, do pretty well for that kind of thing, but I'm also a fan of EMG 81's because they also have a lot of low-mid push without being overly bright. Little ice-picky for the frequencies I like to boost, though.

Other guys who tune that low, but also play faster, more defined metal, would probably want a pickup with a much different sound profile than this one, because it just lacks definition for big chords, and is too chunky in the lows for technical work.

I noticed generally extended frequency response to the Duncan, but a very big low-mid hump that really hammered the amplifier, in a way that is only about 60% flattering.


Proof of Concept:

I installed the pickup both ways, with the wider interstice on the bass side, and once on the treble. The sound different was subtle, but a non-guitarist would probably be able to pick it out.

It's definitely more "focused" when the gap between coils is more narrow, which was terribly shrill on the treble side, and tighter sounding on the bass side.

Wider-spaced side had a more open sound either way, and for the bass side had more shimmering highs, but looser, flabbier lows.

When I make a better-sounding pickup, I'll do some A/B comparisons. This didn't seem like a good comparison because of the overall muddiness.


So, what that tells me is there is definitely something to be gained here, but I need to drastically change the numbers of winds, and experiment with backplates, slug sizes, and coil heights to track down the qualities I'm trying to capitalize on, and really exploit the variable coil gap.


I recorded some tests of the pickup, but it's mostly garbage because of the output. The sound it does well is punchy, dark high-gain sounds, so here's a video in which that is what I use it for. It's not at all a demo of how the pickup sounds, it's just the one thing it does well, done for a couple minutes :P


Enjoy!

[Joe Gwinn]

Those magnets are far too large (in cubic inches) for the purpose. This is why the sound volume is too high and the strings are behaving oddly.

I cannot tell from the photo how thick they are, but I'm pretty sure one can buy thinner neo discs. Nor is it necessary that the discs be the full diameter of the poles to which the discs are glued.