Project: Earthstone

[Prostheta]

Well well well. That time of year again, it would seem.

I've been somewhat out of the game for the last year and a half (at least) with a bunch of projects left hanging for one reason or another. As is customary, adding another one to that list makes everything fine. Changes in my workplace the last two years and moving city sort of bumps you off the track a little, that's for certain.

Anyway. This build is a relatively straightforward set-neck Gibson-type build based around the Ibanez Darkstone shape. I've wanted a nice warm guitar worked around African Mahogany or Sapele (both have warmer characteristics, the latter having a bit more twang/bite/snap/whatever) with wooden binding and purfling everywhere. Contary to my #1 guitar which is a thin Mahogany Ibanez Sabre, I want to have some bulk to this one. Really slow the response. Coupling that with some lower-output A2 pickups and a 24-3/4" scale puts this firmly in the region of a bend-tastic guitar with maybe a softer compressed type of response. We'll see what the wood has to say if/when we get past the post.

The blank I've prepared is a single piece of Sapele with mostly vertical grain, showing the typical ribbon-stripe characteristic.

(wetted with alcohol; will explain the saw marks later)

I want that appearing only on the back, so I laminated a two-piece bookmatched 5mm top of what I think is Khaya Anthotheca (typically sold as "African Mahogany") that has a strong Koa-like mottled flaming to the surface. The top was jointed and lightly glued with a PVAc rub joint. This needs no strength other than to stay put. The quality of jointing is more important than glue quantity or clamping strength in this instance. The top started out as two pieces at 12,5mm each. A 1mm bead of PVAc down the length of one edge is all that is required. Any more and there is a potential for glue contaminating the jointed edge or the grain immediately adjacent to it.

Once dry, the bookmatched top was thickness sanded down to 5mm, all coming off the back with a skim pass of 0,1mm off the top.

 

 

The endgrain of the Sapele from what will be the back end of the body shows a cross-section with mostly vertical growth ring structure. The arched area at the left closer to the pith isn't going to be problematic as there is about 12-15mm coming off either side anyway. Movement in that area is unlikely to cause noticeable effects in anything other than the look of the grain. I've seen better, but equally I've seen much worse

The Khaya top was laminated to the top using Phenol-Resorcinol glue at 50C for 45mins. Anecdotally, I believe Gibson used this for bonding LP tops in the 50s. Whilst interesting if it's true, I used it solely on the basis of it having a glassy non-plastic bond that it pretty much bomb proof. We use this for Teak and all sorts of laminates. It'll probably kill ya though! The other option was to use PVAc, however I'm shying away from that where possible. Cleaning up contamination from squeezeout for transparent finishes is a ballache, especially when you don't see the contamination until the last minute or after you've finished. There's very little squeezeout as I used a glue roller to apply an even film rather than throwing it on.

[Prostheta]

The design work was done in Rhino. A few areas like the volute and heel transition are incomplete, but that's hardly an issue....after all, a rendered preview is to get a feel of how an instrument will look in the real world. Unless you're taking it to CNC or looking for approval on a specific detail, the broad strokes are what matter. Similarly, I was considering using some ammonia-fumed white Oak on the top at this stage but decided against it....since it looks too much like Wengé. Just not what I was looking for.

[Prostheta]

Typically, the top, fingerboard and headstock will be bound with black Walnut and feature a 1,5mm WBW purfling line. The fingerboard has already been prepped off the neck (photos forthcoming) and bound. The headstock will likely have an Ebony headplate.

The back of the body blank still has a few saw marks. These are not an issue, as I am intending on cutting a self-grain matching cavity cover. The first step of this is to make the body blank thicker than is actually required (hence 45mm+5mm=50mm). The cavity will be located in the half of the body blank with exclusively vertical grain structure which helps maintain a good match. The process for cutting this is to stick a cavity cover template to the rear of the blank, rout down ~4mm and clear away material around the cover, remove the plate with a flush-cut veneer or dowel saw, then thickness the back of the body down until clear. The cavity plate is then laminated to copper or brass sheet for strength. All good.

We'll talk other tricks a bit further into the game....my template is currently in the press....

[willliam_q]

I’ll look forward to seeing this develop, I really like the drawing and the wood combinations.

[Prostheta]

Thanks man. If my reputation for glacial progress holds, you'll need superhuman patience!

[Prostheta]

So this is the fingerboard, radiused to 16" with a zero fret....as you'd expect from me by now. Also, the headplate which is itself a bookmatch. The dark patch on the board is a bit of oil I tried to see how the Rosewood looks.

[Bizman62]

Mmm-mm... Starting to look delicious.

 

[komodo]

Glacial?! You're almost done! This is going to be a treat to watch Carl.

Gold hardware - bold choice.
"If someone gave me free gold hardware I would turn the torch on it and bake the electroplating off... and if it melted I wouldn't even feel remorse as I saved some poor unfortunate soul from owning a guitar with Gold hardware."  -RAD

[Prostheta]

Yeah, RAD has always been pretty anti-gold, to put it mildly....! I have a gold ABM 2506 tune-o-matic, and @ScottR really sold me on the Gotoh 510 Deltas. Wood and gold tie nicely as "warm".

Whilst I'm at a keyboard, I never explained about the A2 pickups. I've grown accustomed (no pun intended) to the Seymour Duncan Custom Custom, which is an A2 mag. Contrary to the '59 in my #1 which is A5 I think some slightly squishier A2 equivalent like a Pearly Gates in the neck would work nicely. I really dig low output pickups for their dynamics and feel. Output can be boosted if you want a hot signal, and lower output pickups with the right pot values dial back nicely if you like cleaning a tone up using the guitar controls. That's where my head is with that anyway.

[mistermikev]

lovely

[komodo]

4 hours ago, Prostheta said:

I've grown accustomed (no pun intended) to the Seymour Duncan Custom Custom, which is an A2 mag. Contrary to the '59 in my #1 which is A5 I think some slightly squishier A2 equivalent like a Pearly Gates in the neck would work nicely. I really dig low output pickups for their dynamics and feel. Output can be boosted if you want a hot signal, and lower output pickups with the right pot values dial back nicely if you like cleaning a tone up using the guitar controls. That's where my head is with that anyway.

I did that same pickup merry go round with that Dragon guitar. The SD SH-5 (Cer), SH-11 (A2), SH-14 (A5) are all in that ballpark. Not really low output though at 14k? In the end I used a C8 which is a custom SH-5 using an Alnico 8. Interesting, high output, bell-like. It would pair well with a thick mahog body.

[Andyjr1515]

Excellent! He's back and building!

[curtisa]

Welcome back to the fold, Carl. I should probably take a leaf from your book and consider documenting some of my goings-on too

[Prostheta]

Thanks guys. Given my propensity for long timescales, I figured that fully documenting my thought processes would be an interesting approach. Often certain design decisions are driven by necessity or a desire to try something outside of my comfort zone. Here, there's a lot of forming and mitre cutting needed in the Walnut binding. That should be a task.

I've been using Titebond hide glue a lot recently as well. I have to say that I like it, even though it is a bit expensive. 

[Prostheta]

Been off work for a week with tendonitis, which also means I can't easily do work at home either. *~%'/

Anyway. Today I copied the cavity cover template into 25mm MDF in order to cut the grain-matched cavity cover.

Cavity cover template, plus a piece of scrap from the same stock. This is used to balance the opposite edge of the router base rather than leaving it "tippy" over the edge. It makes the process a bit slower (I would argue that this is a good thing) but it protects against mishaps which can happen on small surfaces such as these!

The round mark in the wood is from when I was photographing the blank and protecting the surfaces off my bench using a couple of rolls of tape. That surface had alcohol spritzed on previously....currently the top is completely taped over. Stray bits of waste easily score or gouge your work!

 

The template was double stick taped into place. The surfaces were both cleaned of any dust prior to application. As you probably know, I think the whole "masking tape and superglue trick" is a liability waiting to go wrong, and I've never had this 3M DST fail with good clean flat surfaces. I use what has reliance and is made for the job.

Firstly, I open up the router base's depth stop and plunge the bit to the surface of the workpiece, then lock it off. This is our zero.

 

Between the appropriate turret post and depth stop, I use something that represents the final depth of cut I want. In this instance, 5mm is adequate so I popped in a piece of 5mm template acrylic. The depth stop can then be locked off. This approach is extremely useful for things such as binding or other work where you can use the work material itself to set a depth of cut within a couple of tenths of a mm.

 

Lots of sawdust later, we have a clear area around the template at exactly 5mm depth of cut. Even though 5mm isn't a deep cut, I still did this in light passes. The first establishes a clean cut in the surface, which is liable to tearout at points of delicate unsupported grain. The last passes take it to depth. I'm likely going to need to take out a bit more to make the sawing operation easier, but that's for later.

 

Template removed, fuzzies knocked off with 240 and any tape residue cleaned off with alcohol. The main purpose of the match is to align with the Sapele's ribbon striping, and if possible the growth ring lines on the surface. The former is fairly easy but the second might be a bit of a trick. We'll see.

 

I might do some work on the front prior to cutting the cavity cover off the back. I need some of the machinery at work to reclean the back after doing the sawing, and also some primer for glueing a metal sheet backer to the wood cover....

[killemall8]

Your planning and methods are always so precise!

[Prostheta]

Thanks Luis. I do have the benefit of time on my hands, what with having no reasonable shop space these days! It's often a case that there's more planning time available than actual physical work time....still....

That reminds me....I sort of forgot about that headstock work. Apologies! Let me get back on the horse....

[Prostheta]

Thinking about it, I might cut the initial cavity cover rout prior to knocking the entire back of the blank flat. That way I have solid reference to where the cover came from in the first place rather than trying to regain a match. 5mm through a blank with vertical grain shouldn't need much (if any modification. Hmm. I'll put some more thought into this since it does sort of commit me one way or the other....

[Prostheta]

Okay, I dragged myself away from food, brandy, whiskies and wine and got some work done at my workplace.

Firstly, a bit of material removal. Outline cut within 1-2mm on a bandsaw. Could have done with a narrower blade (15mm I think) but changing is an absolute ballache. All cuts were done top up as scratches on the back (they happen) will be sanded out later.

 

Next, I removed a little more material around the cavity cover with a router riding on the blank with the cutter set to depth. This had to be done with infinite care as any catches will gouge the blank or worse, damage the bit you're trying to recover.

 

A good arm workout later with a Japanese flush cut pull saw (no, I don't know the names of Japanese saws, sorry) freed the cavity cover. This left me free to flush up the back of the body down to an effective 43mm (43,2mm with a hair for finish sanding). All material removed from the back with a paranoia 0,1mm cleanup of the top since I've been handling the part a fair bit.

 

(no, not one pass....)

 

Back at home, after flatting the back of the cavity cover I drew on a few marks that represent the as-designed location of the cavity. The design has enough leeway in that minor adjustment of the cavity around the workpiece itself is not a big deal. The three parallel marks top/right/bottom left show the designed location whereas the outline describes the position that matches the grain. In theory with this being a single-piece body, I could have done the cavity cover work first before fitting the top, then located that in a placement derived from the matched location of the cavity cover. This wasn't a priority given the amount of leeway available. A judgement call based on what keeps things moving forward.

 

Unfortunately at home, my "workshop" lights are LED panels almost overhead, so taking a face-on shot gives me a big ol' shadow. Still, you can see how the grain patterning (more accurately, "growth ring expression") didn't change much in character through 5mm allowing for a very very nice match. Very pleased with this.

 

My next job is to thickness the cavity cover and laminate it onto a piece of brass or copper sheet for strength. Mini neodymium magnets will be set between both. Strength doesn't drop much tested with 1mm of brass. The internal cavity ledge design will be modified to have a low point acting as a fulcrum, so that pushing on one corner of the cover will tilt it up.

I'll rout the outline of the body first....let's see if this Sapele has any hidden tension surprises that kicks out!

[Prostheta]

Incidentally, has anybody else noticed how much this Sapele is bleeding out? I commonly use alcohol to wet down wood (but not oily ones) and I've never had Sapele having its colours mobilise anywhere near as much as I have with this piece....

[Prostheta]

Mega update. Hopefully this will help illustrate my working and thought processes....

So anyway, on with routing the outline. I find that the way I approach this changes dependent on the wood, the amount of material that needs removing and the design requirements. Some woods are less than ideal, the amount of time cutting close to the finished size and any obscure fragile details affect things greatly. Sharp clean cutters, good feed rates and several pass approaches minimise issues in fractious pedantic woods. Routing is a shaping process, not one for removing material by hogging.

My first job was to rout the outline first, rather than the pixie cavity. I fabricated an acrylic offset base for my palm router many moons ago, and this is idea for edge routing. Most importantly it moves the retention point of the router into the workpiece, allowing one hand to keep it flat to the work face and the other to rotate it around. Generally I move the hand over the body very little, instead using the hand that holds the motor to rotate against the retaining pressure. This keeps inwards pressure against the body even and increases control. My first passes are always establishing cuts, made with the minimum depth possible. Grain on the surface of the workpiece is less supported and easier to tear out.

 

I'll be routing from the top down using a 12mm template, which may actually be 15mm....can't remember right now. Still, it makes the first cut reasonable shallow with the bearing set at minimum depth.

 

Most people know that I do not subscribe to the "trick" of using masking tape and superglue. It comes down to what works for you, however for my part I prefer using something that is designed for a purpose and has predictable results. 3M DSTs are thin poly films that adhere stronger with pressure. They can have problems with adhesion to dusty surfaces (the most common failure mode people encounter) and difficult materials such as MDF. It's also a lot cheaper. Four pieces this size hold an entire body template with zero movement or potential for it. Done properly, this is the way. I have spoken.

 

If unsupported grain is an issue - Sapele can be problematic with the grain being crosslinked, therefore "always against" - then I find that routing up into areas where it might be an issue first works better. The opposite way would remove material that would otherwise be supporting shorter grained areas, increasing the risk of blowout. Sapele can be a bit tear-y on shorter grain, so removing as much material to the line prior to any routing improves your chances of gold, silver at worst.

 

The problems are weird changes in direction on endgrain such as "1", where you need to lead in and around corners whilst maintaining inward pressure on the router. The offset base is invaluable in these instances. Really, "2" should have the same approach. The cut is started with the router moving in the cut direction whilst simultaneously being moved into the workpiece. This way, when the cutter first engages into the cut, the tool pressure doesn't pull the router backwards into a climb cut and starts with actual cuts rather than parting the material. Around double the forward motion as the introducing motion is what I aim for. From there, consistent controlled progress with a similar lead out of the cut.

 

Cuts like at "2" are pretty hellish. Without an offset base, the router can get very tippy. One catch and you're toast.

 

This however, is not toast. I decided to take the cut from "1" to "2" through since it wasn't important to not do so....

 

Yes, this all produces craploads of chips and dust. Sapele sticks to everything. Chip management with routers on edge cuts is terrible at best, so I took everything in several stages, rotating the workpiece and setting up a dam on the bench to control most of it. Typically, this ended up being ten seconds of work and a few minutes of cleaning. It's a good habit to form though, as wood dust should be controlled where possible otherwise you know, death and pestilence and all that.

 

As things progressed, dpeth of cut was increased gradually. Differences in bearing/cutter diameter, concentricity, cleanliness of bearings/bearing surfaces, etc. leave step lines, but nothing that can't be sanded out later. Burn marks tend to be worse and require a fair degree of work to sand out compared to these....

 

Eventually I needed to change bit from my ø11mm x 15mm short cutter to a longer ø19mm x 25mm cutter. I load these with an additional bearing to improve cutting flexibility and reduce bearing pressure marks. By this point I'm also removing the template to ride off the cut itself, so I added tape to the router base to reduce marks from the acrylic.

The final few mm were completed from the bottom face using a bottom-bearing cutter. Sorry for not photographing....

 

Beh, Sapele. It's in a high point so it'll sand out. This was the only tearout I got though. It seems that the bearing in my bottom bearing cutter is marginally oversized leading to a marginally shallower cut. That's a good thing in this instance.

 

So, onto the pixie manipulation chamber. I copied my acrylic template into thick (19mm?) MDF, allowing me a shallow cut without a specialised milling cutter....one of those is on my wishlist though....again, four small pieces of DST. MDF tears really easily and doesn't stick strongly, so one has to be a little more careful.

 

As before, sinking the bit through to the work surface, locking off and setting the depth stop with an "example" of the required depth of cut works wonders.

 

Result.

 

....and the acrylic negative fits (albeit with a fraction of a mm kerf from the laser and the internal corner radii).

 

....so let's do something about them.

 

Sever the endgrain with a sharp chisel....

 

....then pare with the grain.

 

The cavity cover isn't entirely flat at this stage, but fits into the recess perfectly. I will need to repair that slight tear in the corner though. In hindsight I could have matched the cover to the routed recess's internal corner radii and knocked it back, but hey. That's a grain-matched cavity cover.

 

Next I need to get the internal shaping done. Again, routing is not hogging. A combination of Forstner and drill bits, plus a bit of hammer and chisel work got rid of most of the material.

 

This was the only problem area. Oooooops. Also, I wouldn't normally rout against an acrylic template on a workpiece like this as copying into plywood, etc. is a far safer approach. In this instance since the internal template fits perfectly, I figured "why not". It worked a treat, however if your bearings are not clean, not running smoothly or your router speeds are too high you'll end up with melty acrylic. Then you're toast.

 

Rounding over that corner with a chisel and sandpaper is an easy fix. I had a few chips that might have been possible to replace, or I could have filled with polyester. After all, this cavity will be dressed with copper foil at some point. The resultant new shaping was then copied to depth.

 

....like this. The cavity is nowhere near full depth, however at this point I'm considering fitting the pots and switch into individual recesses rather than the entire cavity being dropped down to a few mm from the front.

[Prostheta]

Overall we should be aiming for this....

[Prostheta]

So I have a few items to gather here and there for this. Pickupwise, I'm looking to snag a Seymour Duncan SH-11 Custom Custom. The spacing of the ABM 2506G Nashville-style TOM bridge requires a pole spacing of around 51mm which is right on the cusp of whether I need a TB-11 trembucker version or straight Gibson-spaced poles. I think that on the basis I'll be making custom wooden pickup rings, making one size seems less hassle than two. I'll take whichever I can get first at a good price with a gold cover, trembucker or not.

https://abm-guitarpartsshop.com/ABM-GUITAR-PARTS/Tuneomatic-Stoptail/Nashville-Style/ABM-2506/Glocken-Messing/ABM-2506g-Gold::277.html

I like the Custom Custom's A2 mag as it's worked beautifully for me over the years in my #1, paired with an A5 SH-1n in the neck. In this instance, I think I'm going to take that down a notch to an A2....either a Pearly Gates or an APH-II. Either way, A2. In my experience, lower wound A2s made to a "less modern" spec are more responsive to touch and the interaction with the wood of the instrument itself.

To go with the ABM 2506G, I'll need to pick up an ABM 3020G stoptail. Both are made of bell brass, which as most know, is a semi-psychotic fetish of mine. Screw Zamak and other zinc alloys. Aluminium, meh. It's a costly item, pretty much the price of a pickup or a reasonable TOM/stoptail set on its own.

Tuners....I'd like to say Gotoh SGL510Z with black plastic tulip buttons. Chances are I'll have to take whatever buttons come with the tuners because of price. I'm sure I can retrofit some Ebony buttons at a later point.

I'm unsure as to what value of pots to use with the A2s. I'll certainly be wiring a straight vol/tone with the tone control acting on the neck pickup only. Basically this, but with the tone pot tap wire reconnected to the neck hot so that it only becomes effective whenever the neck pickup is in circuit:

 

I may need to experiment with the pot values so that the bridge isn't too bright when the volume is fully open, then balance off the tone pot/cap to allow the range I want from that. If 500k works for the bridge, I'll likely use 500k for the tone pot with a 220-330nF cap. 470nF is probably a bit too deep of a low pass filter for a less than stellar bright neck pickup.

[komodo]

I could probably live with a guitar that had strings too close together, high action, weak pickups and a loose jack . . .as long as it had a grain matched cover. Did you just drop a Magic the Gathering reference in there? - pixie manipulation chamber

[Prostheta]

Okay, let's call it a day. I found an ugly scratch on the front and remembered somebody mentioning steaming them out here. It isn't 100% necessary in this case, but it was an opportunity.

I fired up the 25W soldering iron and soaked a sheet of kitchen paper. Not dripping, but wet. Enough so that squeezing it produces free water. Pinching off a small roll of paper, squished up maybe 1/8" x 1", paper is applied along the scratch. Press the paper so that the water wets the wood. The soldering iron is held onto the paper for a few seconds at each point to produce steam. Enough so that the steam heats the water in the wood but not so much that the paper dries or scorches, and certainly not the wood. Repeat this along the scratch. Wood takes a few minutes to swell back out, but as long as the fibres haven't been cut or damaged and only crushed/displaced, they should expand back out.

 

I very quickly ran over the surface with 400 grit to remove fuzzies. If you try and knock back recently swelled wood, it will sink back in when dry, leaving a permanent divot where you knocked back fibres that should otherwise be the flat surface.

 

I couldn't help but give it a quick spray of DNA to identify any other areas needing work. As it evaporates, scratches and dings will be visible as pools of wet alcohol.