Prostheta

Cheers! The finishing resin worked out like a charm. I sanded off the first layer from the back and sides through to the bare wood, and the surface is already almost glass smooth. Amazing. The resin does sand back reasonably well, but scraping seems to be the best way of spotting and shaving back high edges. Being (what I think is) epoxy, I opted to use Abranet to eliminate clogging and in three grades; 120, 180 and 240. A small plywood sanding block (maybe 30mm x 90mm, 18mm thick) is nimble enough to get over the convex edges and small enough to treat the flat areas around the heel. A couple of rubber drums from my spindle sander works excellently in the concave areas. All in all, the back and sides took maybe 90mins to go back to bare smooth wood at 240 grit. Sprayed with DNA (Sinol) just before I wiped it down to remove dust.... Pretty smooth, but a few small scratches to be removed later. For the rear contour, I'd made a handle with a radiused edge matching the contour minus 5mm for a plywood plate and sandpaper:

I'm testing a product that is new to me to pore filling; Z-Poxy Finishing Resin (PT-40). This is widely used by acoustic makers for the backs and sides prior to the application of a clearcoat, so I think it should be great as the basis for producing a flat surface ready for solid colours. 8g of resin/hardener made its way nicely around the back and sides of the Invaders body, so I'd estimate that ~14g should be good for an entire body. I've not yet done the top processes yet, so I'm using the "downtime" to pick up on a few jobs whilst I get opportunity to sort that all out. The principle of using finishing resin as a grain filler is that it's stable and relatively easy to sand. It does smell like piss though. I worked the resin around the body using a 2" rubber squeegee until the surfaces were sufficiently wetted. I anticipate that it might take a couple of sessions around the body to fully fill all the pores. One this first layer is dry, it will be completely sanded and scraped back to bare wood, leaving only the pores filled. It seems to be working nicely at this stage, however the proof will be shown once it's sanded back and the second layer completed. I can see some popping of air from the pores so it's clearly not going to be a single stage job. I also found that the store I bought this from (Hobbylinna, or "hobby castle") also sell water-thin ZAP CA. Sweeeeet.

I'll go in and do that for him just in case he doesn't get notifications. AIA, etc.

It used to be a third party mapping app many years ago. It no longer exists.

All cool, and yeah, it's a weird time and finding some good way through it all is a difficult task for many.

Making chips and dust resurfacing the heel contour of epoxy. Ferrule surrounds milled out to 14,2mm diameter. Not enough shellac to prevent epoxy ingress, but hey ho. This will be painted over anyway. A handful of old standard 14x5 ferrules sit nicely. The depths of each pocket were selected so that the ferrules are only as deep as they need to be without protruding. Only the inner surface of these ferrules are a little marked up, so they're still eminently usable.

Thanks Andy. I think I'm only pushing my own envelope really. Most of what I've done is to backwards engineer an instrument into the most likely methods that were used to generate its design in the first place. This is the complete opposite of @ScottR's work where he pulls a guitar out of the wood. Manufacturing and process design is something that appeals to a certain part of me, so examining the steps and divining the decisions of the original designers is almost like being a historian of sorts. Japanese manufacture and the methodology behind it is fascinating, and I really should spend more time reading about W. Edwards Deming, TQM, the "Toyota Way", etc. That absolute focus on product excellence through reasoned process design top to bottom is amazing to me. This all started when I first explored Aria Pro II a decade back. I came across more questions than I did answers, so I sought to broaden that understanding and ended up rediscovering things like how special my old Mirage was. That's when the regret hit a little harder. In 2000-2001 I was discussing a trade with Devin Townsend to swap that Mirage with one of his ESP Explorers which (I believe) was in turn one of James Hetfield's at some point. That never worked out, Devin started playing Strats (Fizzy to Terria era) and I think that EXP went to Beav. Yeah, there's a lot of story behind that old Mirage and there's a lot I would do or pay to see it back. I'm sure that my version will be a pale reflection, however that's just one of the things that drives me to progress, develop and learn. Everybody has their motivators.

Funny, because one of the things I try and keep reminding myself is a quote that is constantly re-told in different forms from different sources. "He who stops being better stops being good", Oliver Cromwell "Once you stop learning you start dying", Albert Einstein (probably, probably not) At some point the fatuously-obvious gets elevated into some snappy pseudo-intellectual motivational poster or meme. Still, the base truth is there. I prefer Nick Offerman's dryness anyway. "Always maintain the attitude of a student. When a person thinks they have finished learning, that is when bitterness and disappointment set in" The last thing I want to do is to sit on my arse and be happy with what I've done, thinking it to be complete and absolute.

Thanks Mike. I think that modelling in 3D often leads towards treating the end product as a single arbitrarily-shaped object rather than a discrete set of milling operations with context. The work I do in Rhino is specifically geared towards developed a specific set of drilling, trimming and pocketing operations. This project has given me chance to practice producing arbitrary working planes, for example the rear contour. I derived that as being a basic cylindrical section in line with the Y axis, rotated around X slightly. Simple tasks and challenges such as this are helping to build my vocabulary of CNC tasks that are more complex in result but simple and defensible in the desktop product end of things. Adding tabs would be possible with what I do also, however I've never really gone down that road. The lever holddowns work exceptionally well, and having the workpiece mounted within a specifically-reference auxiliary bed reduces vibration very well. The machine is more than capable of doing what I am expecting of it, and in a lot of ways it is like an athlete practicing away from the competition. This all adds up to making my skills that my employer(s) can leverage that much higher. I've never liked the idea of taking a ready-cooked model into the CNC programming software as a STEP model or similar. Even though there is good geometry to work from, it feels like a lot of context gets side-stepped. If there's one thing that successful CNCing is built on, and that's defensible, built from the ground up, geometrically/mathematically consistent programming. Everything less than that quickly becomes akin to a poor grainy Xerox/photocopy. Yep, I definitely agree about how one builds a personal working strategy. I'm still balancing off best practice versus that, and it's a task sometimes since this is entirely self-taught. I do thoroughly enjoy it though, because it's all 100% achievement based on what I put in. Those are the best ventures in learning.

That's very true. Wood is just as likely to splinter, tear out, burn or have a poor finish on a CNC as it is with manual machining tools. It sits as a part of a larger process, and hence affects how other work needs to be done. It's very easy to cause more work or make other work harder by not giving it full consideration during CNC machining.

As mentioned, I tend to get very caught up in ideas and have to remind myself how the hell I got here. Well, this time it's not so convoluted. The principle design was to develop a working method for producing a carved top instrument, and to a degree this is pushing away from that, but not unmanageably so. Financially the hardware on one guitar is a high hit (over €300 on the pickups and tremolo alone) however by comparison to the thousands one pays for a high value production or a name custom, it's better. Being a working CNC operator and programmer with potential capacity to hit the highest levels of production quality comparable to top-end guitar houses, I shouldn't be aiming lower. You don't just dip your toes in, right? The way this clockspring unwinds is to produce the Invaders guitar (at the very least) and then develop the carved top programming on what has been built via that journey. This of itself has consequences towards what I was hoping to be the last build of this year, the Aria Pro II PE-type guitar. Mainly my thoughts have been centred around pushing myself up through the 5-axis CNC learning curve, which I think (without arrogance) I've managed to do exceptionally-well in 12mths. As far as carved tops go, I am still leaning towards the simpler "contour level map" approach rather than simultaneous 5-axis work. Mostly this is down to it being a more efficient process timewise, plus the end product still needs significant hand-sanding work regardless. Additionally, I feel that a contour map is easier to gauge when sanding than ball-nose endmill tracks. The rear contour of the first body shows significant compression marks along the path of the endmill where the absolute rotational centre does no useful work. That's a lot of sanding work, and in the context of a constantly-varying carved top contour, it could lead to inconsistency through over-concentration of required work in one area compared to another, especially endgrain. A cleaner contour map in the order of 0,25mm - 0,5mm per level sounds overkill, however being done as a pure 2,5D process using a simple endmill produces almost zero compression artifacting. I do realise that a lot of these issues are rarely considered or discussed amongst builders, and there's high possibility that they're not even as considerable an issue as I describe them (@MiKro?) however when wearing my manufacturing/optimising hat, ideas that are tabled and explored always help rattle out potential gains in efficiency, cost-effectiveness, reduction of labour, base quality off the machine or simple pragmatic rethinks. They matter. Future explorations will look at better workholding (possibly with auxiliary vacuum hookup) and 3D profiling processes. I'd like to see how far I can drive precision, such as adding 2mm corner radii around non-flat surfaces, etc. The binding around the forearm contour in the Invaders instrument is an excellent step towards that goal. Hopefully this train of consciousness text wall is meaningful....!

A small situational update to bring things together since I tend to get scatty at best; making without objective needs to be repeated consistently.... So this first body that has been going into the CNC processes is one of three Sapele blanks I made up. One has pretty nice figure presentation, however I want to keep that as the third option as I never tabled more than one guitar, never mind two or three! The current one is going to be the Invaders body, hence the glow rings around the ferrules. The programming for milling the back of the body is what I would consider mostly complete. I changed the electronics cavity processes slightly to increase the thickness of the thinnest part from 3mm to 5mm and added in one additional process to create a recess purely for the switch which benefits from thinner exposure through the body. The neck for the Invaders guitar is ready to be drilled for the locking nut (on the way) and the neck profile. Since there will be some delay on acquiring the black Gotoh GE1996T tremolo unit and pickups (global supply chain issues) this gives me plenty of time to concentrate on getting paint done, plus making a neck carving jig that can work with necks that have a fingerboard in place. The second body is the most plain Sapele blank with vertical growth ring alignment that has a small knot disturbance in the grain presenting sideways across the front. This led to a small bit of tearout, but is no big deal in the slightest since this was always a paint grade blank from the outset. This one will be made to the same specifications as the Invaders guitar but painted Porsche Pearl White, bound cream with side block pearl markers. This neck will be made from the Maple/Wengé laminate blank. Pickups are likely going to be passive with the possibility of me incorporating the 18v differential preamp I have been making for years, maybe with an active LPF varitone in place of the tone control. In principle, the programming for this body should be identical to that of the Invaders guitar. If anything rattles out that I'm not happy with in that, this will be improved upon here. The third body isn't the most spectacular figured Sapele ever, but does have some nice broad waving/clouding that should look nice under a clear finish. I might do something minor to accentuate the figuring and/or the colour such as a mild dye or oiling to increase contrast but nothing major. This one is a little more of a wild card or even a backup if anything goes tits up. By this point I'm three sets of hardware and electronics in, which is costly to say the least. Two ideas I've had for this one were to go fixed bridge, either a Hipshot multiscale bridge or a TOM/string-through arrangement. Since this isn't a priority in any way, that will be left floating until I've moved the Invaders and pearl white guitars along.

Probably not, but who knows? I usually crank up the reflectivity as a way of examining how light interacts with surfaces to help gauge how the virtual object will feel as a real one. It goes 75% of the way, but at least help you align yourself in the right direction. That heel feels very nice in the hand. Very pleased with that in spite of it being a purely geometric shape as opposed to something more organic.

I forgot to mention that I rubbed a bit of shellac around the recesses, because why not.

To be honest, I don't put much stock in truss rods these days. There's a lot of cheap imports with shady welds that rust really quickly. I'm partial to Gotoh aluminium channel rods or making my own. I just wish I could turn a better thread on stainless because I just fail. Give the rod a stress test in the vise just to make sure, clean it off and yep, lube 'er up.

This.

I can't say that I've ever had a problem with PVAc glue locking up a truss rod. Not with the amounts that a proper application involves. Epoxy, definitely. That's trouble. Generally it's a good idea to lubricate the rod before encasing it, taking care not to introduce contamination to the glueing areas of course. That of itself should be enough to prevent any errant glue from causing any problems.

I absolutely love how this heel contour plays out. It's a mild 10 degree plane with the transition being in line with the rear of the lower cutout. A simple 10mm radius endmill follows the bass side edge of this. I think this would be tricky to do manually, but once you've got a simple angled routing jig in place for the plane itself the edge treatment is far simpler. . Also, some clumsy asshole blew Sapele dust over the wet epoxy. It'll sit on the surface, but still. What an asshole, eh?

So where are we today? Well, I tweaked the rear contour programming a bit. Nobody mentioned the oversize neck ferrule pockets Contour and Gotoh dual battery box routing. Neck ferrule locations dammed off for an epoxy ring pour.

I updated my workholding for the CNC. Scrap plywood, M6 allthread, M6 threaded inserts, leather grip pads and cheap plastic knobs. One will be removed for when the jack socket hole is milled through the edge. ....and this is what the CNC programming looks like....

Not that one! Spede for other people's context:

Nah, it's from a bunch of cheap wooden picnic forks My plastic cocktail sticks aren't calibrated either!

Freshly bound. Hopefully the inlays pop a bit better once I can get sanding the board and cleaning it up with alcohol. At this stage I think it'll burn the surface.

Absolutely not. It would be a fine entry!

Usually 360-400 if I'm using Abralon. 320 with paper. I do a lot of wetting with alcohol to reveal scratches from lower grits because oil will make them pop out.