Voicing the Top - Intuition and Technology

Shaving the soundboard braces to bring out the guitar's tonal potential is one of the more intuitive and mysterious aspects of guitar construction.  For me, it's also one of the more tactile and enjoyable stages when the use of hand tools is mandatory and the senses come strongly into play.  A cycle of tapping, flexing and brace shaving is involved, with a somewhat nebulous goal in mind despite having read everything I can on the subject!

Unfortunately, my hobby builder status dictates that it's often many months after voicing a top then going on to complete and evaluate the instrument before I'm able to repeat the exercise - hardly an ideal situation in terms of being able to build any sort of muscle memory or draw meaningful conclusions. The pertinent questions whose answers seem elusive in terms of being able to adequately articulate them are:

• What did the top feel like as I flexed it, i.e., how much resistance was there when bending it across the grain and with the grain?
• How did the top ring as I tapped it and how was the response influenced by the way the top was held or suspended and where it was tapped?
• What was the correlation between those very subjective assessments and the success of the completed instrument from a tonal perspective?  
• To what extent did the materials, density and stiffness of the guitar's other components combine with those of the top and contribute to the outcome?

As I review that list of questions, it strikes me that the challenges they pose seem almost insurmountable.  It's obvious that relying on my memory from one instrument to the next isn't a satisfactory approach given the subjective nature of the assessments involved and how infrequently I'm able to repeat this part of the process.

Perhaps the use of technology offers a way forward and can lend a degree of objectivity to what would otherwise be limited to some vague, transient sensory experience. Measuring top deflection both parallel to and across the grain prior to gluing the braces, photographing the top bracing and noting brace heights, recording tap tones with a decent microphone (or perhaps one of the hand-held Zoom recorders) or even videoing myself as I hold the top and tap it seem likely candidates as I seek to minimise the guesswork involved in the voicing process and develop a baseline for subsequent instruments.  Or perhaps I should learn to trust my senses and develop a greater appreciation for the delightful uncertainties inherent in wooden instrument construction!

Cheers
Pete

Glue Clean Up

With the advent of the soundport, the standard of workmanship inside the guitar body is more easily scrutinized and there's perhaps an even greater incentive to clean up surplus glue as the various component parts are assembled - not that I've neglected this previously. If some vintage instruments can be taken as a guide, it seems this wasn't always a priority, even for well-respected manufacturers. Times change though, and as I inspected a new Martin in a local music shop not long ago, I must say I was impressed by the meticulous attention to detail within the guitar body including the fact that all traces of excess glue had been carefully removed from the intersection of the back plate with the linings and back braces. I vowed then and there to emulate this in my own instruments to the best of my ability in recognition of the fact that it's often the little things that make a big impression.

There are several approaches to cleaning up excess glue once clamping pressure is applied to a freshly-glued joint. Regardless of the method used, I always try to plan ahead and give some thought to the arrangement of clamps to ensure unhindered access to glue squeeze-out after they're in place. I'm also careful not to apply an excessive amount of glue in the first place. A 1-inch foam roller is helpful in that regard, with the added advantage that an even coverage of glue is more quickly and easily achieved.

I know that the preferred approach for many builders, whichever glue is used, is to leave glue squeeze-out undisturbed, letting it partially set or gel before attempting to remove the excess. I use Titebond for the majority of gluing tasks, and find that a flattened plastic drinking straw with the end cut at something like a 45 degree angle is a good first step towards removing unwanted glue as soon as clamps are in place. This approach works particularly well when gluing braces to the soundboard or back. The straw is pushed cut-end-first along the freshly-glued joint and much of the surplus glue finds its way inside the straw which is then discarded. It helps of course to have several pre-prepared sections of drinking straw on hand.  Chisel-shaped slivers of spruce or cedar, moistened Q-Tips (cotton buds or ear buds) or a moistened sponge can also be useful as a follow-up and, using a combination of these methods, it's usually possible to remove every last vestige of excess glue without compromising the joint itself.

When gluing braces to the back and soundboard, I exercise patience and generally glue one - or perhaps two - braces at a time. Not only does this slow and steady approach give me ample time to clean up after myself, but I have the luxury of a clear view and unobstructed access to squeeze-out.

Another strategy which contributes to a clean look inside the guitar body is to attach the rim assembly to the back first, rather than to the top. I find it helpful to have a second chance to tidy things up before the top is attached and while the various nooks and crannies which will remain visible after final assembly are still easily accessible.

Cheers

Pete

Two Rosettes

The finished rosettes:

I'm pretty happy with them both, and they provide an attractive alternative to my more usual abalone rosettes.

Click on the image for a better view.

Cheers
Pete

A New Rosette

This rosette is a departure from the abalone versions I've recently installed. It's a simple radial design in Macassar ebony which I hope will work well in the visual sense; the final verdict will have to wait until I can marry it with the guitar's other decorative elements and see it as part of an overall design.

Dimensions marked on 1/8" MDF

Sliced and diced: ready for tapering

Tapering the wedges on the disk sander

Wedges tapered and glued

Routed to size - a spiral down-cut router bit keeps things tidy

Ready for installation in a redwood top

I have two of these rosettes ready to install along with their edge purflings. I'll post pictures when they're done.

Cheers

Pete

Lessons in Finishing

Having completed only a handful of instruments with sprayed finishes, it's not terribly difficult to identify flaws and aspects of the process I've yet to come to terms with. Having said that, it helps to keep things in perspective and remind myself occasionally that the perfect finish - if there is such a thing - is an impossible dream for us mere mortals, a fact apparently acknowledged by many U.S.-based builders who happily farm out the finishing of their instruments to a professional. I'm sure the health and safety aspects of working with solvent-based materials play a significant part in that decision too. Down here at the bottom of the world, however, I'm left with the task of identifying problems with my technique, improving my finishing processes, finding materials that suit my situation and hobbyist status and developing the most effective means to combine and apply them. Although it's been a laborious and sometimes frustrating process, I'm cautiously optimistic about my next guitar - but then I always am!

I've made some mental notes of late where finishing is concerned:

• Do a thorough job when pore-filling. At this point in time I'm using epoxy, as do many others. Aside from its effectiveness in filling the pores, I like the way it enhances the figure and colour of the wood it's applied to. Briefly described, pore-filling consists of applying the epoxy to the wood with an old credit card or spatula, then removing as much as possible from the wood's surface, leaving epoxy only in the pores. It often takes several sessions, allowing the epoxy to cure and sanding off dust nibs and ridges before the next application, before all visible pores are filled, but it's worth the extra effort given all the hard work that's preceded it.
• Wipe down cured epoxy thoroughly. A by-product produced as epoxy cures is something called "amine blush", a waxy residue that can have an adverse effect on subsequent coatings. Thankfully, a good rub down with a damp cloth, or one dampened with a mix of denatured alcohol and water is sufficient to remove it. From what I've read, it's worth spending a little extra on a good quality epoxy and applying it in a warm, low-humidity environment to reduce the impact of blushing. The manufacturer of the boat-building epoxy I use contends that their product is not prone to amine blush, but I tend to give it a precautionary wipe down anyway as added insurance.
• Find a compatible sealer. Oily species such as rosewood can be particularly problematic and can adversely affect or even prevent curing of top coats. It's worth experimenting with sealers or barrier coats to find one compatible with epoxy and your chosen top coat material. While some builders sand the epoxy pore fill back to the wood, in which case an effective sealer becomes particularly vital when using any of the oily wood species, I choose to leave a thin layer of epoxy on the wood - excluding the soundboard. The epoxy serves to lock in any oils present in the substrate and the task remaining for me has been to find a effective product to act as a tie-coat between the epoxy and the top coats of KTM-SV. Adhesion of my top coats has been a recent concern, but hopefully I've identified a compatible set of finishing products as I described in an earlier post.
• Find a suitable top-coat. Having set myself the long-term goal of selling my instruments, I've pondered the various features that a prospective purchaser might commonly be looking for. A durable, high-gloss finish is one them, and is one aspect of instrument building that continues to test me, particularly as I've vowed to avoid solvent-based materials such as the much-used but far more hazardous nitrocellulose lacquer.
• Drop-fill early. Any small gaps in the bindings are evident once the first top coats are applied. It's worthwhile dealing with gaps and pin holes while there are plenty of remaining top coats to even out these repairs, the caution being that with only a couple of coats of finish in place, extra care needs to be taken so as not to sand through to the wood when levelling.
• Level sand with the finest grade of abrasive paper which will get the job done, i.e., top-coats are necessarily thin and remnant scratches from coarser grades of paper are difficult to fill. With practice, sprayed finishes shouldn't need much in the way of leveling and, as I've become more comfortable with the settings on my spray gun, I've not found it necessary to level sand with anything coarser than 600-grit.
• Final level sanding of the cured top coats can't be rushed, i.e., don't be in too much of a hurry to crank up the buffer! Patience and a generous helping of elbow grease are pre-requisites as the instrument is wet-sanded in readiness for final buffing.
• Be careful when sanding the edges of the guitar, and especially so around the headstock, where it's easy to "roll-over" an edge with the sanding block and remove more finish than is necessary or, heaven forbid, sand through to the wood. I haven't sanded through yet, but I'm keenly aware of the potential to do so.

Cheers

Pete

Continuous Improvement - a Hard Road

I tend to be hyper-critical of my own work, so it's gratifying to be able to identify at least some aspects of my instruments which I have difficulty finding fault with. Perhaps after 20 years of trying I've simply had enough practice, or perhaps as I've slowly accumulated a decent collection of tools and equipment, developed a range of useful jigs and constantly reassessed and refined my processes, a satisfactory outcome is now more likely.

Now that I've finished congratulating myself, I might add that there's plenty of scope for improvement in several important areas including fretting, setup and finishing. Even more significantly, having any real control over the tone of the finished instrument - except in the broadest sense - seems the most formidable challenge and is one reason I'd be reluctant to take on commissioned work should I ever decide I'm ready to sell my instruments.

An important part of the process of continuous improvement is an dispassionate self-appraisal of my work, involving a comparison of actual results with the work of others and the benchmarks I've set for myself. If my own personal vision of perfection is a valid goal - as it almost certainly should be - it helps to be mindful of the old conundrum involving travel: if we manage only to halve the remainder of the journey each day, we'll approach but never quite reach our target destination. It's a useful metaphor to reflect on when striving towards mastery of any skill; the parallel where my guitar building efforts are concerned is that rapid progress seemed far easier to achieve as I began this adventure, but as time has passed, improvements in the quality of my work seem less significant and more infrequent and difficult to achieve. I do wonder at times whether it's the challenges I constantly confront that form an important part of lutherie's appeal.

While I'm certainly in a position to cast a critical eye over the standard of my own workmanship with a reasonable degree of objectivity, my own limited playing skills will stand in the way of any useful self-critique where tone and playability are concerned. Bearing in mind that it's the views of the guitarists I build for that should be of primary concern, I really do need to seek constructive criticism from skilled players if I'm to make progress with those all-important aspects of the craft. Soon, perhaps, I'll hunt down a willing local participant with that goal in mind.

Cheers

Pete

Grafted Coatings' KTM-SV - A Sticky Problem

With two guitar necks and a body almost ready for finishing, it's an ideal opportunity to refine my spray finishing techniques as well as experiment with a new seal coat material.

As I've reported previously, I'm pretty happy with Grafted Coatings' KTM-SV, the finish material I've used on my last two guitars, however, as I removed the finish in readiness for gluing the bridge in place on both guitars, I was a little concerned by the poor adhesion of the top-coats to the shellac beneath it. Shellac is well known as a material that readily sticks to almost anything, however, it seems the same can't always be said for the adhesive properties of other coatings when applied to shellac. Most notably perhaps, shellac adheres well to epoxy, but epoxy cannot be used on top of shellac. Go figure.

A recent thread on the Official Luthier's Forum discussed this issue and attracted some interesting responses. It's comforting in a way to read that others have had similar experiences and are also looking for solutions. It's also pleasing to see that Grafted Coatings themselves are taking a keen interest and have volunteered to prepare some test panels with a view to recommending a suitable sealer for their KTM-SV product.

I'm looking forward to Grafted Coatings presenting their findings, but in the meantime, I'm keen to try a two-pack polyurethane barrier coat manufactured by Ilva which luthier Randy Muth uses and recommends. I'm pleasantly surprised to have discovered a local source for Ilva products and relieved that the distributor - who commonly deals in commercial quantities - could be persuaded to decant a couple of litres each of the resin, hardener and thinners for me.

If I have any qualms, it's that I'll be reverting to use of a solvent-based product when it was the safe, environmentally-friendly aspects of KTM-SV that attracted my attention in the first place. Thankfully, use of this product is likely to be limited to two light coats, and with the usual precautions of good ventilation and a twin-filter respirator with vapor cartridges, I'm prepared to give it a trial run.

Cheers

Pete