Spirit varnishes consist simply of one or more resins dissolved in a solvent, which is typically some form of alcohol. These varnishes dry mainly via evaporation, and once the solvent is gone the resins regain their original nature. As such, spirit varnishes are typically more brittle and have less water resistance than an equivalent film of oil varnish. However, they tend to dry very quickly, have a brilliant and sparkling gloss, and can form a hard and glassy surface which is often desirable for high-quality decorative work. The most common spirit varnish is simply a solution of shellac in ethanol, and I have used it extensively in the past; however, it has an unfortunate combination of brushing and drying characteristics which make it challenging to achieve a fine surface finish. First, the liquid varnish is both low-viscosity and slightly thixotropic, leading it to constantly form runs and ridges unless extreme care is taken with the brushwork. This is compounded by rapid surface-drying, which turns these defects into pockets of wet varnish, such that any attempts at retouching only make the problem worse when the pocket is torn open by the brush. Finally these pockets, which do not level out and cannot be brushed out, remain soft (often indefinitely) and can therefore not be sanded out either, without ruining the full thickness of the varnish film. The only remaining options are to accept a relatively crude finish, or to turn to french polishing, a technique likely invented to counteract the faults of shellac.

   Alternatively, the properties of shellac can be modified by the addition of other resins. While shellac is tough and elastic (at least compared to most resins), these properties alone make a poor varnish, similar to making an oil varnish with oil alone. Instead I have found that if (as in oil varnishes) a hard resin is mixed with shellac (taking the role of an oil) at 25-50% of the total resin composition, the resulting varnish dries more slowly, becomes self-leveling, and is capable of being retouched without tearing. Even more surprisingly, after experimenting with a wide variety of exotic resins, I have found that ordinary pine rosin produces the best results. Rosin and shellac in a 1:1 ratio produce a slow-drying varnish with good brushability but poor durability, while a 1:3 ratio produces the exact opposite characteristics, though still more brushable than pure shellac. Accordingly, a 1:2 ratio of rosin to shellac achieves a balance of these properties. Specifically, I have found that a good general-purpose spirit varnish can be made from one part rosin, two parts shellac, and six parts 190-proof (95%) ethanol, by weight.

   This varnish can be further modified and made flexible by the addition of castor oil, which can be added as a plasticizer at up to 10% of the weight of the shellac component. This limit was identified by making 25% solutions of either castor oil or shellac in ethanol, mixing them in various proportions, and brushing the mixtures on microscope slides to dry. Compositions containing over 10% oil became hazy and rubbery, indicating partial insolubility of the oil component. This is especially important to consider when adding castor oil to varnish containing rosin, since this incompatible excess oil combines with the rosin instead, making the varnish permanently tacky. Castor oil can also be used as a pigment carrier for tints, though the pigments used should be light enough to remain suspended in the relatively thin varnish, with dye pigments and lampblack being ideal. These tints can either be ground directly into the castor oil (as in making oil paint), or more crudely, tubed linseed oil paint can be mixed with castor oil to enable it to dissolve in alcohol, though this somewhat weakens the varnish film. This, as well as the possibility of using different grades of shellac, allows for varnishes to be made in a wide variety of colors; these can be seen at the top of the page, applied in seven coats to pieces of radiata pine, followed by polishing with a few drops of linseed oil. The resins used in this varnish can also be seen below, alongside the minimal equipment needed to apply it. Note that fairly fine-bristled brushes should be used, to minimize the layer thickness and the need for self-leveling. Also note that these brushes can be cleaned obviously with alcohol, but also more cheaply with ammonia, which saponifies the resins and oil, allowing the brush to be rinsed with water.

   Curiously, it may have been noticed that the ingredients for this varnish, and the ratios thereof, are identical to those used to make rosin-shellac glue. In fact, this varnish can actually be made by melting all of the ingredients together, breaking the cooled mixture into chips, and dissolving these chips in twice their weight of ethanol. This is unnecessary under normal circumstances, but I have found that by this method it is possible to revive long-expired shellac flakes which would otherwise be completely insoluble. This also allows for a preview of the final varnish film characteristics, assuming the varnish is applied in thin coats and given enough time to dry completely. This is important, as often the varnish will seem far too soft when freshly applied. Although fresh coats of varnish can be applied every few hours, the resulting varnish film typically takes a few days to fully harden, though it can usually be handled gently (and even be wet-sanded) after drying overnight.

   Interestingly, this extended hardening process seems to involve some oxidation of the rosin, as colorless varnish will yellow somewhat during this stage of drying, and the melting point of the varnish increases such that it no longer softens when left in direct summer sunlight. For this reason, hardened rosin is somewhat unnecessary, though it may be used for its color. In that case it should be made without any iron catalyst however, since this would darken any tannins in the wood due to their mutual solubility in ethanol. Likewise, other hard resins (such as copal or sandarac) can be used to either partially or totally replace rosin; however, I have found little benefit to their use. Soft resins (such as mastic or elemi) should be avoided entirely however, as their softness is derived from essential oils which are insoluble in shellac. This weakens the varnish film in a similar manner to linseed oil, but also ultimately destroys it (causing cracking) when these volatile oils evaporate over the course of years or decades.

   Overall, I find it highly amusing that the simplest ingredients produce the best varnish, which I believe is the best possible outcome in the context of this website. I look forward to using this varnish in future projects, particularly on decorative electronic devices such as my earlier carbon microphone, where a high gloss on wood should pair well with a high polish on metal, and where its poor water resistance should hopefully not be an issue. It may also find use in more practical scenarios, such as on tables or floors, where its surface hardness might give it an advantage over oil varnish in terms of durability and cleanliness, and where its low cost, simple preparation, and lack of harmful fumes would facilitate varnishing large areas. In any case, this varnish now forms a key component of my range of coatings for wood, and also serves as a simple introduction to varnish-making for other amateur experimenters who may be interested in the subject.