hide glue

The Starting Point

While undertaking some recent reorganization of my basement workshop in Elderbarndottir’s former house (she got married in November and moved to her husband’s house) I came across this nostalgic and beat-up picture from early 1977, hiding behind one of the shelving units.  This French secretaire was my first really high-profile/high-value furniture restoration project at Schindler & Son of the Palm Beaches, where I started working in late 1974.  Unfortunately I did not take any detailed pictures of projects at this point of my career – had this happened once I was a museum conservator the project would have been documented with hundreds of photographic images to go along with the written reports.  I probably took a picture or two of the interior, it was spectacular.  Bat that was almost 50years ago and cannot recall that detail.

Somewhere I have a picture of a Riesener cabinet from the same client, but that picture has not turned up yet.

Prior to this I did a lot of run-of-the-mill restoration for “ordinary” antiques along with a boatload of custom finishing and refinishing; before Schindler’s I was a “scratch and dent man” at a couple of furniture stores.

This project arrived in pieces in the back of Ambassador So-and-so’s Mercedes station wagon.  The secretaire bore the inventory stamp of the Chateau de Saint Cloud.  It should come as no surprise that the culture from which the word bureaucrat is derived should be punctilious about household inventories, but there you have it.

Over a period of a couple weeks, I reassembled it and made repairs to the rosewood and tulipwood veneers, then finally a couple of days of shellac pad polishing.  “Pop” Schindler came almost every day to watch and guide me, it was on this project that he introduced me to hot hide glue.

This was a seed for my fascination with exquisite European furniture, especially of the French variety (along with our company’s work at the Wrightsman estate in Palm Beach and their furniture collection), and was truly the acorn from which my fascination with Roubo sprouted.  It was also the prompting for me to embark down the career path of conservation; I entered that stream in 1981 at Winterthur Museum while I was a student in college.

Prepping for Tordonshell Demo

Labor Day weekend will be the inaugural 18th Century Craft shindig here in Highland County, where I have been asked to demonstrate tortoiseshell craft.  Events | Fair Lawn Farm (visitfairlawnfarm.com

Since tortoiseshell is a restricted material I will be focusing my efforts on the making and working of tordonshell, a convincing substitute I invented many years ago so that I could make Boullework, starting with a pot of hot hide glue and a few select additives.  At the time I was contemplating patenting the formula, but then I con$ulted with a patent attorney and found out how much it would co$t to go through with it.  Phooey on that, I just published the paper and moved on.  (The estimate wa$ for a completed application and patent to co$t $25-50k.)

Along the way I made a lot of tordonshell things to make sure it mimicked tortoiseshell accurately, which will be of great interest to the powder horn crowd.

For the rest of this week and into next week I will be gathering the materials and rehearsing the demonstrations, beginning with discussions of the materials themselves and the making of tordonshell so that all the attendees can go home and make it themselves to use in their own craft work.  It will also be a dry run for my next video project which I hope to begin filming this winter.  I keep telling myself I cannot start a new one until I get the Gragg Chair video edited and posted on line.

An Exellent Series On Liquid Hide Glue

Salko Safic has been covering some important territory regarding liquid hide glue over on his always interesting shop blog, culminating the series with a brief but enlightening video.    If the topic interests you — and it should — go give it a look.

A Reader’s Adventures With Gelatin Molds

I got this note recently, and it encouraged me immensely. This is the fellow who planted the seed in the first place in some correspondence going back to last autumn. We have continued to communicate with our respective tinkering with the process.

Hi Don,

I’ve been meaning to follow up with you on my experimentation with the gelatin molds.

In short, at least for now, I’ve changed my expectations for the gelatin mold. I have had excellent results but have changed my goal of utilizing the gelatin mold for “mass” production. Rather, I think these molds work well for making a few casts.

While the gelatin may not last for years like a silicone or urethane rubber mold, it’s key advantage is that it can be melted down after use and reused. Given the cost of quality silicon and urethane this is noteworthy.

I ended up with a basic formula of 2 parts liquid hide glue to 1 part glycerin (by volume). I bought some hardener but haven’t used it.

I enlisted the help of my 6 year old daughter. We started with making a mold of a quarter and then making a cast in chocolate.

Then during Valentine’s Day she made a clay sculpture and then we made another chocolate cast

Then I moved on to a simple wood carving. The gelatin mold worked well here. Unfortunately I seem to have deleted the picture of the cast. Detail replication still very high.

I have kept my recycled gelatin mold material in my refrigerator for about 3-4 months and there has been no sign of spoilage.

More experimentation to come. I’ll keep you posted on the results.


Salvaging A Gelatin Mold. Not!

I mentioned earlier the failure of a plaster casting in a gelatin mold when I had to step back from work in the shop for a couple weeks during my vertigo incident (very much improved, only an occasional moment of lightheadedness), and my desire to try to salvage the warped, shrunken and hardened mold.

To manipulate and try to regain the utility of the mold I immersed it in hot water to see if that would do the trick. The experiment was both a complete failure and an enlightening success. Yes, the mold being manipulated started dissolving and slumping immediately, with, um, some loss of definition, resulting in it becoming a useless blob of goo.

That said it was elucidative of a correct direction in the overall enterprise, and enlightening to the future modifications. The fact that the mold could be manipulated by the hot water was instructive in confirming this strategy for a dried mold. That said it was immediately clear that the proportion of crosslinker in the original formulation was too low, so I will make a new mold with twice as much crosslinking additive. The mold had very little fungal attack so I know I am about at the right place for that additive.

So, sometimes ven failures keep you moving in the right direction.

Stay tuned.

More Improvements to Gelatin Molds For Plaster Casting II

Armed with the knowledge and experience of recent attempts at making gelatin molds for plaster castings I charged forward with some new modifications to the formulation of the gelatin.

In this iteration I used the following recipe:

1 part 135 gws glue (dry granules)

1 part water, soaked overnight and cooked twice

3% glycerin

3% gelatin hardener

0.5% borate preservative

*Note: it is important to actually record the weight of the dry glue granules as the other additives are based on that number.*

Once the glue was well-cooked I added 3% glycerin and 35 of the hardener.  These numbers were based on the dry weight of the granules, in other words for 200 grams of glue granules I added ex poste 6 grams of glycerin and 6 grams of hardener.  I then added 1 gram (0.5%) of borate complex powder as a fungicide to extend the lifespan of the mold itself (not borax from the hardware store).

The working properties of the gelatin mold were excellent, although the time required for it to lose the requisite moisture  to become robust enough to use as a mold took longer than I expected.  I waiting a day before demolding from the pattern, two days would have been better.  But the resulting mold was extremely tough and utilitarian.

I cast plaster into the mold and demolded that the following day. Like the original demolding, longer would have been better.  Next time I will try 36 hours rather than~20.

The plaster casting was still pretty green since it took so long to slough off the water, so I placed it into a desiccation chamber to draw out the water.

The mold itself was placed into a sealed kitchen container with a damp sponge to maintain the water content of the mold, and thus its viability as a flexible and functioning plaster mold.

After three weeks I checked the mold and found a little bit of surface mold.  This was a useful observation.  The initial mold from many moons ago was a lump of goo after two weeks, this was still sort of viable after three.  I think next time I will jump the borate salt content to 0.75% or even 1.0% to see the result.

All in all I was very pleased with the progress being made, both conceptually and practically.

Stay tuned.

More Improvements to Gelatin Molds For Plaster Casting I

Last month I brought together all the new information I had derived from the initial gelatin molds exercises to see if I could nail it on a more complex application.  For this project I used a pattern derived from a cast pot-metal satyr’s head mount from a 19th century pastiche of a 17th century French desk.  The hardware on the desk was cheezy and poorly finished (I was only interested in the desk itself as it was the earliest intact example of “mastic tortoiseshell”; another tale for another time).  This was probably a poor choice for a pattern because its level of “finishing” made it difficult to ascertain the ultimate success of the new mold and casting.

Just before this endeavor I was doing some winterizing and had a bolt-of-lightening stroke of inspiration.  As I was affixing the shroud around the window air conditioner with rope caulk I suddenly noticed the similarity of the rope caulk to the gaskets I made from modeling clay for my molding and casting.  Could it work the same way, but without the time involved in making the the initial gaskets?

I could hardly wait to get back home to try it out.


I built the form-fitted Lego casting dam for the pattern/mold and quickly laid in a bead  of the rope caulk as the gasket on the inner edge of the underside, and pressed the assembly together.  The adhesion and squeeze-out was basically perfect.  Rope caulk is formulated to stick just enough to make it through the winter and then be peeled off with no residue in the spring.  Cleaning off the squeeze-out required only a pass with a boxwood sculpting tool configured  with a knife edge.

Suddenly I was in possession of a new technique to cut many minutes out of my pattern/casting dam set-up for the remainder of my working life.

It was now time to move forward with the modifications to the gelatin formulation and make the mold.


Gelatin Molds For Plaster Casts III

With a number of data points in my hand it was time to give this  gelatin mold thing a trial run.  For the master pattern I selected one from my inventory of such things, an epoxy replica of a carved element.

I first affixed the pattern to a flat board using sulfur-free modeling clay as my gasket and adhesive.  I have used this method as a rock-solid tried-and-true technique for decades, the sulfur-free aspect is critical whenever the mold-making material is silicon rubber (sulfur inhibits the silicon rubber from setting), and I just use this as my default for every similar application.  I cleaned off the excess and brushed the thinnest possible coating of petroleum jelly onto the pattern to assure perfect separation, then dammed the pattern with a rectangle of Lego blocks.

For the gelatin mold material I used 192gws Standard glue, soaked overnight with the water level equal to the height of the glue in the glass jar.  I wanted the gelatin to be on the viscous side, remembering that all the water going in has to come out and there are shrinkage issues to consider with that.

Once the glue was cooked and ready, I added 3% of glycerin to it to serve as a plasticizer  followed by 2% protein hardener to crosslink the glue (both additions were by weight in proportion to the dry glue granules).  I had already learned to allow the glue to cool a bit before adding the hardener as higher temps make the crosslinking go too fast, turning the glue into an heterogeneous lumpy mixture.

Once everything was ready I simply poured the glue over the pattern until it stood proud of the surface by about 3/8″ and set it aside overnight.

The next day I dismantled the Lego dam and the rubbery block was just what I wanted, it peeled away from the pattern easily.  Clearly the hardener was imparting toughness from the git-go and the glycerin enhanced the flexibility of the still-swollen block.


I whipped up a batch of plaster and poured it into the gelatin mold with outstanding results.  Well, as outstanding as I could have expected given my impatience, which caused me to de-mold the casting too soon.  I should have waited 36 hours instead of 12; the high moisture content remaining in the gelatin mold retarded the setting of the plaster.

I immediately double bagged the gelatin mold to keep it from drying out, thinking the crosslinker would inhibit mold growth.

Not so much.  Two weeks later when I opened it to use it again it was something akin to a slimy special effect for a space alien movie.

Still more territory for improvement on the next try.

“Tortoiseshell and Imitation Tortoiseshell” Monograph — New To The Archive

Recently while working to impose order to the library of the Barn I came across a pile of articles needing scanning and formatting for posting to the web.  “Tortoiseshell and Imitation Tortoiseshell” was my contribution to a 2002 conference that required travel to Amsterdam for the presentation itself, in complete disregard to one of my personal mottoes, “If I ain’t at home, I’m in the wrong place.”

The scanned article is now in the “Conservation” section of the Writings section of the web site.  There are two versions, one about 4.5 megs and another about 1.5 megs.  I’m still working through the idiosyncrasies of my scanner and compewder, figuring out what settings work best.  If I can get this better I will upload that version later.

Gelatin Molds For Plaster Casts II – Ye Olde (and new) Apothacary-ness

Developing, or more precisely re-developing, technology that was once common practice requires lots of mental noodling sprouting from the question, “Really, how did they do that?”  In the end it all comes down the the world we artisans inhabit, the world of Applied Materials Science.  Fortunately for me the base material, hide glue, was plentiful in the shop so I had plenty of raw material to work with.  In the pursuit of gelatin molds for cast ornamental plaster my proof-of-concept work revolved around the observations of how molten hot hide glue (the “gelatin”) actually cures into a rigid adhesive layer and the changes in physical properties while en route.

As cooked glue goes from hot liquid to hardened mass the first step is the one of greatest importance for this undertaking.  At a particular point in the process — exactly where and when depends on a number of factors including the concentration of glue solids dissolved in the water, the grade of glue, the ambient temperature, relative humidity, barometric pressure, and probably phase of the moon — the molten glue mass forms a semi-solid gelatin, which is the stage I am interested in both exploiting and maintaining in stasis.  As long as the mass stays flexible it can be used and re-used as a molding material.

A  major concern for this practice is that the gelled/flexible mass be water resistant.  Though that seems, and is, obvious the accomplishment of that feature requires a bit of forethought.  If casting plaster is the end goal, and for me and the ancients it was, the mold for holding the plaster as it was poured and cured had better be able to withstand the incursion of the copious quantities of water involved.

A second consideration is that the gelled mass, once transformed from an amorphous blob into a detailed mold, be tough enough to impart said details and allow for the set plaster casting to be de-molded without destroying the mold itself.

And finally, in order to be used repeatedly the gelatin mold must be preserved over some indefinite period of time.

Reflecting back on protein chemistry and historical craft/art practice I conclude that the ancients accomplished all three of the previous items more-or-less in one step: they added the most potent protein crosslinker/preservative they could find, formaldehyde.  It has long been understood that exposing or incorporating formaldehyde into collagen matrices renders them water-resistant, or “hardened.”  Think embalming.  Think taxidermy.  Think of the ancient practice of exposing bowstrings and wrapping to the smoke of a wood fire, from which formaldehyde is a by-product.

Since formaldehyde is so noxious I do not have it in my chemical inventory and instead relied on another chemical from the world of film-based photography that performs an analogous function of crosslinking or “hardening” gelatin films.  In practice I found this to be an admirable option for a couple reasons, and not so good in another,  Not problematic per se, but requiring another consideration.

To be sure the photographic gelatin hardener performs admirably in imparting water-resistance and integral toughness to a gelled mass.  I observed it also extends the timeline for the gelled state considerably, apparently retarding the water egress that turns gelled collagen into a hard, glass-like film.   This was perhaps my (and their) first ace in the hole.  A second observation beneficial to the process was that the plaster itself, being integrated into and reacted by water, served to moisten and thus keep the mold flexible   Every time plaster was cast into the mold, it re-plasticized the mold mass.

However, the current safer chemistry of the gelatin hardener does not impart the same biocide/preservative effect that was accomplished previously by formaldehyde.  Thus my initial mold attempt turned into a rotting mass of oozing slime in a few days.  Not an un-solvable problem, but a stinky, sticky mess.

Back to the drawing board.  Sorta.

Next time – Ye Whole Sheebang.  This time with lots of pctures.