Friday, December 27, 2013

Why Not Period Glue?

Synthetic Glue Sucks
I joined the Society of American Period Furniture Makers in 2000, when they first got together.  I really enjoyed going to Williamsburg, even though it was in January, and I had been there dozens of times before.  The reason I had so much fun was the experience of spending a week with like minded furniture makers, from all over the country.  We talked about furniture styles, different methods of construction and tools from different periods, finishes and just about every aspect of our craft that was important.

Usually, when I am in a social environment and others find out I work in wood, the topic of conversation ends up with someone discussing their efforts to build a coffee table or birdhouse.  It is a completely different situation at these SAPFM events.  Furniture design and construction is a real passion with this group, and nothing is too esoteric or obscure to merit hours of intense dialogue.

What fun.

I was an active member in the early years.  I was fortunate to be asked by Roy Underhill to tape a segment on his show, The Woodwright's Shop, which required me to ship a large container of tools and materials back East.  At the same time, since I was there with all my stuff, I demonstrated on stage during the SAPFM conference and made a short video for them about the chevalet.

It was interesting, since there are two back to back sessions of the "Working Wood in the 18th Century" event, and I had a conflict with the second week.  Therefore, I asked Silas Kopf to stand in for me and use my props.  During my presentation, I had each segment of the talk prepared in boxes ahead of time.  Each box was numbered so all I had to do was reach under the bench and pull out the next box which had the materials for that segment of the talk.

I worked fine for me, since I was familiar with all the props, and had a time tested presentation developed over several years of talks at the Getty museum.  It was not so easy for Silas, who does his marquetry using a completely different method.

I need to stress at this point that I think Silas is the greatest marquetry artist working in the US.

As it turned out, he made the effort to use my props and present the talk on French marquetry methods that we had agreed to.  But, after about 15 minutes of his talk, he abruptly changed direction.  "That's the way Patrick and the French do it.  Now I want to talk about how I do it."  His presentation was excellent, but not what we had planned.

I also wanted to contribute to the new Journal of the SAPFM, called "American Period Furniture."  For issue #1 I wrote "Form Follows Process," which analyzed the different methods of work used by craftsmen before and after the Industrial Revolution.  For issue #3 I wrote about my research into the Price Guides of the early 19th century, documenting the time required to make each aspect of furniture using hand tools, "Period Productivity".

1820 Cuban Mahogany with 1980 Synthetic Glue
In Issue #2 I did something different.  During the first session I attended, I was amazed that all the woodworkers there were passionate about choosing the right woods, following period design exactly, understanding period finishes, and so on.  There was a certain amount of divergence in whether or not to use only hand tools, and that I understand.  Most of the members had access to power tools and the general consensus was that it was ok to use power as long as a certain amount of hand finishing was involved.

The area which startled me involved glues.  Practically every person I talked with used modern synthetic glues to make their period furniture.  I could not understand this "blind spot" in an otherwise very academic group of individuals.

So I wrote an article for issue #2 called "Why Not Period Glue"  I took the position that traditional glues were used for centuries and worked fine.  If there was a modern glue which did something better than these traditional glues, show me the advantage and I will use it.  The only one which comes to mind is epoxy, which can be used to repair metal parts.  Of course epoxy should never be used for wood repairs, or worse, tortoise shell, ivory or any other material.

Missing Tenon/ Covered in Plastic Glue
Most of the woodworkers used glues like yellow glue, carpenter's glue, white glue, and other glues with fancy chemical names I never heard of.  Hardly a glue pot among the entire group.

Only those who made their living restoring antiques seemed to understand why it is important to use traditional glues.

Cleaning Surface with Toothing Iron
I was reminded of this article today, as I worked to repair a broken leg on a period Baltimore dining table.  The leg was originally attached with a double tenon to the apron.  One of the tenons had broken and the repair was made by adding yellow glue and clamping.  That repair did not work, of course.

Old Brown Glue Ready for Use
In typical fashion, the clamping was not done properly.  The wood surface was not cleaned in advance.  The missing tenon was not replaced.  And on other legs of this table, the repair was "enhanced" by a series of nails, which did nothing but damage the wood.

As I removed the leg to begin the repair, it was obvious that the synthetic glue did not stick to the wood.  Using a sharp chisel, I was able to pick off most of the glue chips, which came away like flakes of paint.  I think of synthetic glue as a plastic, and plastic does not stick well to wood.  I typically use a toothing iron to scrape away the glue residue and tooth the surface.  You must be careful to not remove wood; just glue.  It takes a bit of care, and the toothing iron helps.

Tenon Replaced, Ready for OBG
 I just felt good, when I was ready to apply the glue for the repair.  All the wood surfaces were toothed and clean of plastic glue.  The tenon was made and installed.  The clamps were ready and clamping blocks were installed.  (See the post on Vector Clamping)

I warmed the glue, brushed it on and applied the clamps.  Another antique repair done professionally, using the same glue that the original maker would have used some 200 years ago.  I'm sure he would approve.

Gigi Inspecting Protein Glue


Chuck said...


A very fine article with some excellent tips on how to do certain operations. I have encountered "inappropriate" glues used on some of the music boxes that have come my way. I am curious about the woods used in the Baltimore table you are working on. The legs appear to be mahogany. Is the apron veneered with thick veneer? What is the secondary wood used/ Sort of looks like southern yellow pine but there is not enough detail to tell. What wood did you use to replace the tenon?

I can tell from what little I have done that Old Brown Glue is a wonderful help in doing repairs. Hot hide glue gels way too fast for some of these operations.

Thanks for passing along the information on removing plastic glue!

HandMadeInWood said...


A very thoughtful article.

If we presume that these pieces have intrinsic historic as well as practical interest it would seem inappropriate to use, as far as possible in restoration, anything other than materials that would have been available to the original maker.

Not only ensuring authentic materials, but I'd suggest that the restoration processes should be reversible. Hide glues can be dis-assembled with care - PVA and most modern adhesives cannot.

All best regards.


W. Patrick Edwards said...


I call it a Baltimore table since it has the distinctive pointed reeded leg detail. However, it is made of all types of American secondary woods.

Of course the top and legs are Cuban mahogany. Single boards which are 23" wide. The apron is veneered with sawn veneer that is still over 1mm thick.

The apron is not brick laid, but single pine boards, and there is a medial brace dovetailed front to back, also pine. The fixed rear apron, with the broken tenon is walnut, and I replaced the tenon with a piece of mahogany.

The fly leg apron is white oak, which is a surprise.

Of course the legs are beautiful Cuban mahogany.

The entire table was refinished with shellac some 30 years ago, and that was when repairs were made to the legs and apron, some of which failed.

There are two identical tables, each with a large single drop leaf at the rear. I assume there was also a third table, with drop leaves on both sides, which would have completed the set.

If all three tables were set up, with leaves extended, it would reach 14 feet in length.

W. Patrick Edwards said...

I did not intend this post to repeat the various features of protein glues that my SAPFM article included.

I simply wanted to point out that plastic glues do not work well for repairing wood.

However, you are correct that period glues, which are reversible, are essential in restoration and conservation.

I will add that these glues are also non toxic, transparent to stains and finishes, easy to use and clean up, can be modified with a wide range of additives, and are cost effective.

They have stood the test of time for centuries, with the first evidence of their use over 8,000 years ago.

What condition will synthetic glues be in after 8,000 years? Just saying.

JC said...

Patrick, I wonder if you could write some sort of post dealing with problems associated with using hide glue.

I have been using it for several years to restore broken pieces and veneers one antique clock cases, but I often have problems with the glue, and I'm still not sure why.

As an example, I reattached the crest from an 1830's wooden works clock using hide glue. The top includes to chimney caps with 3 decorative (thin) boards, in addition to glue blocks.

I had all the parts covered in hide glue, and clamped over night. The repair seemed good until I accidentally knocked the front, and the entire assembly just popped right off.

I've also had similar problems on another clock, where a block holding a finial (under a Vienna clock) simply fell off the case after a few months. I re-repaired it with a fresh batch of hide glue), and the piece fell off again later.

I have read dozens of articles, watched all your videos, and I'm still not sure what I'm doing wrong. I still continue to use hide glue, but I'm always only 50% sure whether or not the repair is solid or not.

I use pearl hide glue sold from Lee Valley Tools, and I make sure not to overheat the glue, using a "bain marie" (double boiler over the stove on the lowest heat).

W. Patrick Edwards said...


First of all, I am surprised at the problems you are having. That kind of problem is unusual.

I do not use Pearl glue. I use granulated glue made by Milligan and Higgins, 192 gram strength.

Be careful of mold in the glue, caused by letting hydrated glue sit too long without heating. Look closely and you can see the fibers of mold. Also, the glue is not a uniform viscosity at 140 degrees.

From time to time (perhaps monthly) take the glue pot and brushes and boil them. Start over with fresh glue.

On the other side, wood surfaces need to be clean and fresh, preferably toothed. Proper clamping is essential.

Glue must be applied when liquid (hot). If the glue gels during application, you get something like a cold solder joint, where it looks good but fails. Kind of like you describe.

When working in a cold shop, pre set the wood surface with a very hot damp rag. This puts hot water in the pores of the wood, preventing the glue from gelling too quickly. Remember the heat in the glue is in the water, so preventing the water from dissipating into the wood prolongs the open time.

When a protein glue joint fails, simply apply a hot wet rag compress to the glue surface, then more glue and clamp it. Sometimes it takes more than one application to size the wood grain and make a strong joint, in particular on end grain.

I used this method on a Grecian sabre leg chair, in mahogany. All the legs were curved and the ends were cut down 2 inches. I used mahogany pieces on the ends of the legs and simply rubbed them in place with hot glue. No clamps were used and the chair supports a large man.

I hate to say this, JC, but you may have "operator error!"

JC said...

For the most part, I've been correctly following the items in your comment.

I usually make fresh glue, and I've seen when it gets mouldy (not a pretty sight). I usually always use a clean brush, since I don't keep a dedicated glue brush. I just wash it out while it's still hot when I'm done my gluing operations.

It's possible that on occasion, I don't have a toothed or 100% clean wood surface. Some of the time, I will simply remove excess old glue, (only if it's old hide glue and hasn't been messed with), and add fresh glue to the existing joint (with the understanding that fresh hot hide glue will re-bind with old glue).

Other times, the new parts may have some stain on them, which could potentially cause adhesion problems.

I usually work indoors (heated house), and I apply the glue & clamp as quickly as possible, to avoid gelling issues. I have also recently started to pre-moisten the wood joints with hot water from the double boiler to see if it would improve the glue joint.

The only other problems I think I could be having would be either with the glue-to-water ratio, or the temperature. I used to use a candy thermometer until I figured out where to set my burner, but I've stopped using it.


W. Patrick Edwards said...

I do not like fresh glue, and the only time I have new glue is after I boil the glue pot and brush and start over...every few months or so.

I like glue which has bee properly cooked every day or every other day for some time. It gets darker and stronger, as long as it is kept at 140 degrees.

Glue doesn't stick on oil or stain. Must be a fresh wood surface.

If old protein glue residue is present, place a hot wet rag on the glue for a few minutes and scrape carefully with a toothed plane iron. If you do not have such a tool, a small piece of hack saw blade works fine as a scraper with teeth. Do not remove wood, just glue.

The glue to water ratio is determined by the drip test. Watch the video. And always keep a stainless steel meat thermometer in the glue. Watching the temperature often is the best way to guarantee success.

Get Milligan and Higgins glue.