An art conservator's paper on the strength of hot vs liquid hide glues

[Dennis Duross]

This morning I found a paper titled "A Study of The Properties of Commercial Liquid Hide Glue and Traditional Hot Hide Glue in Response to Changes in Relative Humidity and Temperature" at the following link: http://cool.conservation-us.org/coolaic ... 0_buck.pdf

If I read the paper's conclusions correctly, both hot and liquid hide glues joints will fail if enough humidity or heat is applied (obviously), but at pressures that wildly exceed those exerted on stringed instruments.

Wouldn't this indicate (suggest?) that there is no practical difference in the strength of the one vs the other for the purposes of stringed instrument building? Is the paper not saying that a properly prepared joint will fail only under conditions that would crush the instrument completely?

Or am I wrong?

I mean obviously the test did not apply 200 lb of pressure on the test joints for 30 years in a room at 70 degrees and 45% RH, so in that sense it's not an exact replication of the conditions that would be considered normal or reasonable in this context.

Any thoughts?

[Steve Senseney]

Early in the paper, they discuss that this is for conservation of Furniture. Their interests and needs are a little different from instrument construction.

I am surprised that the strength was as good as it was on the liquid hide glue.

The other comment is that the product from Franklin industries was "fresh". It is hard to know if the product degrades rapidly in storage or not. It is unclear what the results would be if the bottle were 2 months or 6 months or a year out of date.

The materials used to keep the liquid hide glue in a liquid form are not discussed. Many companies keep changing their products. The product tested in this study done in 1990 may not be the product being produced and marketed now.

[Dennis Duross]

All true.

But even if newer formulations (since 1990) were weaker by half what they were in 1990 (which doesn't seem likely), they'd still be many times stronger than the forces applied by a 12-string guitar, wouldn't they?

I mean, apply 1,200 PSI of force to a guitar and tell me how great it is that (theoretically) the glue joints wouldn't have failed for another 500 PSI (whether they were liquid or hot hide)---after you've finished sweeping up the matchsticks that your instrument has been reduced to.

I'm not trying to start a fight---and I'll quit arguing after this---but I'm just wondering why liquid hide glue gets a relatively bum rap in the world of luthiery if it is (or appears to be) many times stronger than required, is equally reversible and repairable as hot hide, and has all the advantages of other premixed glues?

[Woodrow Brackett]

Do your own research. Liquid hide glue doesn't dry hard at all. I've tested hot hide glue I added salt to (to make liquid hide glue), as well as store bought liquid hide glue. It's kinda spongey. I would think freshness could be an issue too. I mix a fresh (small) bottle of hot hide glue every week. Knowing that I'm always using fresh glue is one of my reasons for using hot hide glue.

[Steve Senseney]

No disrespect intended, and no interest in any confrontations.

I agree with Woodrow that the hot hide glue sets up harder, and seems to be a different product. I have used a little, but quickly moved to using Hot hide glue.

The main reason this subject comes up often is the need to move quickly when gluing with hot hide glue.

I have managed to use it for quite a while. The need to have every thing ready, use a hair blower, or heat lamp, work in a relatively warm room, turn off the cell phone, and move quickly is what is needed.

[Alan Carruth]

Dennis Duross asked:
"...I'm just wondering why liquid hide glue gets a relatively bum rap in the world of luthiery..."

Because many of us who have used it have had bad experiences.

Basically, the operative word in that report was 'fresh': old liquid hide glue tends to set up soft, if at all, and to be fairly sensitive to high humidity, if my experinece is any guide. I've heard much the same from other folks who have used it. It _can_ work, it's just that you can''t be absolutely sure that it _will_ work every time.

Alan Carruth / Luthier

[Greg Robinson]

I also did not see any discussion I thermoplastic creep in that article. I only skimmed it though...

[Dennis Duross]

I think it's (obviously) reasonable to expect that a test like this would be between two "fresh" products---but whereas fresh hot hide glue means just that, "fresh" liquid hide glue could mean just about anything short of the expiration date. One would assume the product would be a long way from its expiration date as opposed to half-way there, but it's true that the article doesn't make that clear.

But it also comes as no surprise that an "old" glue (of any kind) might give sketchy results---so is it not possible to tell when a glue has lost it's ability to do its job prior to using it? Alan's experience would suggest that it's not possible to know this with respect to liquid hide glue.

Does liquid hide glue lose its strength as it nears its expiration date, assuming it has otherwise been treated properly?

I think I'll ask the folks at Old Brown Glue a few questions. The only thing they put in their glue is urea, whereas it's not clear what goes into Franklin's liquid hide glue.

And yes, there was no discussion of cold creep in the article.

[Waddy Thomson]

Franklin's glue has all kinds of additives to keep the glue from spoiling, in addition to the urea to keep it from gelling at room temp. OBG, only has urea, but it's a lot, and my experience is that it really never gets hard. I bought some and let a film dry in a plastic container. It remained pliable and similar to rubber for a month at which time I threw it away. When you do this with HHG, it turns hard like glass. Same for LMI White and Fish Glue. I have never used any OBG on a guitar. I have had confirmation of my experiment from a few others, but, I have also heard from some that had a different experience. If it's dependent on the batch, then I won't use it.

[John Hamlett]

...Wouldn't this indicate (suggest?) that there is no practical difference in the strength of the one vs the other for the purposes of stringed instrument building? Is the paper not saying that a properly prepared joint will fail only under conditions that would crush the instrument completely?

The simple answer to that question is "yes". Liquid hide glue is strong enough for any joint we use if handled well. That's not the whole story, however, and when we read scientific papers we are reading a specific test. We would have to do a test of the two glues in response to humidity over time, sheer force over time, tensile force over time, etc. etc., and once we have dozens (or more) papers testing different aspects of the two glues, followed by other papers supporting (not refuting) those results, we can be fairly certain that liquid hide glue -used exactly as it was in the tests- is good enough for our purposes.
Meanwhile, there is ample anecdotal evidence to the contrary, as Al mentioned, to keep many of us from using it.

When news comes out of some scientific study suggesting that drug X or nutrient Y has health effect Z, we need to understand that that is one study and others may be needed to support the findings. The effect may or may not be small, the study could be in error, and surely it is a very narrow study with only one or two goals. We shouldn't rush out to the health food store and grab nutrient Y or run to the doctor and plead for a prescription for drug X anymore than we should throw out our glue pots (and 'Crockettes' and bottle warmers...) and convert to liquid hide glue based on the findings of one study.

[Greg Robinson]

Waddy, can you please spell-out OBG and HHG for us? We're trying to discourage the use of acronyms so it's easier for those who aren't familiar with them to follow discussions.
It's best if you first state the full name/term and then follow with the acronym in brackets like this:

Old Brown Glue (OBG)

Then you can continue using the acronym through the rest of your post without potential for confusion.

Thanks.

[Dennis Duross]

I got ahold of William Patrick Edwards today, the maker of Old Brown Glue (OBG), and he very graciously responded to a number of questions. I'll present the entire text of our two emails below. His responses follow my questions (in red type).

[Here's the first email exchange]

Patrick:

I’ve read through your site’s Q&A section, so I know you’ve addressed this type of question before, but I’m wondering if you can shed any more light on it.

There are also several magazine articles on the website that relate to glue.

There’s a discussion on a luthier’s forum regarding the potential for failure in the use of liquid hide glue in stringed instrument construction.

The consensus there is essentially that liquid hide glue is inferior to hot hide glue.

The testing done by Fine Woodworking of OBG vrs. Hot hide glue showed that, in some cases, the liquid glue was stronger than the hot glue. Their test showed that the Old Brown glue was, overall, 3% stronger. The variables were: species of wood and how tight the joint was. Since all the glues which tested stronger are not reversible, it seems that this indicates Old Brown Glue might be the best choice for instruments or antiques.


Some report that (like pva glues) dried LHG does not harden like HHG, making it as prone to cold creep as pva’s, and others claim to have experienced joint failures under normal usage conditions.

I find that the glue which squeezes out of the joint, or the glue which I spill on the workbench remains softer for several days. It will eventually harden, but it dries by loss of moisture. I have cut open joints and found that the glue inside the joint is rigid since the pressure of clamping pushes the water out of the glue into the wood, where it dissipates. It is a fact that OBG dries harder, faster than the Franklin or Titebond, simply since it has fewer chemicals modifying the protein.

Normal use can be considered 250 lbs of constant force. [Note: You all on the MIMF would know better than I whether this is an accurate representation of the forces at work.]

In years of repairing antiques, I have been constantly amazed at the success of difficult repairs using OBG. Chairs, for example, do not have a constant stress, but when a large person sits down there is tremendous stress to the structure. I have repaired structural elements of chairs, which were damaged by previous repairs, and, in many cases, fractured into dozens of elements. Each time I am able to put them back together, carefully cleaning the pieces and building up the structure over several days. In all of these repairs, none have returned.

In fairness, those responses appear to be in relation to the Franklin product and not OBG, but the experiences have caused most to steer clear of LHG altogether.

My early experience with Franklin and Titebond liquid glues made me want to create a better product.


Can you provide any info regarding OBG’s suitability for this kind of use?

I am not in the business of instrument repair. I am hoping that testing and discussion among luthiers on the web will prove or disprove the use of OBG for this work. There are many different elements of instruments, including pianos, and I believe that liquid hide glue must be appropriate for some or all of this work. I trust experienced luthiers to test the application thoroughly.


Does LHG lose its strength as it approaches its expiration date?

The indicators of failure are: extremely liquid at room temperature, strong ammonia smell and indicators of surface mold. My shelf testing has shown that it remains good for over a year in all storage conditions. Longer in the refrigerator. I have some bottles which still test good at 2 years or more. I only guarantee it good for 14 months, and that is the date on the label.


Does it dry hard like HHG?

See above. It dries hard but takes longer.


Does it resist cold creep, and do you know of any tests that might confirm that?

I have not tested creep, except in my business repairing furniture. I know that hide glues in general have the strongest resistance to sheer forces than any PVA glue.


Can a user know beforehand if LHG is still usable or when it should be discarded?

See above. If it is beyond the "best before" date, always do a simple test on scrap wood before using.


Thanks in advance for any help you can provide.

[Here's the second email exchange]

Thanks very much for such a quick (and thorough) response. Would it be ok if I shared your responses on the luthier’s forum?

I hope you do. I was educated as a scientist; I believe knowledge and experience should be shared.


Also, would you be willing to answer a follow-up or two if someone were to ask one?

I learn from questions. Always available by email.


I did read the FWW article, and found your video on hot hide glue very informative.

My original question on the forum was in relation to an art conservation paper that was published in 1990 that tested the strength of LHG and HHG under (separate) conditions of 85% RH and 150 degrees F.

Can you share the reference? Was it an AIC paper?


The joints failed at between 1,200 and 3,000 psi (if I remember correctly). My argument was that since the application of even a fraction of 1,200 lb of force on an acoustic guitar would reduce it to a pile of matchsticks, the question of which was the stronger glue joint was largely (if not completely) academic.

I agree.


That’s when several related their experiences relating to joint failures, and their observations that the LHG they’d used did not appear to dry hard.

Joint failure is often caused by improper joint construction, dirt, grease or oxidation on the wood surface, improper clamping, and failure to use a toothing plane iron. I have never had a failure when I addressed all these issues. Glue failure can be caused, also, by using the glue when it is in the gel state. It must be hot and liquid during application.


They also noted that a catastrophic joint failure under a high load was one thing, and cold creep over time was another.

As I said, I doubt that a proper glue application would result in any glue creep.


. . . .

So I don't know if any of that will change anyone's mind, but he seemed very willing to address the questions and any follow up questions that might arise.

[Bill Hicklin]

Any firsthand experience with liquid fish glue, anyone?

[Alan Carruth]

I've been using liquid fish glue for a few years. Mostly I use it for binding, as the long working time (around 45 minutes!) is helpful. It seems strong enough.

I've used it a few times to glue down the 'tongue' of the fingerboard when assembling a bolt-on neck; again, for the long working time. If anything I find it harder to get apart than hot hide glue (hhg) or Titebond in the same application.

I have heard of some testing of liquid fish glue under stress, and when subjected to heat and moisture, which have found that it's not inferior to hhg. I'm going to reserve judgement on that for a while.

Alan Carruth / Luthier

[Nicholas Blanton]

"Joint failure is often caused by improper joint construction". I think this is easily misinterpreted. Some glues are good at filling gaps- epoxy is so good that some wooden boats might be called Wood Reinforced Plastic. Hide glue is not. Titebond is better than hide glue. So, a joint that's proper for epoxy may not be proper for Titebond, and a joint that's proper for Titebond might not be proper for hide glue.

Also, hide glue and old Titebond let go when they get wet- especially hide glue. The edges of the joint absorb water the fastest, and the joint only has to fail around the edges to completely, eventually fail. So the figures of 1,200-3,000 psi , applied to real glue joint, might only apply to part of the joint, and for a limited time. There are a lot of old wood things out there- furniture,pianos, instruments, WWI airplanes- that have had failed hide glue joints. They were all fine for some years, and then little by little the joints spread apart. This is why so many hammered dulcimers in the mid 1800's were assembled with casein glue- it took much longer for it to fail. Not that there isn't a minimum strength for hide glue- lots of 17th century fiddles still out there- but it's going to be lower than 1,200-3,000 psi if you can follow it over a long time.

Now, the question of whether you WANT all those instruments to last forever is something else. Somebody asked Norman Rockwell about whether he shouldn't be using more permanent techniques so people in the future could see his work ( he mixed ink, pencil, lacquer, oil, or whatever to meet his deadlines) and he said, let 'em paint their own pictures. Maybe we should be letting people 150 years from now build their own instruments, too.