Official Luthiers Forum!

It's a PVA of some sort. Any details on his methodology? Shore hardness can be used to loosely correlate to Young's modulus... But the films need to be awfully thick to get a proper reading. It would likely need to start off about 1/2" thick and spend several days at 200 F to bake out all the moisture. Thermal plasticity would be a better determinination.

David, that looks pretty sweet! How did it work out and what did you use for those little shafts?

_________________
Pat

It's not mine, but it is an interesting design, isn't it? It uses piano wire pins, and was designed by James White.

http://www.jameswhiteguitars.com/design/design.html

_________________
David Morris

Found it http://mcknightguitars.com/glue-hardness.html. The test method was what I would consider invalid. He routed slots in a piece of ash to a depth of 1/8", filled with adhesive, waited 15 days, refilled, waited 15 days, planed to expose a flat surface, took three Shore-D readings of each sample and reported raw data with ranges. PVA doesn't adhere to itself, exactly what has been known in the adhesive and furniture industries since the 1940s. Liquid polyurethane foams and had a Shore-D hardness of 0... As much as I dislike liquid polyurethanes, the foam encountered in this test series is not what happens inside a bonding area and this part of the data reinforces my rationale for testing thermal plasticity instead of attempting to test film hardness: by testing plasticity, a real wood to wood bond is being used so the adhesive is being tested "in situ." Despite my qualms with his test method I applied statistical analysis, because I suspected most of the differences were insignificant.

From the three measurements I obtained the mean and 90% confidence interval, which was added and subtracted to the mean in order to form upper and lower bounds (two tailed t-test). For those of you who are unfamiliar with confidence intervals, allow me to explain them in layman's terms. Based on the standard deviation and number of samples, one can state that with a level of confidence (expressed as a percentage value) that the actual value falls within mean ± confidence interval. Generally speaking, with a larger number of trials or extremely clean data, the true mean will converge. Shore-D tests should be very consistent, a properly conducted test will usually vary by 1 or 2 (it's a dimensionless value) while some of the trials in McKnight varied by 15. I would have called that a red flag.

Here is my quick analysis of the data. Colored boxes on the far right indicate that all products coded with the same color had statistically insignificant differences at 90% confidence.



If anyone is feeling diabolical and is up to the task of requesting competitive tests that would be conducted free of charge by a major adhesives company, send me a PM and I can walk you through the process of making it happen. There are several reasons I am suspicious of LMII white. If they want to make the most outstanding PVA ever optimized for luthiers, I would be happy to provide my services as a professional consultant. The same goes for any luthiery supplier, although a minimum production run of 500 gallons is generally required. It would feature a long open time of 15 to 20 minutes, colorless composition, minimal plasticizers, tackifying agents, emulsifying agents, perhaps a bit of sawdust (although this will cause *slight* yellow/brown discoloration.) If I had a lab and appropriate chemicals sitting around I would throw together a batch myself. I'm dead serious, if anyone wants to make the perfect luthier's PVA, let me know.

I hope my explanation and analysis are useful or at least enlightening to most of you. If there are any questions regarding my analysis I'm happy to discuss it.

Hugh,
Once again thank you for the informative posts. You shine the light on what is usually a subject that draws much commentary but not all of it is well informed. It has been reaffirming to me to know my woodworking empirical glue knowledge is accurate. Most of what you have written of has been out there to glean as well but I have never seen it in one place so definitively and authoritatively. From the horses mouth so to speak. I and most of my peers found out pretty quickly that Titebond 2 was not the best for furniture and cabinetry. It is too bad because I have seen many folks think that Titebond 2 and now Titebond 3 represent some kind of advancement. Heck the third generation should be even better right. Can't wait until they come out with 4.
I went through one bottle of LMI white and decided to just use titebond org. One place I will use 2 or even 3 is on exterior doors. With the large cross grain glue surfaces of the rails tenons, (even if split you will still have up to 3 1/2" tenons) I think a little give in my glue is a good thing.
The characteristics of Titebond extend was something I was not aware of.
In regards to the "the perfect luthier's PVA " how close is Titebond extend to that ideal ?
Thanks,
Link

_________________
Cut to size.....Beat to fit.....Paint to match.

Actually, I dangled some concepts in front of them and guided some basic research on adhesives derived from modified biological by-products from other industries which would free them from price increases tied to the petrochemical industry. I insisted that my technology would become TB IV. I told the R&D guys just enough so they were able to make a batch of modified soy flour that was not only a strong adhesive, it was water resistant. What I withheld was the information they needed to make it water proof and structural. They're smart guys, but they are also organic polymer chemists and biochemistry is foreign to them (but not to me.) My goal at the time was to keep doing what I enjoyed while getting out of the tech group, because I knew I was going to get screwed out of my job. I accurately predicted the exact manner in which I was eventually canned. My new boss, who management informed that I was to be groomed to take over management of the group, had woodworking experience limited to 8th grade shop class and his knowledge of adhesives came from his first job out of college over 30 years ago and which lasted for only 3 or 4 years (although if you talk to him on the phone he will claim decades of experience and since my departure has self-proclaimed himself as "resident glue guru.") Immediately after his promotion it was obvious he wanted me out. Unfortunate for the company and customers. My attempt to escape into development failed to pan out. And my attempts at networking with other local adhesives companies who would likely be interested in my novel biochemical synthesis processes have gone nowhere. Even attempting to contact a fellow ex-Franklin employee who is now the technical director at Elmer's failed to get a response, although I would love to get back into the industry. If anyone has any connections, I would love to hear about them.

Sorry, for that tangent. First, I strongly advise that you use only TB II instead of III for exterior doors. TB III has excessive thermal plasticity and I have seen failures resulting from a combination of factors that point back to true adhesive failure on the part of III. This is particularly common on doors that are made from darker woods or than have a dark finish. Many companies used II for years and were convinced to switch over to III by their sales rep (seriously? when is the last time you exposed a door to boiling water?) and following lab analysis and switching back to II the problem spontaneously resolved. Unless you are making wooden cutting boards, or skateboard decks, you don't need or want TB III.

Titebond Extend is the closest to my concept that is currently available. I can't go over the specific differences without getting into formula information, which I still cannot discuss. Without naming names, the person mentioned above is known to leak proprietary information. I have heard it myself, and even had to report an information leak to the company months after I left when someone learned a lot of details from said individual at the IWF booth (out of fear that a leak would point back to me since I was separated from the company a few months earlier and had full access to all formulas due to my position on the MSDS task force.)

Yes, it's LMI White. It's a proprietary glue. " this polyvinyl glue is the only one on the market made especially for instrument makers."(from the website.) Good stuff.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Hugh,
So far cutting boards are the only thing I have used Titebond 3 on. I have only used 2 on doors so I will sleep well at night.
thanks once again, good stuff.
P.S. the tangents are interesting!

Link

_________________
Cut to size.....Beat to fit.....Paint to match.