Note on Wood Glue and Gorllia Glue

[---k---]

Because the subject of glue comes up often around here, I wanted to mention to everyone that this months Fine Woodworking Magazine did a test of several of our favorite glues. The results were somewhat surprising (or not), and exposed something that is commonly repeated to be false.

Anyway, they tested:
Elmers Wood Glue
Tightbond III
Gorilla Glue
Slow Epoxy
Hide Glue

The joints were testing using an exposed tendon joint, so all glue strength. They cut the tenon 3 different ways: loose (1/64" gap - wow that's tight to me!), snug (finger pressure), and tight (requires a hammer to tap together). They repeated the tests with Oak, Maple, and Ipe. Testing was done at Case Western University, by University staff.

Winner: Tightbond III
Extremely Close 2nd: Elmers Wood Glue & Epoxy
Worst Performer: Gorilla Glue

The difference between Tightbond III, Elmers, and the Epoxy was very small. Hide glue wasn't far behind either.

The big surprise (or not), and this is really the big thing I want to point out, was that Gorilla Glue performed terrible in the Loose configuration. I often hear people recommend to use gorilla glue with poorly cut joints because it will expand and fill the gaps (I think partially for strength and partially to acoustically seal?). Based on this testing, I wouldn't recommend this. The strength of the joint was extremely week. It was about 1/2 of the Tightbond joint. Ouch! And as for acoustically sealing, well it is an open cell, porous foam, so I have my doubts about how well it would work.

So, just use regular old wood glue. If your cuts are really bad and you have gaps larger than 1/64 maybe use epoxy.

[cjd]

This isn't the first time they've done this test (or similar tests)

Titebond and Elmers both produce a few variants of their wood glues. I know that Titebond II is NOT as good as I or III for most situations (I think it's a waterproofness difference, may be an operating temp difference - I honestly don't remember)

A poly based glue such as Gorilla glue (the Elmers poly glue costs less and works better IIRC) tends to stick to more substances, is waterproof, and expands to fill gaps. It's a question of application: Use the right stuff for the job on hand.

C

[Chris7]

It's too bad they didn't also evaluate a fast-curing epoxy. I wonder how much of a strength difference there would be between a fast and slow epoxy.

[Brian Bunge]

I actually prefer Titebond II over I. At least when it comes to veneering with it. It dries faster and seems to re-activate with the heat of the iron better than Titebond I. I'm not sure what that means in the long run of the effectiveness of it for veneering, I just know it seems to be easier to work with from my experience.

[cjd]

So, veneer work doesn't rely on strength nearly so much. And that jives with my memory of Titebond II being a more waterproof PVA variant. So, go ahead and use it for iron-on PVA veneer work.

C

[---k---]

The article mentioned that they did the test about 3 years ago.

gorilla glue has it place, as do all the glues. probably great for bonding wood to non-wood surfaces. But, for simple box building it seems like regular wood glue is best. I was just really surprised by the Loose joint results, since I've seen it repeated so to use it with bad joints because it expands to fill the gaps.

I don't think there would be a difference between slow and fast epoxy. The article wasn't long enough and didn't fully describe the failure mode, other than some were wood failure, some wood and glue, and some just glue. If you got a wood failure mode, it doesn't matter how strong the glue is. Just guessing based on how closely the results were grouped, the PVA and Epoxy were primarily failing the wood.

just thought I would pass the info along.

[ahaik]

I am not so sure it really matters when using MDF because MDF is pretty weak and it will fall apart before the glue will break.
Also with MDF, because it falls apart very easily more important IMO is which glue gets absorbs deeper.

[JohnL]

I have almost exclusively used Titebond II over the past several years with no problems at all. The only time I've used Gorilla glue was in bonding corian baffles to MDF boxes, and then I wiped everything down with a (very) damp rag before I started. I've had no bond issues with those boxes either. YMMV

John

[kingpin]

Damn. I hope my speakers don't fall apart. LOL LOL

Mike

[Ray Collins]

I wonder how liquid nails would rate?

Ray

[Dennis H]

I wonder how liquid nails would rate?

Ray

It's very weak. Mostly it's used under the plywood on floors to prevent squeaks but you still need just as many nails.

[---k---]

There are like 10 different Liquid Nails formulas. Many of them are like Dennis says - very weak. I believe if you look at the tube, many say that they are formulated to never fully harden.

[cjd]

Kingpin: just because it's a weak glue relatively speaking doesn't mean you'll have problems. This was a torture test.

Understanding what glues are suited to what purpose is something most people never even stop to consider. I had a friend assemble his plastic models with Elmers white glue. Of course, they didn't hold together so well. Sometimes we think about it, of course.

[technimac]

Check out the video to see how they did the test.

http://www.taunton.com/finewoodworking/Materials/MaterialsArticle.aspx?id=28853

Gee......and I use Henkel PL Premium Polyurethane Construction Adhesive all the time This is a higher viscosity poly glue than the Gorilla poly that they tested.

I noticed in the video that the Fine Woodworking test was basically a shear test and the Henkel poly is designed to remain flexible, so beyond a certain point it will fail sooner than a rigid glue. I'm not convinced that this is the best way to measure the overall capabilities of a glue. Don't get me wrong, I use PVA glue all the time too, but haven't noticed any major differences between the holding power of it and poly. In fact, when having to rip something apart, both glues are usually stronger than the substrate, resulting in that failing first, not the glueline.

Just to be sure though, I'm going to test one of my finished Modula MT's the same way by using a hydraulic jack to try to crunch it. I'll place the bottom left front corner on the jack and the diagonally opposite back right corner directly above, jammed under a concrete beam. Then jack away until I hear crunching.........

That should approximate the normal stresses that speakers are subjected to.......shouldn't it?

[Ray Collins]

It's very weak. Mostly it's used under the plywood on floors prevent squeaks but you still need just as many nails.

True, true....I walk across an example every day Dennis.

How would it be for constrained layers?

Ray

[cjd]

As I mentioned...

Each glue has its purpose. That includes polyurethane based glues.

However, Gorilla has been making people think it's the stuff for *everything*

*shrug*

C

[---k---]

yeah. wood glue is the cheapest and appears to work the best, so why bother with more expensive glues when building speaker boxes. thats all.

The liquid nails product might work well too. It won't be the strongest joint, but it will probably work for everyone but Jon and Thomas.

[WillyD]

The liquid nails product might work well too. It won't be the strongest joint, but it will probably work for everyone but Jon and Thomas.

God knows it is holding my endcaps together well. I also used Liquid Nails to permanently glue them to the sonotube...and let me tell you, they aren't coming out.

[joecarrow]

I've actually used Liquid Nails formula LN-700 to perform some prototyping. I don't have access to a drill press, mill, welder, or other things that would be nice for building gadgets, so I have to get creative.

One of my old methods was to wrap a structure in kite string and then massage in a liberal dose of two-part epoxy. My new trick is to wrap the structure in baling wire (gives most of the strength), and then smear on a healthy portion of liquid nails. It takes a couple days to fully harden, but when it does there's no way the wire is shifting away from the way I put it. I used this method for attaching a small C-clamp to a DC fan that I was using to cool the motor on a power wheelchair.

It's been said, and I'll say it again- each glue has its strengths and weaknesses. Why would there be hundreds of different commercially successful glues around if that wasn't the case?

In short, and this is starting to have even less to do with building speakers, data sheets are a really good thing. They allow you to go through a rough selection process for candidate glues, and they are often a great resource for figuring out how and how to not to treat the glue and glue surfaces. Sometimes a little heat and pressure make all the difference for a good bond; other times you need water or light or fire. Ha! Those sound like the ancient concept of elements.

Seriously though, preparations that help the bond for one glue and material might totally sabotage another combination. I think that if your machining and clamping are halfway decent, then plain wood glue has proven itself time and again.

[JonMarsh]

yeah. wood glue is the cheapest and appears to work the best, so why bother with more expensive glues when building speaker boxes. thats all.

The liquid nails product might work well too. It won't be the strongest joint, but it will probably work for everyone but Jon and Thomas.

Hey, I resemble that remark! :W

Mostly I'm a Titebond III and slow epoxy user, the latter particularly with the more exotic hardwoods

You know, one of the things I like about DIY speakers is getting to use whatever I want for what ever illogical but emotionally satisfying reason I want to use it, very different from the SMPS engineering I do in my day job.

Interesting to see they did test with some fairly dense hardwoods like Oak and Maple. Now, how about the follow up test with Dalbergia Retusa, so I'll know what glue is best for the Isiris project? (which will be getting some long overdue attention in July) At this point my understanding is acetone to de-oil the wood, and epoxy for when you need to glue... let's face it, when you're dealing with a wood twice as dense as Walnut that doesn't even float, the rules are a little different.

Oh well, last full day in Austria, coming home tomorrow - SFO by 8 PM. PST, that is, which is about 5AM relative to the time zone I'm leaving...

[cjd]

They did test Ipe.

Nails are an interesting fastener. They're actually *intended* to be semi-weak holding flexible joiners. That's why your house uses nails, not screws. So it can bend in the wind, not snap... Liquid Nails has its uses. I bet end-caps on a wood to fiberboard joint with a slightly less than perfect joint (your endcaps Thomas) might be an ideal use for something like this.

C

[kingpin]

If anyone is interested in using a poly or urethane construction type glue, Loctite has a water-based construction adhesive in a caulking tube that I used to glue the front baffle on my speakers. I have also used it to glue the corner bead for my drop ceiling to a steel beam with no clamps.

Cleans up with water easily.

Mike

[Brian Bunge]

Mike,

I like using that stuff to glue down crossover components as well.

[Scott Simonian]

Hmmm...

I built my vented Tempest sub, Adire 281's and the LCC with Liquid Nails. All of them turned out alright. No leaks. I am currently putting together two RLp18 cabs. I have three Liquid Nails bottles. Two are "Heavy Duty" and the other is "Project". Should I use something else?

[Brian Walter]

Hmmm...

I built my vented Tempest sub, Adire 281's and the LCC with Liquid Nails. All of them turned out alright. No leaks. I am currently putting together two RLp18 cabs. I have three Liquid Nails bottles. Two are "Heavy Duty" and the other is "Project". Should I use something else?

Assuming your are building the RLp18 cabs out of MDF or plywood, I would use plain old wood glue unless your joints don't fit very tight, then I might suggest using the Liquid Nails. But personally I wouldn't use Liquid Nails or any construction adhesive for cabinet work, that is not what it's intended to be used for. I think you'll find the reason Liquid Nails wasn't one of the glues FWW tested is because it never occurred to them that anyone would use it for that purpose.

Brian Walter

[JoshK]

I never understood how to use Gorilla glue effectively. Everytime I've used it its made a total mess of the joint. I've tried very little, just a drop or two in a dowel tenon, bonding two 2" maple boards together, and it still leaked out and made an ugly joint. I just use wood glue because it is easy to use and I don't make a mess with it.

[cjd]

Poly glues like slightly damp wood at the joint. And they'll make a mess - that's how they work.

You clean it up after.

It's also waterproof, something PVA is less good at.

C

[PMazz]

Titebond III is waterproof.

[yousuredo2]

Has anyone Noticed A difference in glues when laminating sheets together ?
any specifics on what glue yielded the deadest result ?

Does the glue need to be diluted, etc.

does anything else help to deaden a cabinet ?(sander/Sealer, etc.)

[Rolex]

If you are laminating layers together, Titebond makes a cold press glue that would work well in that application.

[mlw]

now this is what I call construction adhesive

I am really new at the speaker building thing, but thought I would offer a thought on adhesive, with which I have a lot of experience.

I am fully in agreement with assembling MDF with titebond II. I have tried the polyurethanes in many different type tests, and they have their place, but not as often as I thought when I started using them

I am building my own 50L JonM MTM (waiting for the drivers to show, after months of browsing (If my wife hadn't caught me on this forum, I sure she would have been imagining a less savory late night internet obsession will post design, BOM and more than a couple questions in a week or so.

But on bracing inside the cabinet, I am looking at using a little smaller braces ( say a square donut of !/2 baltic birch, about an inch deep in the web, but really attaching them to the cabinet inside., with corner blocks and an occasional verticle rib. My glue of choice at the moment is cartridge tubes of two part, serious Simpson contruction adhesives.

http://www.simpsonanchors.com/catalog/adhesives/at/index.html

They are a little spendy at $18-20 per tube kit - which fits the two parts in one tube for standard caulking guns, with a mixing nozzle. the material has a more goopy, less sticky quality than liquid nails. a little brad nailer o stapler to locate things may be handy. This stuff should be available in most large Building Supply box store, or most places catering to contractors.

I tube would probabibly do the inside of two small speakers, or 1 tube per larger ones. An architect by training, remodeler and cabinet (ie kitchen, bath, etc).
I am thinking a really solid grip on the bracing would be advantagous

a little strength comparison:
Acylic-tie bond strength: 24hr - 2900psi, 14 day 2970psi, compressive yield 10,210 psi
SET Epoxy bond strength: 2 days-3,218 psi 3,366 psi 14 days compressive yield 12,650 psi
Liquid Nails Heavy Duty - shear strength, I week, 240 psi
Titebond II, can reach a bond strength of 3700 psi if clamped at 150 psi at 70deg. - great for clamping cabinets together, but based on a different testing methodology. For bracing, I think the Simpson products would provide a very strong weld of the brace to the cabinet side, even if clamping was not the best.


regards, Michael

[---k---]

I'm a structural engineer by trade and specialize in large-project restoration, so I'm very familiar with the Simpson product you are referring too. I got the full Simpson catalog just to my left. I once used some Hilti HY150 to repair a wood bracket for a bed and Powers A500 to repair a rubber stamp. Works extremely well.

I've spent countless hours contemplating using various construction adhesives. I do agree that they would work wonderfully. Similar acrylic products would be Powers A500 or Epcon A7 but these require special guns. Sika also has several epoxies I've thought about. However, I typically end up in a rush and do a "good enough" job. With a slightly loose joint, the PL Polyurethane glue in a tube works really well and is only $4/tube. I've used it in the past to glue braces in, and just did it to glue my baffle on. Works great.

I'm not disagreeing with you at all. I will probably continue to say that for my next project, it will get assembled with something like Hilit RE500 and the insides coated with a layer of Sika Sikadur 55SLV epoxy. But in reality, it will probably again get assembled with Tightbond and PL Poly.

[tpremo55]

<smart a$$ mode - on/> Geez... you guys probably hang pictures on the wall with $150 a tube 2-part fast/slow drying, waterproof, flame proof, non-volatile, non-VOC, non-pigmented, made of a special polymer sufficient to attach itself to wood/paper/Behr premium paint just in case you have an earthquake/flood/tornado or small child hanging from it... after all, a simple nail would be tragically inadequate... :B </smart a$$ mode - off>
Now, for speakers, how often have we seen structural failure or loose joints resulting in resonance or buzzing? Back in the day, I used to throw parties with an old setup designed to invite the neighbors (Ryan, you may remember some of those... :T) including a pair of 'end-table' Heathkit (Altec Lansing - Voice of the Theater - like) speakers that people used to dance on. I don't list this as a requirement for my speakers today :W.


In the end, I've used Tightbond I, Tightbond II, and a long time ago liquid mails (on particle board) and have NEVER had trouble with any enclosure. On my most recent build with curved sides, the construction introduced a little 'pre-stressing' into the cabinet and relied on the glue to hold under that stress. I used Tightbond II and have had no problems yet (knock on wood).


So I guess I would really like to hear about the 'lessons learned' around the use of glues where knowledge of the results would benefit the group here.


Com'mon... don't hold back. There is no shame in learning the hard way is there? After all, I just admited that I used to let people dance on my speakers in college... ops:

[---k---]

I really like these for hanging stuff on drywall (depending on the direction of the loading):

http://www.easyanchors.com/Products.aspx?product=tnl50&info=features

http://www.easyanchors.com/Products.aspx?product=toggle&info=features

Most of the times, I use 2. It makes leveling them much easier and gives the desired factor of safety.

To beat a dead horse. I agree, except for those of us without adequate wood working skills to get tight joints, the epoxy-type works better. I had a tightbond joint pop open on me recently because the pieces didn't match up flush and I tried to overcome this by clamping a bit of a bow into the piece. I fixed it by using a PL poly in to just fill the gaps rather than trying to close them.

Though, if I remember Jon's speaker building article (posted as a sticky), he claims he heard the best results when epoxy was used. But he also beleives in fancy cables. :P

[Dennis H]

The braces in a speaker probably don't need as strong a glue as the box corners. They are there to help damp panel vibrations, not strengthen the box in the conventional sense. The actual stresses and strains on those joints are tiny. Not that you'd do it but you could probably get away with press fitting the braces in with no glue at all as long as they were tight enough that they didn't buzz.

[mlw]

Just for a little conversation, I see the braces (intimately attached) as "thickening" the wall panels (moving the moment of Inertia), and in the corners, helping to make a rigid joint (fixed) - both of which under my limited knowledge, should provide a more rigid box with effectively smaller panels, hence lower vibrations. I've noticed a few speaker builders out there going for really thick walls. Why not accomplish the same thing with good bracing. I had caught Jon's mention of epoxy, but think glue sqeezeout at cabinet corner joints much easier to deal with using Titebond, which led me to think of epoxying the bracing, which led to the convenience of a two part epoxy in a normal caulking gun, admittedly a little pricey, but that doesn't seem to stop many of you folks
Michael

[Dennis H]

I think the main advantage of epoxy over yellow glue is it isn't water based. MDF tends to swell slightly when it's in contact with water and then it shrinks back later as it dries. It can take a few days to shrink completely. As Pete Mazz has pointed out, sanding and finishing before the MDF dries can result in joint lines showing up later. As far as strength, both are way stronger than MDF so the wood will break before the glue. Break a joint apart and you end up with ragged MDF rather than a clean break at the glue line.

[rc white]

glue

A thing that I looked at when choosing glue is that a panel that is loosely suspended at its edges has around 3.8 times the displacement in its center for the same load as a panel rigidly clamped at its edges.
This seems to imply that the nearer the glued joint is to ridgidly clamping the panel around its edge the better.
In a theoretical glue joint with infinitesimally thick glue then any glue should suffice, but in a real diy glued joint a gap filling builders adhesive will probably allow more flex than is really needed so I go for a small a gap as convenient epoxy glued joint on the principle that the glue itself is more rigid than the panel material.
rcw

[Dennis H]

so I go for a small a gap

Huh? I'm guessing the cabinetmakers on the board (okay I was one back when I was young) are cringing at that statement.

Learn to cut, learn to clamp.

[cjd]

Not everything is about stiffness - sometimes mass plays a role. So, thick baffles (in particular) are helpful. We are, after all, converting electric signal into the physical realm with devices that are pistonic in nature - any "flop" and that's a loss. Significant? Then too, you can only get so much bracing in place to make an equivalence to a thicker box.

Oh yeah. Have I mentioned, we're probably a little OCD when it comes to overbuilding?

C

[---k---]

so I go for a small a gap as convenient
rcw

You left off part of it. I think that means he trys to get as good of a joint as he can, but doesn't sweat it and uses epoxy.

I think we have entered into mental masturbation territory with this one.

[Dennis H]

Ryan, yeah, if you blow it, like your front baffle doesn't seal down tight, you can fix it with epoxy. Doesn't make it right or make it good, it's a bandaid. Learn to cut.

[Hank]

FWIW, I've been using PVA (Titebond and Elmers) for over 40 years and have never had a plywood or MDF joint weakness or leak problem. You won't either if your saw cuts are accurate. Now if crossover design were as simple....sigh....