Not necessarily. Some glues are specifically designed to be gap-filling, and most glues have fairly consistent strength throughout, regardless of thickness.

 On the other hand, both polyurethane and hide glue have specific characteristics that weaken them considerably in thicker layers. If polyurethane is too thick, it turns into foam, as you can see in the photo above.  In retrospect, this was the problem with gluing up belts in batches, as I could never get sufficiently hard clamping pressure consistently across all the glue joints.

Hide glue is reasonably strong in thicker layers when new, but dries out over time and crumbles. This is a problem anywhere, but especially in musical instruments where a loose glue joint can vibrate and cause buzzing noises. When I worked in the violin shop, we had to make all of the joints fit absolutely perfectly — even without pressure from a clamp — to prevent such problems from happening in the future. 

I may have to do some more reading but there's a big difference between filling gaps and adhesion, they're two similar but very different things. Glue isn't really intended to be internally strong, it's intended to be strong adhesively. My professional time using adhesives is 50 years old but I haven't seen anything on a tube / bottle, etc. of glue that contradicts the thinner the complete coat the stronger the bond. 

Modern foaming gap filling adhesive bonds are notoriously easy to break. The Blade Forum has or had an extensive section just for adhesives including tests by professionals, companies, etc. One important thing I learned is the competition between makers is so fierce glue formula change constantly, the epoxy you bought last month may be "obsolete" this month. It's crazy.

There was a sub-section of the adhesives section devoted to makers who made their own "hide glue" and that was a really fascinating section. There were some pretty strong disagreements about what to start with, tendon, hide, hoof, hair, milk, blood, what plants, etc. For someone with a natural Boy Scout / survivalist bent I was really absorbed.

Frosty The Lucky.

That last belt did blow apart as predicted, starting at the thick end of the glue joint (R) and moving towards the thin (L). Note the delamination at the upper right and the tearing of the splice tape at the lower left.

So, I modified the caul to pivot on the end of a carriage bolt. First, I cut off the rigid bolt, drilled a recess in the top of the caul, and drilled and tapped holes for bolting on a retaining plate:

Here’s the new assembly, with another belt clamped up:

There’s a rubber washer between the head of the bolt and the retainer plate, holding the bolt in place while still allowing it to pivot. I may make one or two further changes, but we shall see. 

So, the last belt came out of the clamp, and while it’s an improvement, it’s still not perfect. 

Here it is with the clamp removed, but with the layer of baker’s parchment (to prevent glue sticking to the clamp) still in place:

The darker area on the left and towards the bottom show good adhesion, while the lighter area on the top right didn’t bond quite as well. 

Here it is without the parchment:

 Again, adhesion isn’t quite as good in the shiny area to the top right, and I’ve got concerns about adhesion at the edges.

I think the next step would be to replace the layer of cardboard (between the caul and the parchment) with a piece of hard rubber. This would ideally conform a bit better to any irregularities in the backs of the belts, while still giving good hard pressure. We shall see. 

NB: This particular belt has been glued and reglued a few times, which may have created some irregularity on the back. It occurs to me that it might be worthwhile to abrade the backs of the ends after they’ve been clamped in the jig, so that both halves have the same thickness. I might also try experimenting with solvents to see if I can remove the coating on the back without having to abrade.

Another thought: currently, the caul is the same width as the belt, and this may be why we’re not getting enough clamping pressure to seal the edges. 

However, if I turn the caul 90°, it overhangs the edges slightly, while still covering the entire splice.

This should put sufficient pressure to flatten the glue joint all the way out to the edges and thus to eliminate those potential weak spots. 

Update: glued up another belt with the caul turned, but still with the cardboard pad (on the “only change one variable at a time” principle):

Looking MUCH better, although the splicing tape did move slightly before I got the clamp down. Here’s with the parchment still in place:

And without:

Next step: replacing the cardboard with hard rubber. Even if it doesn’t improve the splice significantly, it should last longer than the cardboard, which also tends to compress over time. 

Also, fun fact: the clamp is great for holding down glued belts for cleanup of both front and back. 

This last photo highlights the next challenge: removing or preventing squeeze-out on the face of the belt. Sometimes it just peels off and sometimes it sticks hard, and I’m not sure why. I’m thinking that masking the grit with painter’s tape on either side of the splice might help. 

How much pressure does that clamp apply? 

A thought about the outside edges of the splices being less bonded than the inside there being open sides where PSI will dissipated by some degree of lateral displacement of the belt. Just a thought.

Frosty The Lucky.

It's adjustable, and I have it set so that when I’m gluing a 36 grit cubitron belt, I have to lean my weight on the back lever to get it to close. No idea what that translates to in psi. 

FYI, here’s the most recent splice with the squeeze-out removed:

I don't usually have trouble figuring leverage, it's just fractions. Finding the leverage advantage on that clamp isn't very easy even IF I knew how it worked. It almost looks like a ramp and compound levers. I'd be surprised if you didn't have to put all your weight onto it to get it past the break point where it'd stay closed.  

What kind of cardboard are you using, business card type? 

Frosty The Lucky.

Yeah, it’s from the side of a box that used to hold plastic bags. 

I don't know if it's a workable idea but it just came to me. We used to spend a lot of time on the Alaska Ferry system transporting our equipment and the secret they used on the Ferry to keep cups, dishes, etc. on the tables in rough weather was simple.

They'd wipe a little silicon calking on with a rag and when dry dishes were seriously non-skid. Perhaps a thin skim of silicone or similar calk would provide enough give on your caul?

Frosty The Lucky.

I found the hard rubber and cut myself a piece. 

Unfortunately, when I was testing it, I had the clamp set to give too much pressure, and I cracked one of the welds holding its mounting bracket. I’ll fix that tomorrow. 

Bummer, I hate it when that  happens. How thick is the rubber?

Read you tomorrow, Jer

About 1/8”. 

I think thinner would be better but I'm skyballing.

Frosty The Lucky.

It’s pretty stiff, so I don’t anticipate any problems. We shall see. 

I'd be more concerned with too thick seeing as all that's needed is enough give for the grit to be pushed into it and provide uniform pressure on the joint.

It should do the trick, IF that's the trick that needs doing of course.

Frosty The Lucky.

Welded up the cracks, ground things flat, and added a reinforcing gusset under the point of greatest pressure. 

Interestingly, this last operation gave a nice visual indication of the HAZ.

 Reassembled the clamp and glued up another belt. 

On a final note, the next-to-last belt did reasonably well before self-destructing, and the latest belt is still going strong. 

The gusset would make a big difference equalizing the pressure on the belts.

Frosty The Lucky.

So, here’s the result of this latest iteration. (Note the spring clamps holding down the parchment. This helps with the alignment of the tape on the splice: after glue is applied, the tape is pressed down in its proper place, the parchment is put on top and held down with finger pressure while the spring clamps are applied, and then the caul is positioned and the toggle clamp set.)

I think the differences in color and reflectivity are because of glue seeping between the tape and the parchment, variations in the abrasion of the belt surface, or both.

The texture of the belt cloth showing across the entire splice (apart from one or two tiny spots) seems to indicate that the pressure was equally high everywhere and therefore that we’ve got a good glue joint everywhere.

We’ll see how it behaves in use, but I’m cautiously optimistic. 

In other news, I gave the jig a wipe with ATF before gluing, to keep the glue from sticking to the steel. It worked:

And I think it may have also helped keep the glue from sticking to the grit side of the belt:

How much time are you giving it before testing? Most glues don't reach full strength until all the volatiles have evaporated or finished polymerizing.

Frosty The Lucky.

The Gorilla Glue website says “Clamp objects for 1–2 hours. For best results, allow 24 hours to cure.” All the belts have been getting at least that much clamping time (more usually overnight) and curing time. 

Ah curiosity assuaged, patience engaged. 

Fingers crossed, 

Frosty The Lucky.

The fact that I’m on a business trip through Wednesday makes waiting both easier and unavoidable. 

Serendipitous happenstance or proper preplanning?

Safe journey.

Frosty The Lucky.