www.KitGuitarsForum.com

There are so many threads and so few facts on this subject. I want the information presented here as accurate and factual as possible. Sylvan Well said he will help. We are going to discuss the fact and the fiction surrounding neck joints. Most newbies are exposed to information overflow and on forums so much is myth.
My background is on CF Martin neck joints and dovetails and while I personally prefer dovetails and use them on the majority of builds, I have used bolt ons also. Here we want to talk openly about neck joints and what makes them good or bad.
Since I am well versed in dovetails I'll share my pros and cons.
Pros
It is the only joint that is self sustaining. It is a traditional joint. The force applied to the joint will actually make it tighter. It is a joint that has withstood the test of time. It can be reset if need be. Shimming a dovetail is allowed. No mechanical fastener is needed which makes the dovetail the lightest joint as far as weight goes.

Cons
This is not an easy joint to master and requires skill to reset. It must be glued. There can be confusion on what the cause and effect on the moves are. You can ruin the neck by misadjusting or over trimming the heel.
While it is a traditional joint, I will admit that the dovetail joint, when done well, will not sound better or worse than a well executed bolt on. If the joint is done poorly and a bolt is applied, you are working against the joint and for all intent and purposes, this can weaken the joint.
All neck joints will have to be "set", meaning that you must establish the neck angle and center line. The difference of the dovetail is that the tenon is used to allow the neck to lower into the neck block.
There are 2 methods of dovetail joints, the prepared neck block and neck, and the cut after assembly.

So how can a newbie tell what is best for them? Well I have worked with many of you, and some have gravitated to bolt ons. Now that is fine so long as it is a well executed joint. You see, it's all about process.
What causes a dovetail to fail?
If you have a poorly executed joint, it may contribute to failure. The mating surfaces need to be secure and you need to be aware of the working areas of the joint. Keep the mind set simple. This is a simple pinch joint. Glue is just a freezing agent for lack of a better term. The glue holds the neck in. A well executed neck will be set till it is steamed apart or damaged.
The male part is the tenon and female the mortise. The working part of this is the lower end of the tenon so that as it is pushed into the block it pulls the neck into the block at the same time. I want to see that my witness marks are at least on the lower 1/3rd of the tenon.
You make adjustments on the tenon and you don't work the block. Once you learn to read the joint, you become better at it. Like anything, experience is key. Practice does make perfect.
So what are the major pitfalls? Getting to know what can hold off the joint, not over adjusting and learning the feel of the neck. What tools to use and how to use and sharpen a chisel.
I also have used some bolt ons. The Martin bolt on neck is often the one most of you newbies will be exposed to. While it will work, it has some downside. There are better methods of bolt on. My personal favorite is a 2 bolt system and I use an over-sized head for better mechanical mating. I have had people on other forums state that 2 bolts are no stronger than 1 bolt. To this I will say NO. If a single bolt is on the bottom of the neck, the stresses of the joint may, over time, allow cold creep. A second bolt mid on the neck helps to control this as you are in the thicker part of the neck. Should a failure occur from a fall or shock, that 2nd bolt may be the difference between playing and repair.
Let us also designate a true bolt on will not need glue. Martin is designed to use glue. Of the bolt ons there is a designation of free extension or glued extension. I prefer glued or at least a bolted on fretboard extension.
I will defer to Sylvan Wells on what makes the better or worse designs. In all neck joints, the joinery is critical. Sloppy joints are poor joints. Sylvan Wells and Rick Davis have been invited to participate in this thread. Both are pro builders and I thank them for sharing their experience.

John has asked me to weigh in on the subject of neck joints. I have been building acoustic instruments since 1977 and have used just about any neck joint that has been tried in the past. Epoxy butt joints, pinned (Cumpiano/Gurian) mortise and tenon joints, bolted mortise and tenon joints with the fretboard glued to the top and currently a bolted mortise and tenon and bolted fretboard, i.e. a completely mechanical joint. The gist of what I have learned is that all of the joints work and if you find one that you feel comfortable with, that is the one you should use. Let’s take a look at what is really transpiring in these joints.
Like most things in life all choices have benefits and detriments. The type of neck joint is probably the most important choice in the life of the instrument, although not necessarily for the reasons you might think.
Historically, the classical guitar was first. It’s neck was an integral part of the guitar. It was not a separate piece to be joined later. The neck and block were one. Therefore, it was extremely stable. And nylon or gut strings did not exert the forces on the instrument that steel strings do. To understand that you have to look at the forces that operate on the guitar.
Let’s talk in steel strings. A set of light gauge acoustic strings generates about 200 pounds of pressure on the instrument. It is constant over the life of the instrument. If you think about that force, it is easy to visualize the fact that the neck is trying to come forward and the fretboard is being forced downward, into the instrument. You have to think about what you are doing in your guitar design that will stabilize that force and, ultimately, delay having to do a neck reset because the neck angle has changed. I have yet to see any steel string instrument that ultimately does not need a neck reset. So, under current designs (Martin, Gibson dovetails and other glued necks), the neck reset is delayed for 10-20 years but will ultimately need to be re-adjusted.
Let’s analyze the forces that cause this situation and how they relate to the neck joint. The factors involved include the 1) neck block itself, 2) the neck joint, 3) fretboard, and 4) the upper transverse brace (and, in Martin’s case the so-called popsicle brace).
I have found that a neck block that is less than 2 3/4” wide is asking for instability. Imagine the block. Before the dovetail or tenon is cut into the guitar the sides have a full 2 3/4” of glue surface. Once the tenon (I’ll use 3/4” tenon) is cut into the guitar that leaves a 1” or less glue surface on either side of the block. Assume your block is 1 5/8” thick. That means that, initially, there is a 2 3/4” x 1 5/8” portion of the block that is glued to the top and back and this effectively stabilizes the block from rotating inward. Whether a mortise or a dovetail mortise is cut into the block this takes away from the glued surface of the block which provides stability.
So, strictly in terms of stability of block rotation, a butt joint would provide the most stable solution. The downside to this approach would be that the joint is totally permanent and cannot be adjusted in the future. BUT we know that all necks need to be reset at some point in the guitar’s useful life. So, this is not a practical solution over the long term.
As a result, most builders have gravitated to either the Martin style dovetail or some variation of a mortise and tenon joint. Let’s analyze the both.
As John points out the dovetail is a woodworking joint that sustains itself; the glue just holds its’ position. It is doubtful that it would ever free itself from the guitar. The downside is that it is more difficult to get right in the initial build and, when it needs to be reset, it is a major repair job to do it in a way that is totally invisible. If you can tolerate the ultimate repair it is a great solution.
Now let’s look at the conventional mortise and tenon and bolts. This too, is a joint that works very well. Usually it is two bolts, one at the top of the joint and the other at least greater than halfway along the bottom portion of the tenon. No glue on the mortise and tenon. Done properly, this is an extremely stable joint on par with the dovetail. However, its’ main advantage over the dovetail is that it is easier to adjust in the build process and to reset the neck in the future.
As with the dovetail joint the fretboard is glued to the top. The glued fretboard, in itself, acts as a brace linking the transverse brace to the neck providing great stability. However, gluing the fretboard in place complicates the future neck resets of either the dovetail or mortise and tenon.
In the past fifteen years or so, industrious builders have figured out ways to attach the neck to the body in a purely mechanical fashion, i.e. bolts on the neck and bolt(s) on the fretboard. However, these methods required a rethinking of the basic structure of the guitar/neck mechanism. If the fretboard is not glued down to act as a bridge to the transverse brace then the top itself must thoroughly bridge the gap to link the transverse brace and spread the torsional force to the sides of the guitar. Most instruments that I have examined with the “double mortise and tenon” joint beef up the transverse brace and remove very little material from the ends of the brace so that the forces can be dissipated to the sides. This seems to work very well. However we don’t have 100 years of history to give us an ultimate answer.
We do know, however that this joint can be disassembled in a matter of minutes, the neck angle re-adjusted, and the guitar fixed completely in less than a half an hour without any damage to the finish or anything else on the instrument. This is a huge advantage to builders who have to deal with warranty issues.
So, in the end, any of these joints work well. The reason that there has been experimentation in this area of guitar building is the builder’s quest to find a joint that is easily adjusted, not only in the build process, but down the instrument’s life. You should choose the one that makes the most sense to your woodworking goals.
I hope that all that read these posts ask questions that they may have regarding any of these thoughts which I have posted.

As Sylvan pointed out, the ease of resets and adjustments.
This is from my personal experience doing aftermarket repair work.
Fact is, to be a viable shop, few can survive on building alone. Learning to reset a neck is an economic advantage. An average Martin warranty dovetail reset is $350 and up. This can help pay bills. Taylor, on the other hand, pays much, much less.
I have used bolt ons and while my customer base is about dovetail mostly, you as an individual builder, must make the right decision for you. I use a 2 screw neck tenon and a 1 1/2" thick 3 1/4" wide neck block, a smaller popsicle brace and prefer a higher thinner truss rod brace. 3/8" wide by 3/4" high
(A good discussion for later)
One fact that Sylvan pointed out, and is often not taken into consideration, is the neck block itself. When the guitar evolved from gut to steel string, the stresses increased. At one time, Martin used a thicker neck block of 1 9/16". Today's block is 1 3/8". That small difference in size caused by failures and increased warranty work.
Finally, the quality of work. The joint integrity is paramount to success.

From a true engineering stand point and as a question to Sylvan, what are the advantages of 1 bolt and 2 bolt designs? I've found single bolt designs loosen faster and I've seen the heel break from shock load, commonly referred to as a dropped guitar.
The position of the nut is also very important, the lower on the heel, the less support.
I experimented with 3 necks
A glued Martin bolt on
B glueless modified Martin with 2 bolts
c glued Dovetail
I fastened necks into a neck block and then clamped in a large machinist vise
I applied weight at the nut area with a sling
A failed first
B failed 2nd although the weight it took for failure was more than a guitar body would normally tolerate.
The holding inserts pulled out and cracked the heel
C had weight applied until resultant failure was the neck
There is no sonic difference between dovetails and bolt ons as evidenced by identical guitars I've built, one Martin single bolt on and one dovetail.
I may be wrong but I think the reason dovetails enjoy continued popularity is the desire to preserve tradition.

I can't stress enough that much of the information you glean from the Internet is inaccurate. Somehow, one must learn to separate valid fact from questionable opinion.

Finally, to those of you considering going pro, learn what your market wants and pursue that avenue.

Thank to both of you this is a great informative post.

you are most welcome

that is what we want to provide.

I also enjoyed reading the thoughts and opinions here. It's well thought out and from experienced folks so I don't take the opinions lightly. Hopefully Brian and RD will also chime in with their thoughts. I hope everyone here knows that while I ask questions and try to understand, I do value experienced opinions. I'm afraid sometimes the way I question things doesn't come across the way I intend so I apologize if it seems confrontational.

Would any of you mind weighing in on this topic. One of the "Cons" of the bolt system is the bolts may not remain tight due to working loose on it's own or, due to the neck block contacting with humidity changes. It seems the first issue (working loose on it's own) can be addressed with several good thread lockers (like LockTite). Some are designed to be non-permanent which would seem like the correct approach. Is this a reasonable solution for this issue?

Now how to solve the issue of the bolts loosening due to the neck block shrinking in low humidity conditions. If the bolts are tightened in low humidity conditions is this an issue? It's not difficult to re-tighten the bolts/nuts while doing a string change (2 min of effort if you include 1 min to find the tool) but it's probably not something one can depend on a customer doing. Any estimate on on how much the neck block thickness changes at humidity extremes? I found an on-line calculator that estimates wood movement here:

http://www.woodworkerssource.com/movement.php

If you select true mahogany and then select "Moisture Content" and enter 10% (air dried moisture content) and 28% (fiber saturation moisture content), it calculates a neck block that starts out 1 9/16" thick will increase in thickness by 0.010". So this online calculator estimates 10/1,000" of expansion between these two extremes.

So on a bolt-on neck, if you initially tighten the bolt/nut to the point it compresses the wood slightly, will the bolt loosen from the neck block expansion/contraction? My experience is that sometimes I do indeed need to re-tighten the bolts. Not that they were causing a problem, but a couple of times I've checked them and they weren't very tight. I did not use a thread locker so I can't say for certain if the bolts worked loose or if the neck block shrinking caused this.

Any thoughts on this topic?

Edit: I'll also ask......does using a laminated neck block help eliminate or minimize neck block movement? Is there another material that could be used for a neck block? Or how about a wood treatment like a pressure treated, glue impregnated wood? (not sure if it would still expand/contract or not)

This is wonderful information, thank you guys.
I probably will never build with a dovetail joint, but I still appreciate knowing how they work and how they fit together

The questions about the neck bolts coming loose and the shrinkage factor in humidity changes is something I have not given much thought to Darryl.
If you use a convex washer under the bolt head, the bolts should remain tight at least until the wood shrinks away. If thread lock is used, the bolts may not come loose but the joint might still become loose if the wood shrinks. Great food for thought. It may still be a good idea to apply some. I just used some on a couple of screws that hold my rearview mirror onto the windshield mount. The stuff does help alot.
I build in basically lower than desirable humidity as I do not have a good way to control it in my garage/shop, so perhaps I won't have an issue on my guitars with the shrinkage possibility.
Good thinking and questions Darryl.

Kevin

Correct
The only reason bolts loosen in my experience is wood compression. I don't think it is shrinkage as much , though it may be a part of it.
The best bolts are the largest heads. I hope Sylvan pops in to give his experience. The smaller the sufraces are the more PSI . If you use a smaller head bolt use a washer for as much surface area as possible

If it's compression of the wood fibers over time due to the force of a tighened bolt/screw, would it be possible to saturate the bolt holes with thin CA glue to minimize or prevent compression of the wood? Seems this would help a lot......but I've never tried it. You could also apply thin CA glue to the area around the bolt hole where the washer sits to help prevent it from compressing. Thoughts?

Over time things will compress I prefer the largest surface area but I defer to Sylvan