How I put together a rim

[JRHall]

BTW that rim is bent as it should be -- it snaps right into the mold no need for a zillion spreaders -- NICE!

Thanks, that walnut bends like butter.

[JRHall]

Now I use a 28 ft radius dish to do the top. One reason I do this rather than the flattened upper bout is so I can use a simple radius dish as a clamping caule when I glue the top on.

As I use my radius dish to sand the top edge, I use more pressure on the neck end to bring the angle down to where it needs to be for proper neck angle.

[kencierp]

Seems you have had success doing this:

Now I use a 28 ft radius dish to do the top. One reason I do this rather than the flattened upper bout is so I can use a simple radius dish as a clamping caule when I glue the top on.

As I use my radius dish to sand the top edge, I use more pressure on the neck end to bring the angle down to where it needs to be for proper neck angle.

But the pure geometry does not work out -- if you are flaring the sanding dish down applying pressure that section is now on a different plane then the lower bout -- which is what needs to happen -- the fingerboard plane shoots about 1/2" over the tail block. However, the top actually folds slightly at the intersection of the two planes -- so that means there must be flexing at the same intersection (near the waist curve) of the caul. To get a better match the dish needs to be shimmed on the upper bout where the additional material was removed -- its a math thing. $.02

[JRHall]

Seems you have had success doing this:

Now I use a 28 ft radius dish to do the top. One reason I do this rather than the flattened upper bout is so I can use a simple radius dish as a clamping caule when I glue the top on.

As I use my radius dish to sand the top edge, I use more pressure on the neck end to bring the angle down to where it needs to be for proper neck angle.

But the pure geometry does not work out -- if you are flaring the sanding dish down applying pressure that section is now on a different plane then the lower bout -- which is what needs to happen -- the fingerboard plane shoots about 1/2" over the tail block. However, the top actually folds slightly at the intersection of the two planes -- so that means there must be flexing at the same intersection (near the waist curve) of the caul. To get a better match the dish needs to be shimmed on the upper bout where the additional material was removed -- its a math thing. $.02

Maybe I didn't make myself clear.

I radius the top in one single radius, I just take more from the neck area than the tail. Imagine a 28 ft line extending from the dish down to the point where the center of the whole circle would be. As I work the top I'm actually moving that point back a distance. Still a single point, but shifted.

Jim

[kencierp]

Once you move the dish or rim off parallel or center you are still creating two different planes -- I see this arbitrary movement/machining dimension as a flaw in this particular system -- its why as one guy mentioned he has had 4 out of 100 necks fall in place perfectly right from the start, a practical system should produce perfect results almost always. Why not set up your system so the dish itself produces the correct angle and trajectory, rather then gauging and re-sanding? Just like the Martin factory -- of course they are using a flat sander.

[tippie53]

if that were true Ken , Martin would never have to set a neck. There are side thickness variable and when you place neck block there a variables in the neck block mortise . There are more than one way to do things . I want to control top geometry so the fret board plane works with the saddle height.
The mating of the neck cheeks to the side determine the angle to the top and here is the key .

[kencierp]

John -- you are a skilled trades tooling guy -- I am suggesting that you improve this operation by having simple positive locators just like Martin does (and every body else). As it is now, when you are done sanding and with any process done by eye your results are not consistent, why not make this simple improvement and positively control the angle? As a tooling guy you must know the value and quality to be gained. Its just a suggestion I think would be helpful to less skilled guitar makers.

[tippie53]

I spent an interesting day at CF Martin today to clarify a few misconceptions that seem to be arising. There is in fact a slight radius on the top and an angle applied to the sides to accommodate the neck angle. The radius is very slight and this is to put a radius on the kerfing. When you look at the top in the mold on the production floor you can see this but you have to look very close . I don't think this is anything earth shattering.
The procedure is to remove the flat of the kerfing on the top as this is to match the dome of the top that they expect to occur once the string load is applied. . The top braces Have a 42 foot radius and are glued onto a flat jig and vacuum clamped . The person I was with said that the tops dome but at different rates so they use this procedure for production.
Neck blocks are a 2 piece glue up and are dovetailed after the body is assembled. They want to hand adjust the neck set then and again after final finish is applied. When questioned about all the conflicting information , it is just that this is how they are doing it now and that not all information about how a guitar is made is made public .
I had a camera malfunction today but one of my friends had his cell phone and took pictures of what we were allowed to and I will post them as soon as they are made available to me .
I know many of the fixtures Martin uses and asked why they didn't use a jig for setting the necks , The answer makes perfect sense and that was , We can control machining , we can control RH but we can't control how hands sand wood . The want a higher neck to allow final setting.
I hope that this helps answer a few questions and remember Martin is a production company making over 200 guitars a day.

[deadedith]

John - I want to get this straight. At the Martin shop:
-the sides are sloped from soundhole to neck block to accomodate the neck angle?
-the kerfing is sloped (radiused) a little so when the top domes, there will be a good match?
-bracing that has a 42' radius is forced onto a flat surface for vacuum gluing?


So what is radiused is the kerfing
What is sloped is the rim

Is that right so far?

[tippie53]

yes that is right .
The angle appears to leave a 1/16 inch gap at the neck block
the angle and radius is proprietery but it is there . The one picture my friend took will show the neck angle . I did take apart 3 martins for conversion this year and there was an angle that was visible on the sides and this was part of the reason I wanted to gain this information.