Official Luthiers Forum!

I am working on my very first top.  As a result I don't know how much flex makes for a good piece of wood.

The plate is Engelmann, 2.8mm-2.9 (.11) AAA from LMI (they state it comes from canada and it's better).  It has 30 lines per inch, or more, and a good number of med. rays. Flexing by hand, it feels stiff, it only deflects a few mm. Crossgrain, clearly floppier, but so far I understood it is normal and the bracing plus bridge will fix it.

I also picked bracewood, it is fun to split it and carve struts out of it but then I changed my mind and thought about making the fan braces from the top leftovers.  They are the same exact wood as the plate after all, and splitting neat stripes out of it was a breeze.  I made them about 8-9mm wide, and thickness is 2.8 maybe 2.9mm. 

What sent me in this direction was also Joshua French's comment in an older topic, he stated he is using 7x3mm braces, and then made to a triangular section. I know he has a stash of very high quality Italian spruce, perhaps it is quite stiff, stiffer than my wood I imagine.

My stripes being so thin seem too elastic. I can flex a 25cm long piece up to a 9mm center deflection (i use both hands to  make an arch) before it becomes almost rigid and ready to snap.

I also made a few struts from the bracewood (Engelmann too) at about 6x6mm then shaped to a triangle. They only bend a little then snap.

I won't remove too much material off the top. I am leaving it at 2.8 right under the bridge, and not use a plate.  Around the bridge, perhaps 2.5 and going to about 2.2 at the rim.  There will a few more tenths removed when finished the other side.

I know theres not much one can comment on the all above, but I still hope.   It's not easy when doing the first one :)

Thanks.

_________________
Build log

One more reference to Brunes's article in the Journal of the Guild of American Luthiers Number 79, which includes a qualitative discussion of classical soundboard design.

Thanks for the tip - unfortunately I have no literature at hand, not even the cheapest book on guitar building. I was on a very limited budget lately (bought a new guitar too) so i could only afford materials. And don't ask me what my current tools are :D   I could find a lot of leads in the archive here were i read hundreds of topics. However, when it comes to sharing data on stiffnes, no1 gives exact figures.

I am well prepared for a bad final result, but if I get help it might not sound that bad.  Too much bracing seems to be the most common mistake a beginner makes. I feel like my 7 or 8x2.8mm braces could lead to the other extreme.

_________________
Build log

I have a blueprint of a typical Jose Ramirez III guitar that indicates 2.2mm top thickness for a cedar top. But, there is a lot of variation among builders.

Alex-
You must have some sort of plan or instructions, or are copying an existing guitar?
Yo can download (free) plans for a Torres guitar from
http://mysite.verizon.net/nostberg/

For a first guitar, I'd follow the instructions/plan exactly. You can use the result as a basis for comparison when you do experiments on later instruments.

Wide and flat braces are not the usual thing I've seen in classicals.

John

Thanks guys.  Since I have this great urge to glue some braces today, I cut a new set from a very nice split I found in the bracewood. It turned out to be tight and straight grained and no run out, so i obtained some nice square profile struts quickly. This new set is a lot stiffer than the flat set, at nearly the same profile surface, thus probably about the same weight.


I have pretty much every plan and bracing pic available for free on the net.  They have been a lot of help.  I am more into a Hauser type of construction.  Tores with his ultra thin tops and huge doming is more difficult to do imo.  There has been a lot of interesting data given by Shawn and J french in the Romanillos course topic. I am using some of the stuff learned there too so in the end the top will be a mix of details gathered from here and there :)

If it will sound like a 2k axe, I'll throw a huge party.

_________________
Build log

Bracing done, save for the soundhole patching.  I still got to do some
dressing on the fans.  I really like my harmonic bars, they ended up
very neat and clean which is great considering all i used was a swiss
army knife and sandpaper  



The center fan is flat simply because i started using my flat braces
(top leftovers) then changed my mind, posted this thread etc.  From
center to waist, they were 7x7mm, 6x6 and 5x5 profiles (or about that)
then swissknifed to triangles then tapered.



The top thickness is pretty much as it was in the center, and down to
2.5mm at the waist. I planed the graduating to the end, after i string
her up and hear how she sounds.

_________________
Build log

Alex - Hi and welcome. I'm pretty new to classical guitars myself, so take anything I say with a grain of salt (or two, or three).

If you are at 2.8mm under the bridge, and down to 2.2-2.5 around rim, I'd guess that your top is way overbraced. The fans and particularly the lower closing braces look too tall. (Are Hausers that way too? I know he used a thick top, but he must have compensated for it somewhere.)

Just an opinion. I'd still be shaving. Do you know about where the main top resonance's tone is?

Jim

_________________
Jim Kirby
kirby@udel.edu

If I knock on it, the tuner seems to be picking up smth close to the low E string.  I'm not sure the mic is sensible enough to hear more than that noise similar to when you knock on a door. There is a much lower volume ringing in the top besides the thump. I think I can hear that it is an E when I have my ear close to it while tapping, but maybe I am imagining i hear it. 

_________________
Build log

I thought (and i think heard) that hauser was using small profile braces, that is way i tried using those 7x3s. But for example on that page about the restoration of a hauser II, one can see braces taller than mine are :)

_________________
Build log

I've built only Hauser style classicals and my first thought was that your braces look much too tall. Braces follow the cubed root rule, for example, If you have a brace 2 mm high and another 4mm tall, the 4mm brace will have 8 times the capacity for supporting the load (2x2x2). If you made it 6mm high, it would have 27 times the capacity to support. So in other words, your top gets extremely stiff as the braces go up just a little. If it's too stiff as your appears to be, there won't be enough energy from classical guitar strings to get the top moving, and you will have a very dead sounding guitar. Also, that center brace that seems wider and flatter, for the same amount of weight, you could have a taller and skinnier brace which would add more stiffness without additional weight. That's the whole key here, you want it as stiff as possible, but also as light as possible without compromising the structure. It's a balancing act. It took me a few guitars to even get it close.

Good luck!

John

I just used a signal generator to find out where is the main resonance. I found it vibrates most at 220Hz.  I have been told to try and get smth between 175 and 196.  I'll see how it does after I trim the edges to final shape and shave the V. 

_________________
Build log

Alex, it is looking good. I think that while your fan braces are tall, it should be okay as long as you get the resonant frequency you are going for.

One small suggestion and that is that because you want the top to vibrate as freely as possible, you may want to taper the ends of the fan and angled end braces down to about .5 mm as most of the strength will come from the height in the middle of the brace but if the ends are tapered down the top will be more frre to vibrate in those areas.

Because is difficult to see the width of the center cross brace but it looks a bit wide. With the cross brace you want to balance the stability needed but to not choke off the sound so you can thin the sides of that brace while keeping the height and it could help.

Good luck, the journey is as much fun as the destination when building guitars.

I need to confess the purpose of a wider central fan brace. Before attaching any braces I started to sand the top to get some initial graduation. Between sessions I was measuring the thickness and flexing along grain to see how it changed. It seems I flexed too much, and i got a hairline snap along the central seam, at about 3mm distance, about 3cm long. It is entirely covered by the endblock, V brace and central brace, but just to make sure I glued a wider central brace (8mm)

The middle harmonic bar is 9mm wide and 20mm tall.


Since it is day time I can pump more volume and get a better feeling on resonances. I picked up a new one at 77 Hz, noticeable stronger than the 220 Hz one if I hold the plate by the upper bar or lower waist. 220 Hz is still the stronger one when holding by the middle bar or the areas between fan braces.

_________________
Build log

You might get away with making fan braces and soundhole braces/patches with the left over top material....On some of Hausers guitars the fan braces are only 2-3mm high. However I dont like your chances though of getting enough strength out of such low bracing over the rest of the top. if youre putting an arch into your lower bout then your harmonic bar in the waist area needs to be alot higher than 2.9mm.

Alex;
Ihave repaired many great classicals and the tallest fan for a "Spanish" sounding should be no taller than 6mm -BUT you have to listen & flex the top to get an idea of what you need.
I personally use 5mm & 4mm wide braces depending on what top wood I'm using.
I've used fans from 7mm tall down to 3mm .
It really all depends on the tone(sound) you after.
Light braces give more warmth-stiff ones help the treble and in many cases the bass.
You have to be careful with body volume(Depth)
And width.
A deep body makes for more bass.
Which most nylon strings do not need unless they are over-built.


Your top thickness graduations look great !

www.collinsguitars.com

_________________
Mike Collins

Because I couldn't find any good size C-clamps at the hardware stores yet and I don't have any patience left and for other reasons i decided to glue the bridge before gluing the top 

So, folks, meet Clampzilla    It's made from old oak tiles, and the top one was already bent by drying near the fireplace. All together it's rock solid and can put a good amount of pressure. I've tested and there are no gaps left when tight.

I'll be using LMI white (yes HHG on my next project)  I guess I'll put a liberal amount and let the clamping squeeze the excess out.  But I might overclamp? Perhaps i could count the numbers of turns needed to remove any gaps and use that when there's glue in between ?

_________________
Build log

Hmmm..... Maybe I'm missing something here, but aren't you supposed to glue the bridge to the top?   

Yes, the pic shows a test. Instead of the actual top I use a 3mm spruce stripe to check for gaps. 

_________________
Build log

I'm not sure I understand.
Your plan is to glue the bridge to the top before you attach the top to the sides/rims?
This could work, but you will have to be extremely accurate in all your positioning during further assembly.
The advantage of doing the bridge last is that you can compensate for small errors to get a guitar that will play in tune.

You could use your bridge clamping device on the assembled guitar if you supported the caul very well inside the instrument (with props or small screw-jacks) and used light clamping force-it shouldn't take much force if the bridge is properly fitted anyway. If you got too enthusiastic, you could crush the top, of course.

Cheers

John

What's the shape of the 220Hz resonance, Alex? There's a whole series of them, and the shapes indicate what the mass and stiffness distribution is in the top, while the pitches, along with the overall weight, can give you an idea of whether it's stiff enough. I have been using this information for years to 'tune' tops, and they're coming out OK.

Hmmm.. perhaps I got this all wrong?  I hold the top squeezing tight the mid bar between thumb and index, keep it in front of a loudspeaker and sweep a software tone generator.  At 220 Hz I feel it vibrates more than it does for the other notes. The halftones around 220 move the top a bit more than the other notes as well. Lower than 60hz it does not do much, same for 350Hz or so and up. It has a similar "resonance" at 73 hz if I would hold it by the upper bar or by the V brace joint. 

_________________
Build log

I assume when you say:"hold the top squeezing tight the mid bar between thumb and index," that you're holding it in the center of the brace? What you're getting is probably either the 'ring' or the 'ring+' mode, either of which could be around that pitch.

Try this:
Make some little pads of soft foam, and use two under the upper face brace (shoulder brace) ends, and two on the edge of the top between the wide part of the hips and the tail block to support the top off the bench, with the rosette facing up. Sprinkle some fine powder all around on the top. I use decorative 'glitter', but sawdust and tea leaves also work well.

Hold the speaker right over the top at the bridge location, and as close as you can get it to the surface. Sweep very slowly through the frequency range from, say, 200-250, watching for any signs of movement in the dust on the top. If you only have a couple of watts from your computer sound card, you may not see much. If you can plug that card into an amp that can put out, say, 12 watts, you should be able to get clear patterns. Keep sweeping over smaller frequency ranges around those values where you get movement, and note carefully where the dust moves. With a few trials, even if you don't have a lot of power to use, you should be able to get an idea of the mode shapes. You may find that moving either the pads or the speaker location will make the pattern more active and clearer.    

Dear Alan,

Thanks a lot for the info.  This sounds like a lot of fun to start with.  Will try it tomorrow once I get the thing out of clampzilla. I was just cleaning the glue squeeze  

_________________
Build log