|
|
Post by lavaman on Jun 11, 2015 9:35:08 GMT
Hi Rob
Many, many years ago I was a design engineer working for Rolls-Royce aero engines. I did some work with carbon composites, but much of what I used to know is long forgotten. However, we can apply some basic engineering principles to guide your decisions.
In engineering terms, components have to be both sufficiently strong to withstand the applied loads and sufficiently stiff to control deflections. Guitar strings apply their load parallel to the neck, so the neck is, in engineering terms, a strut subject to bending and buckling loads. Increasing the stiffness of the neck increases its resistance to these bending loads. Stiffness is proportional to the product of EI, where E is the material's Young's Modulus, and I is the 2nd moment of area.
Young's modulus for carbon composite will vary depending on the fibre orientation, and the fibre to resin ratio. So it's very difficult to arrive at a precise figure. Also, there will be a difference in Young's Modulus between your proposed composite tubular section and the Stewmac rectangular neck reinforcement rods due to the unknown differences in material specifications. For your application, I suggest a conservative assumption that Young's Modulus for the rectangular Stewmac composite is, say, four times that of wood and your proposed tubular section is, say, three times that of wood.
I, the 2nd moment of area, is a factor to take account of the geometrical section. The (I) value for your 4 mm by 8 mm double tubular section is about 50% that of the 3 mm x 9 mm rectangular section.
(Note, these figures are for the area of the reinforcement bars only and do not include the stiffness provided by the surrounding wooden neck)
So, in summary, my 'back of a fag packet' calculations reveal:
- Wood neck E = 1 // I= 1 // Stiffness EI = 1
- Double tube composite reinforcement E = 3 // I = 0.5 // Stiffness EI = 1.5
- Rectangular composite reinforcement E =4 // I = 1 // Stiffness EI = 4
When I consider the overall value of the finished guitar, I think the extra cost of the Stewmac reinforcement bars is worthwhile.
Iain
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135 
My main instrument is: bandsaw
|
Post by R the F on Jun 11, 2015 14:29:00 GMT
Now that's what I call an answer! Thanks very much for taking the time, Iain. The only problem is that I gave up on anyone answering earlier today and ordered the box-sections! I can always think of a justification: how about "I want to retain a little flexibility to allow the truss-rod to do its job" or " I could store spare strings down the middle of the box sections" or "it's an attempt to reduce the weight and therefore increase the resonance of the whole instrument". I'll bear your observations in mind in future and try not to be quite so tight about these things.
Thanks again.
Rob
|
|
|
|
Post by lavaman on Jun 11, 2015 14:40:23 GMT
I think the square tubes will do fine. As you say there is a risk of making the neck too stiff for the truss rod. I like the fact that you're pushing back the frontiers of guitar design. Long may it continue.
|
|
Riverman
Artist / Performer
Posts: 7,381
|
Post by Riverman on Jun 11, 2015 17:39:16 GMT
Yeah, what Iain said...  |
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jun 11, 2015 18:21:09 GMT
I don't imagine I'm doing anything that someone hasn't already done; as I move into my dotage, I'm starting to believe that anything that can be imagined has already been done - and probably more than once.
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jun 11, 2015 22:21:01 GMT
2 weeks today I started building these guitars so perhaps time for an update.
The sides have been bent on a bending iron. I’ve not done very much of this and, though I realise this is one of those things which succeeds much better if you get it right first time, I often don’t get it right first time. Furthermore, the bend at the waist is quite severe and the mahogany was particularly reluctant to go that far. After a lot of gentle (but firm) coaxing, I got both sets pretty close to the shape of the mould:
![]() 
You’ll see that I’ve trimmed them back approximately to shape in the photos above so that they more or less fit in the “trough” which will be used to sand them to their final profile. After that I glued in the neck- and tail-blocks with some difficulty. It’s the first major operation I carried out with fish-glue and I found it does tend to live up to its German name (“Fischleim”); trying to keep the block straight and the ends tight together whilst applying pressure was a slightly slippery experience.
Next time I’ll see if I can tack them in place somehow before cramping. I had a similar experience with the linings but the open time of the glue gives plenty of scope to reposition if things don’t go perfectly straight away:
(You’ll observe that the tail-block in this picture above has been stained with bichromate of potash as a colour test and the one below hasn’t).
More generally I’d say that fish-glue is close to perfect for construction of this kind with, of course, the added bonus of the righteous glow it gives you. I would, however, like to know how long to keep it cramped for; the fact sheet from the manufacturer says 12 hours, which makes gluing on 8 linings a rather time consuming process and explains why I have to stop writing at this point since I’m still gluing and waiting…. Lucky I have a day-job to revert to. You may get blurred images in the background of half-polished tables and half-repaired chairs from time to time. Oh and I have also recovered the roller in my sanding machine; doesn’t it look smart and dandy:
All ready to thickness the back and front plates when they’re prepared.
I’ve also (as previously mentioned) ordered my carbon-fibre and 2-way truss-rod for the neck and the machine-heads. This time they are Gotoh SE-700s – yes, you’ve guessed it: they’re a bit cheaper than 510s especially when you’re buying two sets at the same time! (As I write this I’m becoming aware of the fact that I’m a terrible skinflint; or maybe I’m just a shrewd businessman.) All that’s outstanding now is fret-wire and strings. I don’t think it’ll be finished by the weekend, though.
|
|
|
|
Post by lavaman on Jun 11, 2015 23:51:01 GMT
Are you sure that you've got enough clamps??? |
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jun 12, 2015 6:20:07 GMT
Why? Have you got some more in your loft?!
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jun 14, 2015 10:44:33 GMT
Can't stop tinkering with bracing layouts. At the moment I'm going with the following (the one on the left hasn't changed but the one on the right has evolved somewhat from earlier form - but is it stable? This time it may corkscrew instead of folding!):
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jun 27, 2015 15:52:50 GMT
About time I explained my temporary disappearance. Nothing more exciting than a workshop full of other people's furniture, I'm afraid. The guitars have been pushed well aside for a couple of weeks but I'm hoping to make room for them again soon.
Meanwhile, I've been keeping up my interest by designing a label to stick inside. It's based on a running-group T-shirt I designed a year or two ago (Cafe Runners) but only three T-shirts were ever printed so why waste a lot of hard work?!
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jul 15, 2015 10:45:51 GMT
I believe I left off when gluing in the last of the linings with my precious new fish glue… A cautionary tale: never put a large open bottle of fish-glue on a traditionally stuffed dining-chair seat while you turn away to cramp up your glued surfaces. You will find that the bottle will gently and silently keel over and glug more than half its contents on to the floor before you turn back to replace the lid. (I didn’t have the presence of mind to take photos, I’m afraid, but) I managed not to panic, scraping up most of it and leaving the bits to settle to the bottom before decanting it into a spill-proof container. (I did say this would be a warts-and-all account).
Back to the business in hand. Although the sides are already shaped to accommodate the arching of the top and bottom plates, the edges of the new linings still need sanding down to the precise angle and profile and this was done in a “trough” knocked up from ply and hardboard and then covered with 80 grit providing the right arch for the top and bottom (on opposite sides). The lined guitar sides were then slid up and down the trough till the dimensions matched the plans:
I then glued in the side-braces/split-stoppers but made them rather slighter than usual (about 6 x 2.5 mm) – I’m trying to be very weight-aware and I’ve noticed that some respected makers seem unconvinced that they are really necessary.
The sound-port sites were also reinforced with a couple of layers of veneer – maple, I think.
I also took the chance to photograph the edge of the side/lining while waiting for the glue to go off:
That's walnut - douglas fir - douglas fir - douglas fir - mahogany from outside in.
Moving on to the soundboards. Here are the remains of the door (now a motion picture), before and after butchering:
The bits I’m interested in for soundboard are the middle and bottom rail and you may be able to see that they are not perfect; people always have to put screw-holes and nail-holes in these and nature is happy to split the ends, given half a chance, not to mention imperfections in the wood itself. These problems will need dealing with – or I could move on to another door, I suppose. Here is the end of one of these rails:
80-year-old quarter-sawn Douglas Fir. The main problem is that you can’t use much of it because the dowels intrude so I only get 2 half-width boards (= one soundboard) from each rail. These ones only just provide enough length, as well; I need to use the side where the joint-moulding has left most wood (i.e. the part near the surface of the door). I run these through my table saw and then handsaw the bits the circular saw hasn’t reached and end up with this:
Then I plane the dowels out on one side and plane away the paint on the other:
(For those who like to know these things, it’s a Record 04 ½, probably just post-war). Having got them down to the meaty part of what I need, I run them through my sander-thicknesser a few times and they are ready for gluing edge to edge. First, though, how to get the edges nice and straight. Now I know I should be able to “shoot” them with a (t)rusty old plane but I’m afraid I put together a jig and a router and a straight-edge to do the job for me. The fact that it was inspired by an entry in Stefan Sobell’s “news” blog gives it some credence, I feel! Here are some pictures; I hope it explains itself; the two halves of the soundboard are book-folded together along their join and cramped on top of everything:
And then they are glued together. A fairly traditional tent set-up with a large weight (about 25-30 kilos which I found in the garden when we moved into this house) on top to make sure the good side (i.e. the side you can’t see) is flush; I’ve had bad experiences with this in the past since the boards are a bit thinner than they really should be at this stage and 1mm out of kilter means doing it over again. Hot hide glue, by the way, because that’s what people seem to use for this job. I taped both sides of the joint to keep the mess from spreading too far.
If you're eagle-eyed, you'll have spotted 2 holes in one board and even a repaired split at the end of both. The holes have been lined up with the hole in the soundboard and with the bridge area so I should be able to make them disappear. The repaired splits will have to become a "feature".
Not much progress, I suppose, but you should see the amount of furniture I’ve fixed and polished in the meantime! Actually, since it's not really possible to make a proper living from the furniture restoration business, I am tempted to see if I can’t sell a few guitars… so I’ll be looking for advice from the marketing experts on the forum when I’ve finished the build. You have been warned.
|
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Jul 15, 2015 21:47:52 GMT
Just looking at one of the doors in the house (which are the same style and wood as the doors I use for guitars) I noticed that the muntins consistently have exceptionally close grain - 37 lines to the inch in the one I bothered to check. The only problem is that muntins, as we all know, are quite narrow, varying between 65 and 70mm in width on the doors I've checked. The go to waste at the moment but it occurs to me that we frequently use narrow pieces of decorative timber to make up the back of the guitar so would it be possible to use this very high quality timber in narrow strips for the soundboard itself, though it clearly performs a very different function from the back. I, obviously, don't see why not. Does anyone else have a thought on this? It's a terrible waste of great timber and there are no dowels in it. here's how 65mm strips would come out on my current guitars:
and here's a picture of the toilet door muntin (sorry about that):
or am I making a muntin out of a mullion?
|
|
ocarolan
Global Moderator
CURMUDGEONLY OLD GIT (leader - to join, just ask!)
Posts: 35,722
Mini-Profile Background: {"image":"","color":"c0cfe1"}
Mini-Profile Name Color: 182a3f
Mini-Profile Text Color: 733a1c
|
Post by ocarolan on Jul 15, 2015 22:40:20 GMT
.........
or am I making a muntin out of a mullion?  |
|
R the F
Luthier / Guitar Maker
Posts: 1,135
My main instrument is: bandsaw
|
Post by R the F on Oct 16, 2015 8:39:33 GMT
In the summer the dog almost died  and the DWP sent me a booklet with a letter headed “Approaching State Pension age: Important information about your options”. I therefore decided there probably wasn’t enough time to build another guitar with 13 frets to the body and no cutaway. Unfortunately, I was already well into the build (see above) but felt I couldn’t just plough on regardless having lost faith in what I was doing. Since my design involves a "through" neck rather than a separate neck block (jointed or bolted to the neck), I decided to take advantage of this by slotting the cutaway into the side of the extended neck. Here's the plan:  Once I had drawn it, this, in turn, reminded me of A. J. Lucas’s Pavilion Sweep guitar design and made me take notice of his building system using a solera to re-order the build and to make sure that everything is locked exactly in line right from the start; I’ve had problems attaching the neck in a straight line in the past. So that meant building a solera – which took a while - as well as cutting a chunk (cutaway) out of the rims that I had already built. It also meant starting the build with the soundboard and then the neck – so that’s what I did. In fact, the soundboard had already been started (see above) so the first thing to do was to install a rosette. I bought some constructional (i.e. thick) veneers - sapele and maple were the biggest colour contrast I could find – so that I could make my bindings and rosette a little gaudier and chunkier than last time when I used standard 0.6mm veneers:  Here is the making process: > Design it  > Glue together strips of veneer in the designed order; these are about 2mm deep:  > Coat the back with hot hide glue and tape the front the make sure they stay together when cut. Then cut at the designed angle; I used a little Proxxon circular saw since you have to be pretty precise at this stage and there’s a real danger of using truly foul language if such fiddly things don’t work out:  > After a bit of a fiddle the results looks like this:  … which is where I started with the design:  Incidentally, the drawing program I use is by Siemens and is called Solid Edge. Sounds more like a band than cadcam but there you go. The 3D version costs about £7000, I think, but you can download the 2D version for free. (They are probably hoping I’ll convert to the 3D once I see the possibilities!) Now, actually, I don’t make up the full strip (as above) since it falls apart too easily when you try to manipulate it for fitting so you actually install it component by component. Here is a picture of a practice run on a spare bit of Douglas Fir:  This may look like a total shambles but what you can see is the channel full of fresh hot-hide glue. If you’ve ever done any veneering, you’ll recognise this mess and know that all you need to finish the job is something hot and wet to melt everything and press the bits in; below is my modified brass upholstery hammer sitting in near boiling water so that it’s nice and hot and wet:  I should point out that, just because it’s a hammer, that doesn’t mean I use it for hitting anything; it’s used to press the bits into place. Of course, the bits on the outside go round a bigger circle than the bits round the inside so you have to have a very sharp scalpel and a pair of powerful glasses and be prepared to melt and re-melt and slide bits up and down adding and taking away slivers of sapele until it more or less fits:  Notice the deliberate imperfections which reassure the casual observer that this was not done by machine and is certainly not printed! (Have to polish a table now) |
|
ocarolan
Global Moderator
CURMUDGEONLY OLD GIT (leader - to join, just ask!)
Posts: 35,722
Mini-Profile Background: {"image":"","color":"c0cfe1"}
Mini-Profile Name Color: 182a3f
Mini-Profile Text Color: 733a1c
|
Post by ocarolan on Oct 16, 2015 10:30:30 GMT
Looks very fine indeed! Glad the dog was OK too!
Keith
|
|
Jo whoever he is.
