Lattice kerfing router jig (Lattice English sycamore build)

[Nick Royle]

This is my first thread here, so hi to everyone! (Actual question in bold below my babbling! )

I've only made three guitars: one for me, one for m'dad, and one for a mate's daughter, but I have wood for a couple more and I have "the book", so you may well be seeing a lot more of me! The great woodrat invited me to join and I told him that I'll start slow or I may become the most prolific poster here! Well, the time has come for me to start in earnest!

I'm entering the OLF's Local Build Challenge this year (wanted to enter last year but chickened out), and due to the lack of suitable softwoods here, I looked to English sycamore. Trevor was kind enough to post an all-Koa carbon fiber reinforced lattice braced guitar as inspiration and several things about the idea appeal to me, not least the long term stability despite the thin hardwood top.

I don't have any stationary saws, so I'm going to have to make a router table based jig for cutting the kerfs in the lattice elements. Does anyone have any simple fixtures they could share?

And, if I were to aim for 3.1mm thick braces, would a 3mm router bit be ok for cutting the kerfs?

Thanks for taking the time to read this and thanks in advance for any help at all!

All the very best,
Nick


Here's the drawing as it stands if anyone wants to cast an eye over it for me...



92 degree lattice, 7 x 7 (45mm between nodes) Somewhere in the region of 3mm thick and 8mm high.

English Walnut back and sides (got)
English ripple Sycamore top (sourced, may get tomorrow)
English Walnut neck (I had some oak but I think the dark neck will look better)
English Ash and Chesnut for linings and side splints (got)
English bog Oak for the fretboard (sourced, could get for bridge but too heavy I'm sure)
I'm still trying to work out which English wood to use for the braces.

[Nick Royle]

I've been up all night, researching and comparing my options, and I'm wondering if straight grained and knott free "English" cedar of Lebanon may be a usable brace material in this case?

http://www.wood-database.com/lumber-ide ... f-lebanon/

It's very stable. Used in construction. I was thinking of using Doug fir initially and the stats for that aren't too different to the cedar of Lebanon. And, more importantly, I can actually get the cedar!!!

I must admit that I haven't done any real maths, here. And I'm still trying to wrap my head around how the properties of the cedar of Lebanon should affect the height and thickness of the lattice elements. I wish I could find something identical to King Billy pine!

Thanks again in advance for any help at all!

Cheers,
Nick

[Trevor Gore]

I don't have any stationary saws, so I'm going to have to make a router table based jig for cutting the kerfs in the lattice elements. Does anyone have any simple fixtures they could share?

Hi Nick,

The fixture to use on a router table is much the same as you use on a table saw as shown in the book, page 7-8. The construction of a simple router table is shown on page 4-44.

If you have a "proper" router table you can locate the fixture off the miter slot. If you use the home built style you need to set a fence to run along one edge of the router table. Page 7-8 shows a piece of aluminium angle being used as the sliding fence, which helps protect fingers from the saw. With the router table version the brace wood has to run on the table itself, as a 3mm or 1/8th inch cutter seldom has the cut length to protrude though the bottom of the angle to get to the wood. It's critical to get the registration peg in the right place so that the pitching comes out right and also the angle needs to be right! So the basic concept of the fixture is:
1) a fence to slide in the miter slot or against the edge of the router table
2) a cross fence fixed at an angle to 1) to give the lattice angle
3) a location peg so the the first cut (which is done "free") then locates over the peg to set the pitching of the subsequent cuts

If you want to get fancy you could make an adjustable angle one, but the thing to realise is that adjusting the angle also usually means adjusting the pitch which means making a locating peg which is adjustable for both pitch and angle. I have never gone to the trouble of making a fully adjustable one (one day!) so I have 3 or 4 non-adjustable ones. No pictures, because I think it's a good exercise to figure these things out oneself! , but would love to see what you come up with so I can pilfer your ideas! (Young minds are so much better...)

And, if I were to aim for 3.1mm thick braces, would a 3mm router bit be ok for cutting the kerfs?

Basically, just match the sizes. Do a few test cuts in the wood you choose to see what slot size you end up with, which will depend on the wood, the quality of your bit and the run-out (TIR) in your router collet. Don't make them too tight or you risk breaking the lattice as you assemble/disassemble it.

92 degree lattice, 7 x 7 (45mm between nodes) Somewhere in the region of 3mm thick and 8mm high.

Would recommend staying closer to 80 degrees for a regular (equally spaced elements) lattice. Calculate the EI you need or estimate it from sizings that you have experience with. It's easy to make these things come out way too stiff.

I'm still trying to work out which English wood to use for the braces.

If its going to be all English, probably Douglas Fir (also known as Oregon Pine) which I think is plantation grown in the UK or Scots pine ). No harm in trying English cedar of Lebanon if you can get it.

[Nick Royle]

Thanks so much for the reply, Trevor!

I don't have a "proper" router table, I was just going to make the one in the book. I like the idea of the adjustable rig but I think the simplest solution will win the day. I'll start doing some drawings and getting my head around how to do it. I was thinking I should adapt the saddle slotting jig's fence.

but would love to see what you come up with so I can pilfer your ideas! (Young minds are so much better...)

Believe me, my young mind is unlikely to come up with anything better than "good enough"!

Would recommend staying closer to 80 degrees for a regular (equally spaced elements) lattice. Calculate the EI you need or estimate it from sizings that you have experience with. It's easy to make these things come out way too stiff.

I was trying to keep the angle close to an x, may I ask why closer to 80 degrees would be preferable?

I certainly don't want to make it all too stiff:

I'll get that cedar of Lebanon and work out it's stats but based on averages from the wood database, it's going to be heavier and stiffer than WRC or King Billy, and as the pieces are so thin anyway, I could only reduce their height, at which point it's all going to be heavy for its stiffness....

Essentially, I'm wondering if a sparser lattice would be appropriate. Somewhere between the All-Koa rope bound guitar's top and the cf sparse x from the book? Or will the thin top require that fuller support? I'm just worried I won't get the correct 2 degree rotation with such a strong lattice. (I'm going to read the book all night tonight to work out how to make this decision! I know I'm meant to work these things out for myself but my best experience extends to building an L-00 from the Grellier plans and my maths skills are non existant.)

If the sparser lattice were feasible, should that also be 80 degrees? Or even equally spaced?



I quite like the first one on the extreme left. I figure, I'd still be getting some benefits of the cf, still getting experience with epoxy, and it's certainly slightly less daunting (less fiddly, less machining, no issue with bridge pins and K&K pickup installation, less concerns about over-bracing).

Is it as good a plan though?
Nick

[Trevor Gore]

I was trying to keep the angle close to an x, may I ask why closer to 80 degrees would be preferable?

Smallman did a heap of work on this and the answer was 80 degrees. The reason is to keep the ratio of long to cross grain stiffness in the right ball park. I've made the angle smaller when doing 12 stings to counteract the greater torque. If you go bigger on a regular lattice you end up too stiff across the grain. You need to be pretty accurate, too. If you need, say, 78 degrees to fit the body shape, bridge pins etc., 80 degrees doesn't cut it!

Essentially, I'm wondering if a sparser lattice would be appropriate...

I used the dense lattice on the koa guitar because of the heavy figure/short grain. On straighter grained woods a sparse lattice is fine. Figured sycamore can be pretty short grain. On this sort of thing it is very much a judgement call on what the wood feels and works like. Another thing about sycamore is that it's very white. Until you get epoxy on it. Then it heads for amber very quickly. So test to see what colours you end up with and that the epoxy for the bracing doesn't end up penetrating right through. If it does, you're forced into an epoxy fill on the top, and no longer white.

If the sparser lattice were feasible, should that also be 80 degrees?

Preferably. Or at least close.

[Nick Royle]

Until you get epoxy on it. Then it heads for amber very quickly. So test to see what colours you end up with and that the epoxy for the bracing doesn't end up penetrating right through.

Darn! I hadn't thought of that. That could be a major problem with a ~2mm top. And I'd have to buy the epoxy before I know. I definitely want the top as white as possible. Maybe I need to research other approaches to hardwood tops. If I can't use epoxy and cf, I'll be using titebond. Back to the drawing board.

If you go bigger on a regular lattice you end up too stiff across the grain.

Now I understand!


Thanks Trevor!

[Nick Royle]

Here's my first attempt at maths from the book. Till now I've just been trying to wrap my head around concepts. I've made a lot more practical use of the construction volume than the design volume up till now.

I managed to get some half decent looking quartersawn braces out of the cedar of Lebanon. It's on the light side at around 500 kg/m^3

I got a figure of E = 11.36Gpa... But I really didn't feel like I was using the right units or anything, and this is the first equation (4.4-4) I've attempted since I was 16, so I'm going to post my working out below (because I faintly remember maths teachers liking that ) and just about beg for some assistance..... I'm really surprised that it came out with a number in the right region... I tapped the brace as instructed (between finger and thumb at one quarter of the length, and tapped it at half the length) and got two peaks... 441hz was highest amplitude and then a lightly lower peak at 504hz. When I use 441hz, I get E = 11.36. If that's right, that's practically sitka spruce stats! Am I understanding any of this right?

My working out:

Frequency f
441 hz
Length L
0.43 m
L squared
0.1849
fL^2
81.5409
Depth d (in direction of tap)
17 mm
1.0279 x d
17.4743
fL^2 divivded by 1.0227 x d
4.666332843
above squared
21.7746622


all multiplied by p (E!!!!???)
11.36 Gpa


Mass
72 g
Volume
138 cm^3


p (Mass divided by volume)
0.507246377


Did I get even anywhere close to working this out right? Was it just a fluke that I got a number in the right region? Or was it a fluke to find such a good bit of wood?

Thanks in advance as always, Trevor!

I know this is a cheeky request, just a shot in the dark really as I don't have any epoxy... Could anyone be kind enough next time they mix up some epoxy to paint a healthy amount onto a thin maple/sycamore offcut and see if it soaks through and stains it? I'd love to know before I get the epoxy. Just thought I'd ask on the off-chance.

Cheers!

I'm still struggling with fluxural rigidity, too! I need a "for Dummies" version of the book. Or maybe there should be a qualification you have to get before you're allowed to buy it!

Oh, and modulus of rupture for the cedar of Lebanon is high, too!

Species, Density, Young's, rupture, Max
Sitka Spruce, 480, 12, 70, 35
Cedar of Lebanon , 500, 11.36, 82, 41
King William Pine , 350, 5.8, 69 , 34.5
Western Red Cedar, 290, 4.2, 54 , 27

[charangohabsburg]

Did I get even anywhere close to working this out right?

Assuming theses measurements are correct...

Frequency f
441 hz
Length L
0.43 m

Depth d (in direction of tap)
17 mm

Mass
72 g

...and the volume is calculated correctly too (the width of the bar would be 18.8 mm, right?)...

Volume
138 cm^3

then, according to my calculation your result of 11.36 Gpa for Elong is correct, but...

p (Mass divided by volume)
0.507246377

... is obviously wrong.

[johnparchem]

I get 521.74 kg/m^3 for density.

I think that your cm^3 and g are not the right units to use in the division.

[Nick Royle]

Aah, I messed about with the numbers a bit. Was meant to be 70g not 72 (my brain was in bits from trying to understand how to work out fluxural rigidity)... So E = 10.97.... And density of 507g. Still not bad if it's right. I'm frankly amazed that I got so close to being right!

...and the volume is calculated correctly too (the width of the bar would be 18.8 mm, right?)...

It's 19mm but I may have messed up a number somewhere. Or just rounding errors.

But then if John is right, and I've used the wrong units for volume and mass, then this could all still be wrong. If someone could confirm that for me I'd appreciate it! I don't see where it says which units to use so I just guessed till I got the right sort of number!

Thanks so much for the help, guys! I really appreciate it. Luckily I have little shame in exposing my ignorance! I feel like I've made the first step toward using design principles from the book.

I'm going to prepare some more braces, for the back and UTB at least, and give them a test. I'm not doing a live back so I'm sure these'll be fine.

[johnparchem]

But then if John is right, and I've used the wrong units for volume and mass, then this could all still be wrong. If someone could confirm that for me I'd appreciate it! I don't see where it says which units to use so I just guessed till I got the right sort of number! ... .

Any of the units can work as long as you keep track of them and convert as necessary. Your .507 for density did not have units, I just converted to the units I saw in the book. kg/m^3 instead of what appeared to be g/cm^3.

[Nick Royle]

Thanks, John, I get it now. I'm really feeling my way in the dark here. I'll get better!

So it looks like I worked it out correctly. Which is surprising enough, but even more surprising: It wasn't a totally stupid idea to try the cedar of Lebanon... It sands and carves nicely, it has a decent strength to weight ratio, decent modulus of rupture, no knots, and I got enough to get some decent quartersawn, runout free braces! I'm just going to assume I wouldn't have been as lucky with the Douglass fir and count myself pleased.

I suppose I should check its Q, too?

Talking of Q, I have my sycamore "top" now. It's lovely. Very white and decent if not extraordinary, tight, fiddleback ripple. By using it I hope I'm not limiting the potential of this guitar too much, I keep reading that a maple top is really only suitable for an electro-acoustic guitar due to its high Q and relative lack of tonal colour when played unplugged. I suppose I could put a pickup on it and say it was intentional. (Really jealous of people living in countries with decent spruce!)

Being relatively close in stats to a spruce, I'm guessing that this cedar of Lebanon wouldn't be as good a candidate for use in a cf lattice as King Billy or WRC.

I really need to see if the epoxy will ruin the sycamore before I can decide on anything. If epoxy can squeeze though bridge laminates then it probably can through a thin piece of sycamore, and without the cf lattice to support a thin, low mass top, this could all be a really bad idea. Maybe I should just get a non-English top (Alpine spruce) and say I didn't want to build something rubbish.