Jump to content

Prostheta

Veteran Member
  • Posts

    14,753
  • Joined

  • Days Won

    354

Posts posted by Prostheta

  1. So this planned out perfectly. Note that the centreline on the headstock part of the Maple deviated at a very slight angle from the body part. The centreline was extended down the headstock by eye. This isn't easy, however a piece of blue tape provides a high contrast "target" so you can tip the blank up and down to sight whether it's to the left or right of centre. Moving the tape and repeating until you're on the mark isn't too tricky once you get a feel for it. Over such a distance, any error is in the magnitude of a mm or two which can be borne in mind for any futher calculations made around the headstock. Once it's on the CNC, it'll be what it is with reference to the edge (57,5mm parallel off the centreline) however it's enough that the wing tip glueing can be considered. The neck portion of the blank measures 25mm in thickness and the headstock 26,4mm. The final thickness of the neck will be brought down to 21mm on the CNC (after a bit of mass stock removal on the jointer) which will advance the neck to headstock transition line a fair distance (4mm of depth at 13deg is 17,3mm if you do your trig). the headstock itself will only receive a skim cut to establish a plane referenced to the CNC. It's safe to say that the transition will move about 15mm along. This needs to be considered if I am to glue on a piece before the CNC gets its teeth into there. The overly-thick headstock is good for the moment, since it helps with stiffness during initial machining. The first neck process showed that the headstock wanted to bend under tool pressure. This isn't an option. I may still figure out some sort of auxiliary workholding jig that supports the headstock from underneath, however I'd want to make that adjustable rather than a one-shot jig.

    20210925_181142.jpg

  2. Those screws are fine since the thread is 3mm. As long as they seat within the upper pilot and have the right bite within the lower one, they're okay. They were just cinched down so the heads sit flat enough so the pieces can't part. I hate those screw heads anyway....a Torx screw would be much safer both for insertion and not stripping out on removal.

  3. Same procedure as always. Once dry fitted and secured, drill 2mm pilot holes through the edges from one piece to the next. Open out the hole on the upmost piece to 3mm so the screw doesn't grab, but pulls the lower piece upwards.

    20210925_160740.jpg

     

    Given how close these screws are to the perimeter, I would have preferred to use small panheads or maybe add washers so the countersink profile doesn't encourage splitting. I don't have either immediately on hand, but these don't need torquing down. Just enough for location purposes.

    20210925_165357.jpg

    20210925_165408.jpg

  4. The question now is where to located the headstock scarf component of the neck blank. The headstock shape is significantly offset and has a width greater than that of the Maple being used. Placing it dead centre would require additional "wing tip" material to be glued either side. So one option is to line up the headstock so that the lower part of the reverse headstock is a single part, requiring the very end tip to be supplemented with a Maple cutoff. The other is the reverse of this.

    cad_250921_1537.png

     

    Drawing this out, we can see that with a 107mm wide main headstock piece we need a 12,1mm wing glueing either side. In my opinion, it would be better disguised below the 1st string's tuner. Wood chosen with equal growth ring alignment and from close proximity should be close to invisible, or at least harmonious. I recall that my old Mirage had a striking contrast between two adjacent pieces....the only fault that you could really pick, other than those MIDI pickup holes and the banged-up headstock....

    137-8.jpg

     

    So this is the sort of arrangement we're looking at. If I mark the headstock component of the neck less than 38,4mm from the centre then we're golden. I'll make up the missing material from wood immediately adjacent, testing for which direction works best for colour and grain appearance.

    cad_250921_1543.png

  5. So now, we're here.

    After throwing the neck part of the blank over the jointer and through the thickness planer, I drew a line that is parallel to the centreline, close to the lower reference edge of the neck. This is the reference edge we're use from hereonout. With all of the cutting, tapering and laminating, that reference edge is absolutely needed to know where anything is at any time. The edge was tuned on the jointer using the same part cut/flip/tip method as doing the taper until the clean cut edge was parallel to the drawn line. I might fine tune this more after double-checking everything, but for the moment it is good enough that the scarf will be out hundredths of a degree at most. If that edge were out of parallel by the 4-5mm it was initially, that angle would be significant enough to be visible.

    20210925_145745.jpg

     

    Back onto the table saw with the 13deg scarfing jig. This time I swapped out the blade to a 400mm combination/crosscut. The riving knife is further back than I would normally set it myself (typical in a workplace....) but not an immediate concern.

    20210925_145629.jpg

     

    The first cut takes off the extra length from the Bubinga and roughly surfaces the scarf.

    20210925_145754.jpg

     

    A tap on the back end of the blank with a hammer advances it a fraction of a mm into the cut, allowing a clean skim to be taken. This is perfect for glueing with no extra work needed.

    20210925_145935.jpg

  6. It's unlikely that anybody can help answer this question better than Google has provided so far, however I am looking for Martin-style U-channel truss rods within Europe. Generally these'll be listed as "Martin style truss rod", "Gotoh truss rod", "Hosco", etc. €16 for a 430mm rod seems about the right price, with a lot of places inflating that cost by anything up to 100%. I really dislike Stewmac's Hot Rods (way too deep), the cheapy "welded with a sledgehammer" Chinesium blue-plastic rods, and don't want to order from outside of the EU for shipping, import and VAT reasons.

    I think Guitars and Woods are going to be the winner here. By far the best deal all around.
    https://guitarsandwoods.com/materials-and-supplies/truss-rod-and-covers/aluminum-truss-rod-410-guitar-65994988.html
    €16 is a great deal. Shame I can't find any other items to pick up from there for this project....

  7. Whilst not attached to this specific project, I made up a second 13deg scarfed neck blank of Maple and Wengé.

    Same as the Invaders neck blank, this is current 26mm thick on both pieces with the same basic oversize dimensions.

    20210925_094906.jpg

     

    The way I stop pieces sliding is to clamp the layup dry, drill a 2mm pilot hole through both pieces, open out the top to 3/4mm depending on the screw thread size, then drive a screw through into the rear component. All of these are outside of the neck/headstock outline of course. I do two from the rear and two from the top. They don't provide any sort of clamping pressure, simply locating. <edited for clarity>

    As I explained earlier about the quality of cut off the table saw, a combination of the high positive tooth angle of a rip blade and the tool pressure deflecting an open skim cut leaves this sort of machining marks. The mating surface underneath was given a sub-mm slow pass which is adequate in a well-supported cut.

    20210925_094921.jpg

     

    The leading edge here looks dry, however it is not. This seam is outside of the workpiece, and the entire face of the body component was evenly coated with Titebond Original. Squeezeout is seen as a gold standard of sorts, and in many ways its presence (given an evenly wetted coating) shows good clamping pressure and all that. If you know the quality of your glueing surfaces, clamping arrangement, etc. then it isn't always an absolutely necessary sign, I prefer less mess, so spend more time ensuring the layup will be good and establishing which clamping arrangement to use whilst dry, In this case I used two small clamps on the outer edges and two honking clamps across the centre. The parts can be marginally massaged into alignment if things aren't 100% correct through hammer taps whilst the two small clamps are holding the layup.

    20210925_094933.jpg

  8. Your thinking seems sound. My understanding is that heat causes lignin to loosen, allowing the cellulose/hemicellulose structures to slip. Water just stops wood from scorching by reducing localised temperatures. Normally with bending wood against an iron, higher temperatures than the maximum 100C of water are used so I'm unsure how to rationalise this since it's a "low" temperature bend. Clearly, it works. That's the result. It might just be that it works for simpler forming such as this, but more complex and demanding bends - those that require a form to be maintained without much retention - may need forming at higher temps to increase permanence.

  9. Your spindle sander looks like it's the same Triton-branded model that I have. I should really make that modification as well as it's far better than clamping bits of wood every time. Very thorough process on the headstock facing veneer bending also. Does the oven-warming do anything specific other than drive off residual moisture?

  10. First order of business. All parts were brushed clean of dust and then wiped with acetone. This isn't a good idea with oily woods like Padauk unless you want to end up with pink Maple. The Bubinga benefits from a wipe, nonetheless. Not a soaking! Then, the layup is flipped and taped from the back. Clamps lightly cinch everything together as a pack.

    20210924_170314.jpg

     

    I removed the clamps and smoothed the tape down to try and get everything as flat as possible on the lower surface. With a good tape, the layup can be opened like a book between the individual mating surfaces for glue to be added. Titebond Original is the best benchtop glue here. Titebond II and III are inappropriate as they're designed to be more flexible to withstand wet environments.

    A length of plywood serves as a flat base to clamp everything up. The tape stops glue getting out. Two cauls and vertical clamps hold the pack in line whilst the rest of the clamps keep everything silly tight. Let's estimate numbers! Well, that's roughly an inch of thickness over maybe 36" of length, so 36 square inches. Each of the large clamp delivers about 1000-1250PSI, small clamps maybe 250-500. Six clamps of varying size, let's simplify and call them all equal size. So maybe 5000-6000PSI across the whole layup, divided by 36. That's what, about 120-140PSI? It's a little under the 200-250 that I try and aim for, but perfectly adequate. 7-8 clamps would have been better (200-220PSI) but things get crowded very quickly.

    20210924_171246.jpg

    • Like 1
  11. Speaking of which, here's today's work so far.

    13deg scarfing jig (I decided to stay with 13deg as it has a different "reveal" of the truss rod access cavity than 11deg) set up on the table saw. The pieces were cut at 40mm, 80mm and finally 120mm blade height. Being a rip saw, the surface finish is not what I'd call adequate for glueup straight from parting like this. A tap with the hammer on the back of the clamped piece allows for a slow steady pass to be taken at a fraction of a mm which is perhaps okay for glueup. If it's not airtight, it can be better.

    20210924_151732.jpg

    20210924_151746.jpg

     

     

    I didn't get photos of this process, however the neck portion of this blankk (I made three) was split into 30mm, 20mm and about 50mm. I cleaned up each part with a thickness planer to a mm under those sizes. The central laminate was tapered on the jointer using the same sort of technique I've described previously years back. Set the jointer cutting depth to half of the difference from one end to the other, cut the piece halfway from the desired thin end first and stop when the cutter head's top dead centre is at the halfway mark. Reverse the piece back to front and tip the front end up, pivoting off that halfway cut. Cut this over the jointer to achieve the required taper differential, and take finish passes until the whole piece is at the right final sizing.

    In my case it was 16mm to 11mm, so I set the jointer at (16-11)/2 or 2,5mm depth of cut. Those two passes left the part about 19mm at the body end (as it came out of the thickness planer) and 14mm at the headstock end. The process wasn't as perfect as I've described it here, so the finish passes were a combination of fine tuning and final sizing. Generally this works quite well, but I did set myself very specific dimensions to hit, and I'd be disappointed in myself if I did half a job....

    20210924_163850.jpg

    • Like 3
  12. I know, I'm quite liking it as well. I have to see if I can find that particular Subaru paint chip that I found which is white but with a subtle gold fleck in the micas. It'll have to be a project for later in the game, however once the process for this guitar is down pat, it'll be pretty simple to make several without that much effort. Most of it to this point has been refining the details that have bothered me, testing this and that, etc.

  13. I don't use gloves for operating hand tools as I need that tactile feedback and control. A glove won't do much to stop a whizzing router bit, and wearing gloves often reduces your feel for a tool's stability and how close it is to riding out of that safe zone.

    My relationship with plunge routers changed when I tried using a fixed-base router. The tippiness of the tool was way lower, and my control for retaining the machine around grabby cuts far higher. If I can avoid using one, I do. Big plunge routers are a liability until you add a large base for higher stability....factory bases are rarely large enough, but serve to make the tool look more marketable!

  14. Alrighty then. Let's look at how I've marked out the neck for sorting the scarf and laminations. Excuse the mess, do come in:

    20210922_184539.jpg

     

    The entire Maple blank measures just over 870mm in length and about 107mm in width at 26mm thick. I marked it out for maximum yield, or at least minimum chance of screwing up. In CAD I measured out the lengths of each piece rather than marking out a 13deg angle. Measuring that way is much more accurate over short distances and acute angles. The two pieces were marked out from opposite ends with 25mm having been added in to the headstock length (this rapidly loses length with any scarf glueup error) and 10mm on the neck length. Once these lines are established, they're taken over the edges, joined together and then a midpoint between the two draw up. That way, I can laminate the neck part with a good degree of excess (plus that 10mm) and the same with the headstock. Scarfing a neck with tight tolerance is prone to issues, and for the sake of a few Euros of additional length on the blank you can easily build in a bit of insurance. I'm unsure how I'm going to cut that 13deg, however I'm thinking it will be a combination of a table saw plus a 13deg-edged scrap of MDF or plywood. It doesn't need to be accurate as the whole thing will be cleaned up further down the line.

    I'll probably take the neck part from one side of that 107mm piece (or however it ends up after lamination) as the grain is almost perfectly vertical on one side, maybe a few degree off on the other. The waste can easily be used to fill out the headstock tips, however I want to try and keep as much width on the neck as possible so the CNC has plenty of excess to grab. Offsetting the vacuum consoles along the width reduces any sort of twisting (this shouldn't happen anyway) and increases flat referencing.

    cad_220921_1911.png

×
×
  • Create New...