M W Dales Guitars on Michael Winston Dales Guitars

A very brief look into OpenSCAD

Sun, 07 Dec 2025 18:04:03 +0100

I like visual CAD tooling, that is designing things by clicking and dragging and drawing on screen. A lot of what I build has a visual element to it along side any technical requirements, and so this makes sense. But once in a while I design something quite simple where most of the complexity is hidden away in the mini-spreadsheet that is the parameters form, that I might as well be doing computer programming: usually things like this are simple connectors or jigs to help me do the more interesting bits.

To take this to an extreme, there is actually a CAD tool that is based around specifying your part as a program, which is the logical extension of that parameter driven modelling. OpenSCAD is a tool based around letting you specify your model as a set of variables, as you would with parametric design in a traditional CAD program, and then using bits of code to say “make a cube with the dimensions from this parameter, and then shift it over by the amount in this other parameter, and then duplicate 3 times (where 3 comes from yet another parameter)”.

Whilst I don’t think this is ever how I’d build guitars, for certain jobs I think this would be quite a quick and simple way for me to specify the design. For example, this collet holder I made many years ago where I just need a bunch of circles increasing in size would probably have been quicker to make if I didn’t need to draw it by hand and apply the parameters to it.

As it happens, I wanted to 3D-print myself a small thing to keep a bunch of bits in, and so I thought this was the perfect excuse to finally try out OpenSCAD (it’s what you can see me trying to make in the screenshots I’ve included).

The other thing that made me interesting in OpenSCAD is that the language itself is what’s referred to as a functional programming language, which is a style of language that I don’t get to use often for work, but is a style I enjoy programming in when I can, as it eliminates many of the opportunities for coding bugs.

But unfortunately my enthusiasm for OpenSCAD was short lived. My point here isn’t really to complain about OpenSCAD, as I admire what they’re trying to achieve, but rather just to save those who like me were curious to be a little cautious on what you can achieve with it.

The first screenshot is actually based on my final code before I gave up, but this is a more honest view of where I ended up:

The first major program is that unlike other CAD tools I’ve used, if you put two shapes next to each other and tell OpenSCAD you want them to join or one to be used to remove parts from the other, it can’t work out what’s going on where the faces touch, you actually have to ensure that the shapes clip into each other by some small amount. No other CAD tool I’ve used before has made the user do this, and it means your design is no longer this nice mathematically pure model that you’d expect using this numbers based approach, you have to keep adding a fudge factor.

You can see in the above screenshot two issues due to lack of nudge factor: my model for a drill bit is a cylinder with a cone on it, and where the two models are joined with a union operation, although the faces touch perfectly, OpenSCAD still has a surface between them where I’d expect the union to make them into a single object. In trying to debug this I then added another cube to subtract from half the model as a hack to give me a sectional view, that cube has the same faces as the other of my model, and so you get the rest of that weird glitching you see on the bottom left in brown. this is all fixed in the first picture by adding a bunch of 0.01 mm offsets to everything to make the sectional cube slightly bigger, and have the code and the cylinder faces run past each other slightly.

This then means that my parameters can’t be just pure descriptions of how big a thing is, you need to know what it’s going to be used for so I can add or remove little fudges as I go. This crops up again if you say have a cylinder that you want to use either as a positive piece that you want to add to a design, then you specify it one way, but if you want to use it to cut a hole in another piece you have to specify it another way due to how it models cylinders internally. Ultimately, I don’t like how I have to think about all this as I’m trying to design a piece, I want the tool to work that out for me, that’s why I’m using tools, to save me effort.

I spoke with a couple of others that have tried it, and they had some other things that gave me an expectation reset also, such as the lack of filleting (rounding edges over).

I really love the idea of OpenSCAD for a certain class of problems, so I hope it continues to improve. For instance, Fusion 360, for all it’s nice parametric design, can’t handle me specifying the formula for spacing out guitar frets, which is why I have written other tools for that, but OpenSCAD in theory would make that super easy to do. I think if you wanted to explore the bounds of OpenSCAD as a hobby, then I imagine you can have a lot of fun pushing what’s possible via mathematical modelling and combining it with 3D-printing or milling, evident if you scroll down a bit on this tutorial page. But if, like me, it’s just a means to an end for specific challenges, I’d recommend sticking to other more fledged out traditional CAD tools for now.

Workshop notes

Fri, 15 Aug 2025 09:04:03 +0100

Continuing on from last update, these “weeknotes” (ahem) follow the new neck I’m building for the second guitar I built. When I started writing about this neck in my notes here I couldn’t say what it was for, as it was a surprise present for my brother, who is the owner of that guitar. I knew he was bringing the guitar back to me as there was a problem with the electronics, and that seemed as good a time as any to finish the job I started in 2017 when I first made that guitar by building it a neck myself, to replace the one I had bought at the time.

Inspection and cleaning

It was nice to see the guitar again when my brother brought it down to me when my siblings visited earlier this year:

It’s a lovely, albeit heavy, guitar. And it’s an example about why doing commissions is fun, as it’s not a guitar I’d have built for myself: the pickups are not my style, I’m not into gold hardware and so forth. But because Tristan asked for that, I was pushed into new things: I had to replace the metal cap on the neck pickup as I couldn’t get a stacked-humbucker in that style with a gold cap, I was forced to do the fretwork myself as buying a fretted neck with gold fret-wire in it much more expensive. I learned a lot on this instrument, and so despite it not being to my taste, I still am somewhat fond of the guitar for what I learned when building it.

Anyway, as my brother is an actual band playing musician, I could see that this guitar had seen some action since I first built it - though it was less worn that I’d expected, with the oil finish still in good condition. That said, two things caught my attention. Firstly, at some point indeed the pickup selector switch had broken and needed replacing (an easy enough job):

And second, it was in need of a good clean after many stage outings and rehearsal sessions. This is a problem with guitars: they pick up a lot of dust and dirt adn grime and they have lots of hard to clean pieces. So I disassembled the whole thing and gave it a good clean.

In the end even the bridge had to come off to let me get dirt that had build up under the rounded edges and under the saddles. Guitar building is a glamorous life - you get to remove other people’s finger dirt from fretboards…

Radiusing the old school way

So back to the new neck for this guitar. On the last batch of guitar necks I’d built, I stepped away from using radius-sanding blocks for putting the radius into the fretboard, and switched to using a router. You can see a picture from that earlier post below, and there’s also a video on that post if you want to see it in action.

Using a router is such a better way to put the radius onto neck: like we discussed in the last post with the table-saw method of cutting the fret-slots, it’s both more accurate and faster, a total win-win, and something I don’t really want to give up compared to my old method. However…

The jig you see above I used before was made for the shorter 24" scale-length necks on builds like Delfinen, and so I needed to make a new jig for the longer 25.5" scale-length neck for The Chuncaster. However, when I fired up the CAD model to draw up the new jig I discovered that the headstock on the longer necks is slightly wider (the 24" neck is on the left, the 25.5" neck is on the right):

This wasn’t a particularly deliberate change, I suspect I just tweaked the bezier curve slightly when having to update the CAD model from one scale length to another without realising it. Whilst it looks like in CAD I’d get away with it just if I built the jig on the right, I doubt I’d get away with it in practice; when I made the jig for the 24" necks it actually split at the narrow point there, so I think I’d end up fouling on the headstock with the router jig.

I can claim, truthfully, that this is because I designed the 25.5" neck before I changed methods for radiusing, and so isn’t an oversight or mistake on my behalf. But there isn’t really an excuse for not having gone back and updating my 25.5" neck template to accommodate using the router for radiusing the fretboard I started making this neck, that’s just lack of foresight on my behalf, and that I didn’t realise I’d tweaked the headstock profile in this way.

The upshot of this meant I had to go back to putting the fretboard radius on how I used to do it before I had access to fancy power-tools and jigs, and broke out my old radius sanding blocks.

You’ll see here that I’m using some sort of jig to hold the neck and give me a straight reference parallel to the centreline of the neck. This is actually a fret-slot cutting jig that was in the workshop before the table-saw setup we saw last week was built (or perhaps why the table-saw setup was built). I’m using it to keep one side of the radius block along a straight edge parallel to the centreline because if you don’t, as a result of how your body rotates as you reach, you always end up putting a slight bias into the radius making it lower on one set of diagonally opposite corners than the other. Even with the jig you have to be careful: you can see here in this next photo what I’m talking about, as if you look at the dust you can see I’ve sanded less in the top left and the bottom right than I have in the opposite corners:

To combat this I switch sides I push the block against and what hand I’m leading with when pushing the block regularly to even out any bias over time, and checking regularly to spot if I am introducing any bias by looking end on.

Not only is sanding the radius onto the fretboard a lot more fraught in this way than using the router, but it’s also a lot more of a workout. But after an hour (according to the timestamps on the photos) I was done:

Next time I’ll make sure I tweak my headstock design to accommodate the radiusing jig!

New-school holes for old-school tuners

On most of the early guitars I built I used vintage-style tuners, which I really like, both from an aesthetic point of view and I find them easier to string up a guitar with. However, whilst I find easier to use as a player, they presented a challenge for me as a guitar builder when I stepped away from using the CNC-router to build necks, because they require a stepped hole in the headstock to properly centre and support them. You can see here in the CAD model for one of the earlier necks I made:

The tuner-post is quite narrow, and so the lower hole is cut to that diameter, I assume to make them easier to centre, and then the metal plug that goes into the top of the hole that the post passes through requires a wider top section to fit in. The stepped hole is easy to make when you’re CNC-routing your neck, but when I switched to making things without CNC-routing I decided that this stepped hole was not a thing I felt confident making by hand, so I switched to using modern-style tuners, where you drill a single 10mm hole through and more of the tuner sits in the hole to make sure it’s placed correctly and securely. (I’d show you a CAD model of that, but I’ve never made one, as visually the new and old methods look the same on the outside, and given I’m not CNC-routing I’ve had no need to specify it in such detail - I’m a little lazy with my CAD like that :)

The point of this is that Tristan’s guitars has the vintage-style tuners on, which he likes, and so I need to work out how I’m going to hand-build the holes for them to fit into. I was pondering how I might do this when workshop-mate Matt said he’d had success using vintage-style tuners with just straight through holes - so long as the holes are aligned right you just need the plugs on the front for centring them, so it all works out okay. Which makes sense, but I needed the confidence of someone else to make it okay for me to feel I could try that!

Despite the impression I gave earlier about lacking foresight, occasionally I can think ahead, as demonstrated in the photo below. You can see there are two small screw holes in the headstock that are where the outer two tuners will need to be positioned: these are not there by accident! I use screws to mount the initial template to the plank of maple that becomes the neck, and I have two of those screws line up with the centre of the outer tuners to make it easy for me to find the reference for them even after I’ve removed the face of the headstock. I can then connect them with a line and add the marks for the four other holes to give me the references I need to drill the holes.

With the holes marked up, it’s then over to the pillar-drill and using a 9mm forstner bit that I had to buy for this purpose (my others are all multiples of 5mm), 6 new holes are produced:

In that photo tuners are not perfectly lined up there because they’re not attached properly, and the hole being uniformly wide means they’re not locked into place the way a stepped hole would be, but once everything is tightened down they’ll be fine. It’s worth noting that the 9mm hole is fractionally too small for the inserts that go on the top, but I can use a reamer to widen them slightly when I get to final assembly.

Now for a proper workout

In last week’s post I had this photo, which shows how I mark up the neck ready for carving:

I said I’d explain what the lines mean this week, so I guess I’d better do that. I learned this technique early on my guitar building from this video by Fletcher Handcrafted Guitars, and I’ve used it with a little variation on every neck I’ve built.

The process is:

Draw centreline down the length of the rear of the neck
Mark 1st, and 12th frets on the
On the 1st fret draw a mark 5mm either side of the centreline
On the 12th fret draw a mark 10mm either side of the centreline
On either side of the centreline draw a line the lenght of the neck that connects the marks you’ve made on the 1st and 12th fret - this gives you some tapering lines that you can see above.
On either side of the neck draw a line along the length of the neck 2/3rd way up from the rear (which should be just below the fretboard glue line)
On the rear again, mark headstock bounds - I do this by extending the sweep of either side of the headstock as it angles in towards the neck at around the start of the fretboard.
Measure 85mm up from heel of neck on the rear centreline and draw a line across the width of the neck
Measure another 35mm up the centreline and draw a triangle with that last line you drew (again you can see that in the above photo).

The triangle is where the carve of the neck stops before the neck joint with the guitar’s body. These days I normally don’t have a squared neck pocket like The Chuncaster has, but make it more curved for better ergonomics; for those guitars I don’t use that triangle method, but put the neck into the neck pocket, trace the sweep of the joint line, and then extend it 10 to 15 mm and that becomes my no-go area rather than the above triangle.

The line on the side and the tapered line on the back between the 1st fret and the triangle is where you’ll carve out to a facet using a rasp, which you can see in this photo of the neck after I’ve completed that stage:

Once that is done, you then carve at a more shallow angle to the centreline to have very rough curve here. You can control how chunky the finished neck feels by adjusting those 5mm and 10mm marks you make at the 1st and 12th fret - those values above give quite a skinny neck by the time you’re finished, and so I add a couple of mm to both if I want the neck to have a more baseball bat like chunky feel.

At this stage using the rasp I start to work the facets out into a more continuous rounded over shape. The one thing these marks don’t help you with is that the depth of the neck should taper also, with guitar necks being more narrow at the 1st fret than at the 12th etc. So at this stage I use callipers to start measuring the thickness of the neck, and get the values down to a reference neck I know I’m happy with, taking more off at the 1st fret end, then smoothing it out to the 12th fret, and using a ruler to check how well I’ve made the transition.

Then when I’m roughly happy I’ll use a band of fabric backed 80-grit sand paper pulled over the neck to start smoothing it out to a curve, before working up to finer grits to get a smooth finish.

All through this process from the moment I start rounding out the facets I’m regularly taking the neck off the stand and holding it like I was playing it and checking whether it feels right to me. Measurements in millimetres are an important guide, but in the end it’s all about whether it feels good in the hand that is the real measure of whether the carve is right or not. During the sanding stages I’ll also stop occasionally and check for rasp marks that still are showing and highlight those in pencil or chalk so I know where to focus my attention:

It’s a lot of work, and I usually end up splitting the process over two days, as the bulk removal wears me out to the point where I’ll not have the mental focus to do the finishing detail to a level I’d be satisfied with. But it is ultimately very satisfying that you can start with a few crude pencil lines, and end up with something so smooth and beautiful:

It helps that the birds-eye maple is also a beautiful material to work with - this neck is going to look amazing once oiled. That’ll do it for this week - next time we can cover the fretwork!

Some Time in the Shop

Fri, 08 Aug 2025 09:04:03 +0100

Picking up a story of a neck build from a little while ago. This neck is actually a new neck for the second guitar I built, The Chuncaster. That build I made for my brother, and is the first guitar body I made myself rather than buying, but I still bought the neck, which is generally a strategy I recommend to first time guitar builders: you can make a lot of mistakes on the body and still have a great guitar at the end, but building a guitar neck is a lot less forgiving. That said, I started building necks not long after, and I’ve always wanted to go back and make one for my brother’s guitar; finally after all this time, that’s what I’m doing.

Resuming from gluing

When I last posted about progress I’d been trying out my new technique of making jigs by 3D-printing them, building a device that helped with shaping the headstock on a guitar neck. The end result of that work was that I’d cut the profile of the neck into a single piece of maple and got the headstock level cut in, and we finished that post with me setting the fretboard being glued on:

At this point I get nervous about whether I used enough glue and enough clamps to get a uniform seal all the way around the join. I’ve only once failed at this, but it still smarts a little to think about the wasted effort to get to that point of failure. Building a guitar neck is an exercise of investing more and more time into a piece and each next action could void all that and send you back to the start - another reason I tend to caution new builders away from starting with a neck unless they’re already seasoned woodworkers.

Thankfully with the clamps off, everything was looking good:

One thing I didn’t mention in my previous post was that for the first time since I started building guitars I was using a different brand of wood glue. When I started building guitars I bought a large bottle of gorilla wood glue, which has done me since. I picked this because it was what was used in the workshop at the community workshop I was in at the time, and it seemed to do the job. And indeed, has done so ever since. However, in my current workshop, which I share with two professional luthiers, both use Tightbond, which they both swear by, and so I’ve switched over. Partly, because if they are happy with it, then there is probably a reason, but mostly because it makes it easier in a shared space when people use the same things so if you run out half way through a job you can borrow from the others in a pinch.

I bring this up as I was asked by someone at Makefest this year about whether I used fancy glues or not, and the answer is no, and as low-brow as it might seam, I had nothing against gorilla glue and so if that’s what you’ve got to hand, go for it. I did note with the Tightbond it was a little less runny than Gorilla glue, which made it easier to apply and there was a little less sliding around of the fretboard as I tried to clamp if down, so where is a benefit to the switch. Tightbond I think dries quicker too, but given I tend not to have consecutive days in the workshop often, that’s less of a practical concern for me than my professional workshop-mates who are on the clock. So Tightbond does seem better, but you can totally use wood-glue you got from your local hardware store for this purpose if that’s what you have to hand.

I did chat to a long time builder of arch-top guitars, and he strongly advocated using fish-glue, as it was easier to heat treat to undo, so made repairs on older instruments easier: I guess this is the kind of insight you get after building for several decades - and might also have saved that failed neck I mentioned before, and I could have more easily delaminated the fretboard and tried again. So, there is better glue, but again, don’t let not having perfect stop you getting started.

Tidying up

With the clamps off, the next step is to trim the fretboard to the profile of the neck. The first stop for this is the bandsaw to remove as much of it as possible, before then doing a finishing pass with a palm-router (a hand controlled router being a finishing tool, not a bulk-material removal tool).

Even after getting rid of a lot of the material with the bandsaw, you still end up with an impressive amount of mess to clear up after routing, and why I still got into my overalls for a seemingly simple task!

With the fretboard now flush on both the sides and the heel end, I still need to deal with the overhang at the headstock end, for which my 3D-printed jig makes a brief, penultimate, return:

You can see I was going very slowly in my cut by the burn mark there - apologies to my past wood-shop mentors. Normally a burn mark like this is something to avoid, but in this case that material will be removed when I round over the headstock later, and for me the greater risk that I was trying to avoid was pushing through the wood, coming out the other side of the overhang, and then hitting air: I was worried if this happened the piece might slip and I’d end up cutting into the rest of the neck, which would be bad. Thus I took it slowly and cautiously, but with a bandsaw that does lead to friction burns like this.

Fretting, part one: cutting the slots

As I mentioned in the previous post on Makefest, as it was the tenth anniversary I’ve been reflecting a lot about how far I’ve come in that time. When I made my first necks I needed a way to mark out the position of the frets so I could cut them, and being as I was based in a makerspace, I came up with my own way do doing that using a laser-cutter. Using the laser-cutter had the added benefit that I could do more than just mark the position, I could etch it enough that my fret-saw could find its groove, literally. I’m still quite proud of that maker hack, but these days I reap the benefits of being in a shared-workshop with professional luthiers show don’t have time to hand-saw each slot, and have come up with their own way to speed up the process using a table-saw:

You can see the neck is stuck (using double sided sticky tape) onto a metal plate with a series of notches on it, and these notches will let me align the neck up with a pin on the sled that you see on the table saw in just the right position for each fret. We have a different metal template for each of the common scale-lengths guitars come in. And with this what was a very careful hour of saw cutting (as saws like to jump out of grooves), is just a few (still careful) minutes with a table saw:

Again, as with the glue, do you need this to make a guitar? Absolutely not: a ruler, a pencil, and a saw will do the job perfectly fine, just it’ll be slower and a little more stressful perhaps. But at the same time I do feel this is one of the better luxuries I have access to in my current workshop :) The one downside of this otherwise wonderful setup is that the templates don’t include a notch for the other side of the nut slot, which means I need to shimmy the template along a few mm and hope I aligned it correctly before I cut that.

Inlays

With the slots cut, I can now mark out the position of the fret inlays. For those readers that don’t play guitar (of which I know there are some) the inlays that can be seen on the face of a guitar neck and, somewhat less obviously, on the side of the neck, are there to help the player work out where they are on the neck. The double dots on teh 12th fret tell you you’re one octave up, the dot on the 5th fret tells you that’s where A is on the low string (in standard tuning), and so forth.

I have to confess that I’ve stopped doing dot inlays on the face of the fretboard, as I prefer the uncluttered look that shows the wood more, but I’d be remiss if I didn’t point out that this aesthetic choice also means I reap the benefits in not having to drill recesses for them and glue them in. But what I can’t skip is having the side dots on the neck, as those are the ones that are more obvious to the player.

For this new neck for the Chuncaster I decided to follow on the trend I started with Hästen, and do metal inlays on the side. For Hästen I’d used a silverish metal, but Chuncaster is somewhat notable for having all gold hardware, and so I had to use a goldish metal for these ones. Having never used this specific material before, I did as ever and tested it on some scrap before using it on the actual workpiece:

There’s two things I primarily testing for here: what’s the right drill size I need to use to get the closest fit to the dowel (otherwise you can end up with obvious glue lines), and also how hard is this metal when it comes to sanding down. With the silverish one on Hästen it was quite tough, so I had to be careful not to over-sand the wood around it whilst levelling the metal part. Thankfully this one seemed no worse and having used it once I found levelling it worked fine so long as I took care.

With that testing concluded satisfactorily, I could set to drilling the inlay holes:

You can see that I’m installing the inlays on the seam between the fretboard and the main body of the neck, which is the most common way to do, but actually not what I meant to do :) Following the seam has two advantages: you know that your inlays are in a line, and you know that despite not having rounded over the fretboard yet that you’re not too close to where the eventual edge of the fretboard will be. The downside though is it’s really really obvious if you don’t drill perfectly on the line, and being wood, nothing is infinitely perfect, as drill bits lite to follow the grain and so forth. Thus on the last batch of guitars I’d take to putting the inlay into just the fretboard itself: it means a bunch more measuring and marking, but means you can have some tiny amounts of variance in the hole position without it being so obvious. But I forgot, and just went with how I have done it on most my builds out of habit - doh! Thankfully this hadn’t be specified by the recipient of this guitar, so no problem here other than I feel a bit silly that I changed my technique.

With the holes drilled, I can now insert the dowel, using a little superglue to secure it, and some superglue-activator to stop me sticking myself to the workpiece:

And the end result:

At this stage I lightly filed down the faces of the dowels so they weren’t sticking out too much, but I didn’t go all the way, as the entire neck will be sanded somewhat agressively when I do the carving of the neck, and so I’ll let that job serve the purpose of getting the inlays flush. On the topic of carving, whilst I have the neck on the bench with a pencil and ruler to hand, now is also the time to mark up the guide lines for where I’ll need to carve the neck:

I’ll cover what these marks signify in the next post when I do the actual carving.

The final act of the 3D-printed jig

The last job for this trip to the workshop was also the final task for which the 3D-printed jig had been designed: assisting in rounding over the headstock transition:

The jig is doing two things for me here: firstly, it’s providing a guide for how much material to remove, and secondly it’s providing a flat surface I can put on the table of the spindle-sander that is parallel to the neck’s centre-line (because the sides of a guitar neck are tapered, so can’t be used for that).

A few minutes of sanding later, and I have a perfect transition sanded in, in what has to be the most hassle-free that job has ever gone.

I’m very pleased with this 3D-printed jig that serves so many different parts of the neck build in one single print. Now I’ve done this once, I know that I’ll be doing more of this, particularly for the next Älgen style build, where I struggled with several build stages due to the angles involved.

That’s it for this update, but more soon on the progress on this neck and some new designs I’m working on!

Liverpool Makefest 2025

Sun, 20 Jul 2025 11:18:31 +0100

This year was the tenth birthday for Liverpool Makefest, and I was delighted to be able to return as an exhibitor once more and attempt to explain guitar building to the people of Liverpool. I started exhibiting at Liverpool Makefest in 2017, having only just started building guitars the previous year. Whilst it wasn’t quite my tenth anniversary of building guitars, it felt a good time to do a sort of retrospective, given they asked us to do something extra this year to mark the occasion. So I went for scale: normally I take one guitar, perhaps two, or three at the most: this year I took along seven guitars I’ve built: ranging from the very first parts build, to early complete builds, though to the latest conventional and unconventional guitars. Rounding it off I also took along the amp I built and the cigar-box guitar I actually made specifically for Makefest in 2019.

Truth be told, the retrospective wasn’t the original plan. I did have plans to show something new, but nothing came together in time despite a couple of starts at various things, and so this retrospective was something of a fallback position. But as a group of twenty somethings from London once said: you can’t always get what you want, but if you try sometimes, well, you just might find you get what you need; given that I’ve struggled with progress this last year, struggling to get out of the shadow I made for myself with Älgen, a reminder of what it is I do was probably what I needed as much as anything else.

One of the reasons I enjoy Liverpool Makefest each year so much is it’s the chance to share what I do with so many people; not in a “isn’t what I do great” way, but in a way to talk about what interests them, to try inspire them to try making something, or just to share in peoples’ enthusiasm for music or making or both. The audience of people who stop by is quite diverse: some are experienced players interested in the instruments; a few are people who have tried making guitars or related things themselves and are interested in comparing notes; some are interested in learning how guitars are made or how what I do is perhaps different; and some just want to look at the shiny nice objects and move on. Whatever the motivation for people stopping by, the day is generally a blast, and I feel very privileged to get to take part in it every time.

For the last few years I’ve focussed on Älgen, which is quite an unusual guitar and I guess somewhat of an advanced build, being how it brings together a bunch of different worlds into a single instrument - and my focus has been trying to show people that there are new techniques out there for building things that in the past were only possible in advanced industries, but now accessible to hobbyists, like custom 3D-printed metal parts. But this year it was more of a return to the narrative I tried for the first few years of Liverpool Makefest, trying to show that guitar building is an accessible hobby, and that the fancy guitars I’ve shown of late are not where I started. Each year at Makefest I get a few comments like “you must be so clever/smart/talented to do this”, and whilst well intended I find that a little frustrating, as I’m none of those things and I’m actively trying to show that isn’t the case. The one skill I’ve managed in all this is just to start small and each time do a bit more than I can do, and hang out with folk that can help me when I get stuck.

The line up of guitars on the table thus represented my journey. Back in 2016 I knew no woodwork, but wanted a guitar that was like my Fender Telecaster but slightly different, so I just bought the cheapest parts I could, watched some YouTube tutorials and put together my first build. I got help from people in the local community workshop when they saw me struggling, and thus I was able to make the guitar you see being played in the above picture.

My brother Tris, then guitarist in the band IKARI, saw this and asked me to make him one similar, but he had a few differences, including using some different wood for the body. So for that one I bought the neck still, but made the body. Building the body of a solid-body electric guitar is relatively low risk, as if you mess it up it doesn’t really matter too much, and so was a great start to doing some wood work. That instrument, christened the chuncaster by Tris, is clearly not made to the level of a proper luthier, but he played it on an album, in music videos, and most crucially at a gig at King Tut’s Wah Wah Hut in Glasgow: seeing a guitar I’d built being part of a gig where 200 people were having a blast convinced me that guitar building was worth sticking with. Tris was actually at Makefest this year, lending his guitar to the table and helping out along with his photographer wife Ket who took all the photos here (thanks Ket, they’re ace!).

So it went on with the building, over the last 9 years I’ve just kept doing a little bit more on each build: the next two I built fully by hand but were still telecaster shaped, then I made another style of existing guitar by hand, before slowly moving onto my own designs, and trying to integrate more 3D-printed and laser-cut parts to align with my other interests. Each time getting stuck and finding a way forward thanks to others being generous with their time and knowledge.

I think I lose track of how far I’ve come, because it is all incremental improvements instrument by instrument, and because I share a workshop with two excellent full-time builders I know that I’ve got lots to aspire to, and so it has felt like I’ve been a little stagnated. But in trying to show people at Makefest this year that building can be an incremental thing, and it’s okay to grow over time, and then pointing to specific examples of the guitars I was able to assemble for this show: it let me feel a sense of pride in my work that normally I don’t allow myself when focussing on that next increment I need to make.

So my thanks to everyone who stopped by: I very much hope you got something out of stopping by, because I know that I did. I’ve been back in the workshop since, and I’ve been motivated to try get on with the next hurdle that is in front of me, as I can see how clearing those in the past has helped me progress. Similarly, I’ve been doing difficult or tedious bits on one of the builds I have going, but appreciating now that before these would have been a source of angst and stress, but now whilst I’d not say they’re easy, experience means I execute them much more confidently, a think I’d not have thought about without spending a day reflecting on how things have come along.

At the end of Makefest came the announcement that the wonderful folk who have shepherded the event are stepping down: Caroline, Mark, and Denice have done a great job, and I can’t thank them enough for the opportunity Makefest has given me over the years. I’ve had so many interesting conversations, been able to give my guitars over to people to play who never would have expected to get to play a custom hand-built instrument, and best of all once a year I get to stand in a library, normally a sanctum of “shhh”, and tell people it’s okay, you can definitely make some noise today.

I’ve made some friends along the way: I knew some of the DoES Liverpool crew before hand, but I can clearly remember meeting Snoof for the first time as she stopped by to try out one of the guitars, and over the years feel lucky that I’ve become part of that maker community. I remember vividly being chastised on the first year by someone for not knowing Frank Zappa’s work, something I corrected for when they stopped by the next year. I remember one of the steam-punk pirates coming around and offering what I suspect was a hip-flask of something should I be flagging (and the fact that I have anecdotes at all about steam-punk pirates is something to treasure in of itself). This year I suspect the conversation I’ll remember the most is this one:

No, I didn’t ask, I figured that’d be rude, and I like to assume they always dress like that.

Most of all I’m grateful that Makefest reminds me that guitars are fun and an experience best shared. So thanks once more to Caroline, Mark, and Denice, and all the others that have helped over the years and all the volunteer crew, I really appreciate it. I wish the organisers all the best for their well earned retirement from their roles, and will await to see what could be next year. Something I’d be happy to help be a part of make happen again if possible, though it’s hard to currently know what that might be productively that given I live 200 miles away. But it’s a unique event, and worth preserving in some form if possible.

Finally, a shout out to my other half, Laura, who every year supports the madness that is trying to cart all of the stuff from Cambridge up to Liverpool and back again, and makes sure I’m fed and watered throughout the day. In making support of others is everything, and that includes attending maker events like this!

Workshop diary: trying a new approach to headstock shaping

Sun, 23 Feb 2025 10:39:57 +0000

In addition to the bass-guitar build I’ve been writing about, I’ve got a couple of electric-guitar necks I’m building at the same time. Both of these are regular (to me) 25.5" scale-length guitar necks, so in theory nothing new here. But I know that for the bass-guitar’s neck I’ll need to change things up, as some of the techniques I’ve used don’t scale up, and so I’m using these necks as a proving ground for some new approaches. This has the double advantage of it being easier to try out new techniques on something you’ve done before, so you’re not innovating on too many fronts, and it should leave me with an improved workflow that isn’t just suitable for the bass-guitar necks, but is generally better.

My approach to building a neck

I guess on disclaimer before I go on with this. In general I write these workshop progress reports so that others can learn from my mistakes or see a new technique they might use, but I feel this week we’re going into a weird territory this is due to the niche I find myself in: I don’t use a CNC-router, but I have access to some funky 3D-printing stuff, and so my workflow for building is probably starting to reflect this odd point in the building space I occupy :) But hopefully something in here will be of interest, and I do make mistakes for you to learn from along the way, I’m just not sure how applicable these learnings will be to most other builders out there!

Whilst I said in the opening we’re talking about building guitar necks, we’re actually going to focus in this update on how the headstock is made. There’s two dominant styles of guitar headstock style: the flat style that you find on Fenders and others, and the angled style you find on Gibsons and others. There’s advantages and disadvantages to both, and as this will be a long post anyway I’ll skip that discussion for now, and just state that so far I’ve predominantly used the flat style, as that’s because it’s easier and less wasteful to build for someone with my experience and tooling. You can see an example of the sort of neck I build here:

The first stages of my workflow are going to remain unchanged: using a template to cut the outer profile of the neck and the truss-rod channel, which you’ll see me do later. This gives us a flat profile of a guitar neck in wood like so:

To start making the headstock to the right level, I then make two precision cuts with the band-saw, a shallow one perpendicular to the neck and one parallel to it along the face of the headstock just to remove the bulk of the material:

I then glue on the fretboard material and once the glue has dried I trim it to the profile of the neck:

And then finally I use a spindle-sander to round-out the transition:

Hopefully that gives you the general idea of my method.

The challenge is doing those cuts and the rounding on a piece of wood that doesn’t have nice edges parallel to the centreline for resting on the beds of machines like the band-saw and the spindle-sander. The sides of a guitar neck taper, it being thinner at the nut (for me 42 mm) than it is at the heel (for me 56 mm), which takes away the obvious contact point.

In the past I’ve solved this by using a full-neck-sized MDF jig made on the local maker-space’s CNC router to hold the neck for these operations. You can two such jigs in action in this picture from the archives:

These simple jigs, made from an 18 mm thick sheet of MDF, do get the job done, in that they provide me a straight edge that I can rest on the bed of both the band-saw and the spindle sander whilst I make these cuts, but they’re not ideal. Firstly they’re quite bulky to store for all the different variations I need, and most vexingly 18 mm is not deep enough to be super-stable on a spindle-sander without some extra material clamped to it. Again, from another old post:

Not particular proud of the above, but it did work after a fashion :)

Given this method isn’t optimal, and given how much bigger such a jig for the bass-guitar neck will be, I’ve decided I want a different solution. One that has the following properties:

Indicates the various depths and positions I need to cut on it, including the rounding radius for the headstock transition.
Is more stable on the spindle-sander without adhoc extra things being added to it.
Is smaller than the entire size of the neck.
Can be made without a CNC-router…

That last one is a Michael quirk obviously, as someone who is more at home with 3D-printing than CNC-routing these days, but it’s also a fun experiment. Making jigs out of wood usually involves some compromise due to the fact that say MDF comes in sheets of set thicknesses, and so I was interested in what I’d design if I was freed from those constraints. And if this trial works, then perhaps I’ll be emboldened to make more jigs this way and I can try making angled headstocks with scarf-joints by making it quicker and easier to build jigs.

Right, with the scene set, let’s take a look at progress on the guitar neck I’ve been building to try out a new approach!

From nothing to something

Mostly I’m including this early part of the neck build process here as it nicely emphasises what I discussed last time about using the router in different directions to avoid tear out. We’ll get to the jigs shortly!

To make this new neck I’m working on I started start with a 1" think plank of maple I thickness it down to 20 mm. By the time I’m finished the wood should be closer to 19mm thick, but it’ll get sanded and cleaned up as I work on the neck, so I thickness it a little over size. Once I have it the right thickness I attach a template with the profile of the neck and cut off most the excess using the band-saw, leaving a roughly ready necks:

Once that’s done, I get the neck flush with the template using a hand-router. As described for the body I cut last time, this involves cutting from both sides so as to avoid tear-outs as the grain direction changes, but it’s a little easier as the neck is less deep than the body so I just need a single pass from both sides for the relevant sections.

You can see in the two photos I’m using a different router bit: they’re the same diameter, but one has the follow bearing on the top, when the router is cutting from the side with the template on, and the other bit has the follow bearing at the end, for when I’ve had to flip the workpiece over and the template is now on the rear.

The new jig: Version 1

Here you can see the CAD model for my first attempt for a new jig to meet the above design brief:

You can see how it has depth guides to let me how how deep to cut the headstock relative to the rest of the neck, and guides me where to spindle sand the transition to. It also has a clear line indicating that first perpendicular cut I’ll need to make when removing the face of the headstock. If we look at the rear, I’ve effectively built in that extra platform that I kludge in before, but just contoured a little to make it easier to hold on to:

It’s worth noting that the size of the jig is constrained by the size of the bed of the 3D-printer I’m using: the bed is 256x256 mm on the Bambu Carbon X1C. So whilst I did want a smaller jig as part of my motivation, the upper limit of what I can have is constrained for me. I was a little worried about balance whilst using the jig, which is why the jig doesn’t just centre on the headstock, I’ve also supported a section of the neck to try make natural instinct of it to tip over less. To be clear, it will tip over left to its own devices, but the more of the fretboard area of the neck I include the less I’ll have to fight that.

One printing session later:

One thing that wasn’t shown in the CAD model was the supports that are added when printing: because of the flared out base on the rear the printed software has to put the supports there to make everything level. I printed it in this orientation so that there would be no support material in the areas where I’d place the guitar neck, thus ensuring smooth contact points that can’t damage the wood. To be fair to the Bambu printer I’m using and its software, I’ve generally found that supports come away cleanly, but that’s not always the case for some printers, which will leave raised material where the supports join, and I just didn’t want to take that risk.

The supports did indeed come away clean, and it had a nice fit when I tested it with the neck in progress:

The print feels plenty sturdy for workshop abuse even at the default print settings. However, at this point I found two things I didn’t like about my new jig:

Although the wider base did make it nice and stable for use with the spindle-sander, it was foolish of me to taper it to a narrow point as I did, as that means on the bandsaw we just have a very thing edge pressing against the fence 🤦🏼
As mentioned, the first cut I’ll make is the one perpendicular to the headstock face, and because of the size and shape of the new jig, I found I couldn’t see the contact point of the blade when everything was in position to make the cut. Given how unwieldy this set up is anyway, that made me nervous to actually use this new jig.

So, whilst the idea I felt still had promise, this design wasn’t good enough, and so it was back to the drawing board (or the CADing computer I guess).

This is why 3D-printing was originally more commonly described as being a tool for “rapid-prototyping” (before the quality got to the point you’d actually use 3D-prints for production): having learned these lessons it didn’t take me long to re-spin my jig design and try again.

The new jig: Version 2

With all that in mind, I updated my design to address both of those problems. From the front:

And from the rear:

You can see that I’ve added a cut out on the top surface to allow me to see where the blade will make contact with the wood, and on the rear I’ve added some fins that will sit flush with the fence on the band-saw. I also added some holes speculatively (which is a fancy way of saying I didn’t bother measuring things like I should) should I want to use some clamps with the jig.

Another print session later, and I have version two in the workshop ready to test:

The first observation with the new print is that although I didn’t change the recess for the neck, it’s slightly different in size to the last one: 3D-prints are known to have small changes in sizes depending on things like temperature and filament variance, so it’s not the biggest surprise. It’s a small enough change that just some masking tape is enough to remove the slight wobble it caused:

With that fixed, I can now confirm that the new print sits better on the fences on the bandsaw and I can see the contact point with the bandsaw blade for the first cut (and you can see how without that gap the contact point would be obscured):

And in the end those speculative clamp holes did come in handy, given how short the fence for perpendicular cuts is on my band-saw. They’re not ideally positioned truth be told, but good enough to use:

And so with all that, I could finally make the first cut:

All this for a single cut thats less than 5 mm deep! :)

With the first cut done, I could now make the longer cut that removes the rest of the material for the face:

You can see here that the blades that I added to the back are the main guides here for this cut, and they worked a charm to keep things straight, and so a minute or two later I added another headstock blank to my collection:

And we have a level headstock:

However, there was a problem with the cut, and the root cause sits somewhere between design fault (i.e., past me was to blame) and user error (i.e., current me is to blame). You can see in the above photo, and more clearly in the following photo, that the cut was not particularly clean. I mentioned in the last post that the band-saw wasn’t cutting cleanly and I wasn’t sure why, but I had since refitted the blade and tweaked the tension properly and it was cutting nicely before I did this, so I don’t believe it’s the band-saw’s fault. I think the root cause can be seen in the following photo, where you can see the marks in the wood extend into the plastic of the 3D-print:

Basically the blade was too close to the 3D-print, and as the blade comes down to make the cut it’s grazing the surface of the print which is causing the blade to deflect out a bit and wobble. We can put this down to user error in the first instance: I didn’t align the blade and the jig correctly. Because of the limitations on the size of the jig, it doesn’t go all the way to the end of the neck, so I’d drawn on with pencil the cut line all the way to the tip of the neck, but the blade is thicker than the pencil line, and I failed to allow for that.

The design fault is why did I have to do that at all? There’s a UX error here: because I’d clamped the neck to the jig to hold it in place, freeing up my hands, it never occurred to me to remove the neck from the jig to test my alignment. The moment the neck was clamped into the jig that was it, they were one in my mind. Had I not had that in my head, then I could have ran the jig along through to check clearance. The other design fault is I limited myself to just a single printed part: although the size of the printer’s bed constrained the size of the jig, I could have made an extension as a second print to cover the end of the headstock and give me a guide.

All good learning for next time. Thankfully the damage to the neck isn’t that bad, and it’ll sand out, which will be a form of penance for me.

Rest time for the jig

The new jig having proven its worth (but perhaps disproven my worth), it now gets to wait for its final act in the building of this neck, as the next step was to glue on the fretboard:

It got glue added, and then clamped up, ready for my next trip to the workshop, at which point we can see if the jig will fare as well on the spindle-sander as it did on the band-saw:

Workshop catchup: Bass edition

Fri, 03 Jan 2025 10:39:57 +0000

In the previous update I talked about the process of designing my first bass guitar, my Delfinen Bass, looking at how it differs from a regular electric guitar design. This set of notes will catch us up on how the build of that has got going.

Before we start in the workshop side of things, a quick update on the CAD model, as it’s had a couple of tweaks since I started it. On the front:

Visually it’s had a change as the roasted maple wood, which you’ll see below, is quite dark, so I felt I wanted to work that in. Pickup wise I’ve always wanted to try out the Lace Aluma style pickups ever since I saw Philip McKnight demo their P90s on his youtube channel. The pickups have an unusual design which don’t use any copper windings but are still passive pickups, which I find intriguing, so I plan on giving them a go on this build. I feel this style of pickup would work quite well on an Älgen style guitar, and so this is just an excuse to try them out here.

More importantly, before I could hit the workshop I needed to fill in a bunch of details on the neck and body designs that impact the actual early build stages, which starts with some bits visible on the back of the guitar:

Here I added non-visual but technically important bits like the mounting posts for the screws in the cavities; not exciting, but I need these all in place before I can make the templates for cutting the wood. Similarly I added the cavities in the neck for the graphite rods that will go in the neck alongside the truss-rod to deal with the extra strain on a bass from it’s chonky strings. You can see these here with the fretboard hidden:

At this stage it’s not just a matter of drawing them into the CAD model, I do need to know how I plan to make these cavities: what tools will I use, will I need any new bits for the router, what will be my reference points for any cuts, and so forth. You can draw anything you like in CAD, but that doesn’t account for you being able to actually make it, so as I’m filling in these details I need to also work out how I plan to build it.

What you can’t see is that I also had to do a bunch of fiddling with my CAD in Fusion just to get my design into the right state for me to build templates from my model. Fusion 360, the CAD tool I use, is based on the concept of a timeline, which is all those icons you can see at the bottom of the screenshots I’ve posted; that’s the history of all the operations I made to build the design. The order in which you do these can have important consequences, and I often come slightly undone as when building a design I often add early on visual references for myself so I can get a sense of how the instrument will look, even if they’re not technically needed in the CAD model. By adding those early I tend to pollute the timeline with things that get in the way later.

For instance, the rounding over of the body edges, which I’ll do by hand using a palm-router with a round-over bit, isn’t technically needed in the CAD, but it helps me see what the finished guitar will look like. But by adding that feature in CAD, if I want to say make a 2D-template based on the front of the guitar, Fusion will account the rounded over bit, and so my template will be too small being based just on the actual flat surface in the CAD model. If I added all the visual bits and the end I could just disable the last few bits of the timeline and all would be good, but because I do them before say I add the cavities for the pickups, I can’t readily do that. So I spent an hour or two having to unpick my design to “fix” all this.

I shouldn’t know better than this by now, but that needing visual cues as I go along is super important to my design process, so I end up having to do all this tidying at the end.

Materials

Whilst I was doing this second round of design work, I had already secured the materials for this build. My plan was to make the body with poplar, due to its light weight, and the neck out of roasted-maple, again as maple is strong and roasted-maple is lighter weight due to having less moisture in it. Fortuitously for me, my workshop-mate Matt had a spare poplar body blank he’d prepped but in the end didn’t use, which he was happy to trade me:

He’d even done most of the prep, which was great news. Truth be told, workshop time has been scarce of late, and so I’d decided that in order to speed this build up I’d just pay extra for a one-piece body rather than the usual jointed two-piece body approach I use (which is more affordable). Thankfully Matt’s offer of materal also meant I skipped the jointing stage, so I was delighted at this bit of good fortune.

For the neck, I ordered the wood, but wanting roasted-maple I had to switch from my usual wood supplier, Exotic Hardwoods UK, over to another supplier of instrument woods, David Dyke Luthiers’ Supplies. After putting in my order, a few days later a nice parcel of wood arrived:

The wood on the left is the neck and fretboard for the bass, and the wood on the right is for a regular guitar neck. Interestingly both necks are roasted-maple, but clearly one has reacted quite differently to the roasting process.

With that, I have all the wood ready to make a bass. You can hopefully see the resemblance between the material and its final form:

Template time

I was tempted to try doing some of this build on the CNC-router to speed things up a bit, after Matt offered to let me use his, but given I knew I’d end up doing a bunch of the work on the bass over the xmas period when he was unlikely to be around, I opted to just go with my usual approach of making laser-cut templates, and then using a combination of band-saw and palm-router to get the wood into the right shape.

One catch with this for the bass is that the neck is big enough that I can’t use a regular 600x400 mm sheet of acrylic for making the templates as I do for the regular electric-guitar necks, I had to step up to a 1000x600 mm sheet, the next size up, which is notably more expensive that simple math would lead you to assume. I’m also glad that the local maker space got a slightly larger laser-cutter a couple of years ago, as this sheet wouldn’t fit in the older one!

Unfortunately cutting all this didn’t go as smoothly as I’d have hoped, as the laser-cutter seemed to be having power issues. Often laser-cutters tend to lose power if they’ve been heavily used and the lenses/mirrors have got dirty from smoke and not been cleared, but in this instance there was something more odd going on: the power seemed to fade over the duration of the cut. You can see it in the picture here, where the start of the cut is more clearly made than the end (it’s going in a counter clock-wise direction).

This lead to the first attempt to cut a neck-template not working, as it’d not properly cut through the material at the start/stop point, despite me going over it a second time after I spotted the issue, and it made the top of the template ragged:

Whilst I could try sanding that smooth, I’ve learned many times over that any marks in the template show up in the wood and are then a pain to sand out, so I opted just to re-cut this one. But what should have been a fairly straight forward job at the laser-cutter for an hour turned into a half day of prodding and fighting with it.

I often get asked why I don’t do more of my guitar building with CNC things, and whilst partly it’s because I enjoy the manual process, it’s also partly down to nonsense like this. If I made more guitars and had my own dedicated machines that I used regularly, then whilst they’d still have issues occasionally, in general using the machines would be more of a known quantity. But using shared machines like this I just can’t predict how they’re going to behave, and so I prefer to replace that risk with slower but more predictable manual processes. One day I’d like to go back to CNC-routing for some things, but not whilst I do so few instruments each year.

Anyway, machine frustrations aside, I eventually turned one large sheet of acrylic into the templates I need to shape my bass guitar.

This also means for the first time I get a sense of how the guitar is at life size:

Quite pleased with how it feels at full size. It’s not always the case that what looks good on screen will look good in real life, so it’s a nice sanity check at this point before I start making hard-to-undo changes to my raw material.

The acrylic is actually clear, rather than white, that’s just the protective film that is removed before use. I use clear so that I can see the wood grain under the sheet and check that it I’m happy with the bit of the wood I’m using. I also etch on a centre line on the template that I can then line up with the join line on the wood.

Before I can use the templates there’s one final thing I need to do, which is countersink the mounting holes for the screws, so they don’t interfere with the router. Alas, the laser cutter I have doesn’t have a magic multi axis head, so I do this on the pillar drill.

With that done, it’s finally time to get started!

First cuts

Before I can cut the profile with the template, I first need to ensure the body blank is perfectly level and at the right thickness. Although in theory the joint faces should be at 90˚ to the front/back they faces won’t be perfectly so in general (particularly if you used a hand-plane to get the joint faces flat), and so the body will have a slight V shape to it most likely, which isn’t great for attaching templates.

The first step was to remove the excess glue along the seam:

With that tidied it was over to the thicknesser to get it properly flat and down to the thickness I wanted. Matt had already thinknessed the two halves before he jointed them, so I didn’t need to do too much here: the pile of wood to the side you see here is from another workshop-mate working on the table router, not from my thicknessing!

Now with the body prepped, I can finally attach the template, taking care to line up the centre line and check on both sides that I’m going to avoid any features in the wood that I don’t want. In this case there’s a knot on the back towards the bottom edge, which is why I mounted the template quite near the top edge. For those curious about using screws to mount the template, the screw holes are all in areas where later on the wood will be removed: the top one is in the neck pocket at the lower two are where the pickups will go, so no evidence will remain of these screw holes by the time the body is finished.

With the template on, the next step is to remove the bulk of the wood using the band-saw. Normally for bulk material removal I’d use a chunky blade eat through the material, but because of the curves on the body, and because poplar is relatively soft, I opted instead to switch out the blade for a smaller one that is better for following curves. Changing a bandsaw blade is a relatively quick process on the Record bandsaw I have, and so it’s something I’m happy to do for each bit I work on if necessary.

A few minutes later and I had my body roughed out:

You can see that the cut isn’t the best - given I just put on a fresh blade and thought I’d set up the guides right I’d have expected a smoother cut, so I’m not sure what I did wrong there - I set the tension appropriately and adjusted the guides, but clearly the results are not great. Thankfully I’m just roughing at this stage, but that definitely isn’t acceptable for other jobs I used the bandsaw for. After doing this I returned the bandsaw to having a heavier blade on it, as my workshop-mates mostly use it for rough cutting, and that was cutting cleanly.

Mystery aside, the band-saw stage is just to get the bulk of the material off, but not to get the actual body down to size. For that I use a palm-router with a pair of follow-bits that have a guide bearing to follow the outline of the acrylic template exactly. Routing though should generally be seen as a tool for removing just a small amount of material, hence the band-saw stage first.

This is where I make a big mess similar to the pile next to the thicknesser, only I’m doing it by hand rather than with a table-router. From the picture you can see that my router bits aren’t deep enough to cover the body in a single pass, I’ll need to do a first pass with the template on, then I can remove the template and use the guide bearing to follow the upper cleared part when I do the lower half.

Above I said I had a pair of router bits, and some might be wondering why. For most things I use a follow bit, which has a bearing at the top which sits against the acrylic template as a guide as I cut around, tracing out the shape. But just doing this in a single pass has a significant problem: if you want to avoid the wood tearing, you generally want to cut as you “step down” the wood grain. But as you trace around the edge of the body roughly half the time you’ll be stepping up the grain lines. What on earth am I talking about, let me try show you with this annotated picture:

I’ve marked small black ticks around the edge of the template. If you go around the template anti-clockwise (as you would do with a hand router), and observe the grain lines, at each tick the grain goes in a different direction compared to the line I’m cutting. Now a router bit will spin in a clockwise direction (as indicated by the blue circle arrow thing), and so in the blue section where I’m “stepping down”, the cutting edge pushes the grain into the body, giving a clean cut, but in the red section, if I kept going with the same cutting tool going clockwise as before, then the router blades will lift the grain, causing tearing.

Hopefully you can see from that sketch how the blade on the router bit will lift the grains on the red cut. You can see the effect of tearing clearly here on the bass body as I progress:

This is that transition point between blue and red indicated on the annotated diagram, where I’m ramping out my blue cut and I have to go slightly into the red zone. Don’t worry, this ramp out area will be removed when I do the red pass, but it hopefully illustrates that cutting against the wood can cause quite a bit of damage.

Because I can’t change the direction the router cuts, the solution here is that for the red section I use a second router bit that has the follow bearing on the bottom, and I have to flip the workpiece over and cut from the other side, so that the red section is being cut in a grain friendly manner.

So basically to cut the body like this requires four passes: a blue and a red pass on the top half, and similarly on the bottom half. This is also why it’s important to get as much material off with the bands-aw as possible, as otherwise the excess material will foul on the shaft of the router bit I’m using for the first red pass.

It’s a lot of messing around, but on a wood as soft as poplar the tear out damage if I didn’t do this would be quite notable, as you can see. And that would then lead to sanding, and I hate sanding. Even on maple for the necks, where the tear out is much less because the wood is tougher, the little you do get is harder to work out, as the wood is much tougher :) Basically a little more time at this stage saves a lot of effort later on.

Here’s a pic towards the end when you can see I’ve just got the last pass to do:

And finally the outer profile is completed!

Well, I’ve done the outline :D I still have to route out the neck pocket, the control cavities, and the pickup cavities, but that’s a job for another day, as I’d also done a guitar neck at the same time (which I’ll document in the next notes), and I was getting tired, and you don’t want to mix tired and routers. It’s also hard to drink tea in all this PPE.

So with that I’ll sign off here, and we can talk about necks in the next set of notes!