A week in the shop

There was no direct guitar progress this week, with the CNC Router still out of bounds for guitar work. We did fix the USB port failure on the current adapter so it is back in operation, but it’s on the old control electronics so I refuse to put material on it at the moment that I can’t afford to replace, and after two failed attempts to cut a pick guard on that, I’m not going to throw good money after bad.

As a result I concentrated on mostly contract work this last week to let me get ahead of that so that when the CNC router does get it’s new brain I’m ready to take advantage of it.

So, progress on that new brain. Two replacement CNC Controllers have now arrived: the DM500, a new pendant style controller, and the Masso, a more advanced unit that has lots of bells and whistles, but still has the advantage of not being a PC. Adrian, one of the Makespace directors (with minimal help from me in the form of switch pressing occasionally) wired up the first of the new controllers to our test jig (shown below with the old Grbl controller we tested initially).


Whilst I like the idea of a controller with lots of bells and whistles, in the context of Makespace that’s not necessarily optimal. Whilst there’s a small set of users like myself who are pushing to the more advanced end of what can be done, there’s many more users for whom this is their first bit of industrial control kit, and as such whatever solution we get for Makespace has to be robust and easy to comprehend. becoming a trainer on the CNC Router has been quite an eye-opener in terms of what the Makespace requirements are - it has to just work more than it needs to let you control things to the nth degree (no matter my personal opinions on it). At the same time though, we do have some minimum feature requirements, such as the ability to do 3D surfaces, not just 2.5D as per some solution (e.g., the Instructables X-Carve). Thus we’re starting testing with the simpler of the two controllers, the DM500, which I hope to start software testing on today.

In early December I wrote a bit about my side project to make myself a new pedal board. This was partly to fix something that bugged me about my current pedal board, and partly to move me out of my comfort zone in terms of making things and make learn a new skill. The design centred around a folded bit of metal onto which the pedals would attach, along with some nice wood end-caps designed to make it look nice for a home or studio setting. 

Screenshot 2019-01-18 at 15.09.23.png

Having designed the metal work, and consulted with a couple of friends who had done this sort of thing before (many thanks to Mark, Rob, and Adrian!) I commissioned a local firm to build me a one off prototype, and the metal work finally came back:


When this turned up, I was very excited to see how close what I got was to my design: it’s one thing to design an object, it’s another thing for it to be actually manufacturable. I might have designed the radiuses too tight to be folded, or relied on tolerances that were not feasible etc.

Initial impressions were that it looked very much like the design I sent them, but the question is whether it’d mate with the other parts needed to make the entire thing. The main interaction points are the wooden end-caps, and in the rear is a power socket that just passes through to an internal kettle connector, so you can mount your pedal power block under the board but still have an accessible on/off switch on the outside. There are also rubber feed I ordered and the undersides were designed to fit those.


As you can see in the above picture, the power socket and the feet mount perfectly, and the board sits nice and level with the feat on. Very pleased with this. I then cut some mock end caps out of MDF just to see how those would work (demonstrating that our current CNC Router controller knows when I’m using cheap material, as it let these ones cut without incident).  With those on it looks the business:


Not bad for a first go at designing with folded metal!

However, although it looks the business, there’s a few things the prototype indicated I need to tweak. The main one is that the front fold isn’t quite perfect, and so having the end-caps perfectly flush with the metal doesn’t work aesthetically, as they highlight the slight variance. I’m informed that folded metal like this is never absolutely perfect, and you have to allow for little variations like this - sometimes with a mallet ;) However, here I’ll try to adjust the design of the end-caps to make this acceptable in the design. There’s also a few machine marks I’d like to see if I can get removed at the production place, as although I can file them out, I want the production place to also powder coat them before shipping them to me for the planned production run. Before I respin the end-caps I’ll get the guitar cable jacks wired in and pedals mounted so I can check for any other niggles by actually using it.

But for all those niggles, I consider this Achievement Unlocked! I’ve never fabricated things in metal before, and this turned out better than I had any reasonable right to expect for my first go. This opens up the door for other things down the line: for example, if I wanted to make more amps I can now make my own chassis to match my design rather than (as I did last time) building my design around an existing chassis which changed the vision I had quite substantially.

I did find a little time to keep plugging away with my trying to re-learn electronics. I’ve been struggling to get my IO-expander chips to play ball, but didn’t get very far due to my push on contract work. I did however make some strip boards for the relays I want to use with my pedal switcher, as they don’t mount on the breadboard directly.

A week in the shop

Apologies for the lateness of the weeknotes, contracting work has kept me rushed this last few days. But, the week notes must go on: here’s what I did last week.

After the previous week’s setback with the CNC router, I decided to just use the slightly damaged pick guard I had made and spent the better part of a day putting together most of the rest of The Clydesdale. This was partly just to check everything else was okay and there were no more surprises waiting in store as I brought all the bits both made and bought for the guitar together, but it was also a bit of a cathartic process for me to let me see the guitar looking like a guitar.


As is inevitable on a project that goes wrong so many times and drags on long beyond what you expected, this guitar has become something of a weight on my shoulders and mostly a thing that brings me negative emotions, as I view it as a series of personal failures leading it to be over six months behind schedule. Seeing it put together like this though helped remind me why it’s worth all the pain: the guitar just looks beautiful. The dark crimson stain sets off against the cream pick guard just right, the birdseye maple neck with the wenge fretboard is wonderfully rich in texture, and all the bits hang together just right.

I was going to solder up the electronics and string it up, but the Mustang design on which this is based is quite fiddly to do wiring wise and so I’d rather just wire it up once (when I have the correct pick guard) rather than do it now and then have to dismantle it when I replace the pick guard. But for now it was motivational to see it in this form.

A tip I learned whilst putting They Clydesdale together that I didn’t know before: how to shorten screws reliably. When you get pickups they have to come with long screws to allow for whatever mounting you might use, but I find then that they’re too long for the cavities I make in the body for the pickups, and so I have to cut them short with a hacksaw. This is a fiddly process, but a friend of mine (thanks Graeme!) explained his trick to getting reliable cuts on the screws: use two nuts to create a guide for the hacksaw blade.


This lets you get the position right and fixes it so that you don’t lose your place when you measure and then go to the vice for sawing. It also helps keep the ends neat by containing the material in something, so you don’t have ragged ends of your bolts. You can additionally sand the ends whilst in the nut for a better finish.

To do the setup for The Clydesdale and the prototype offset I finally ordered a proper set of nut files. Nut files are really quite eye-wateringly expensive, and so until now I’d been getting by with a set of the blowtorch nozzle cleaners, but whilst they do get the job done, they’re slow and frustrating to use. I think at this stage I want to be making things easier to get right rather than fighting with tools still. I do enough of that with the CNC Router :)

As I mentioned in the last weeknotes, I decided that I needed to rethink my guitar building plans for 2019. Even once the CNC Router is back online, I don’t think I’ll trust it for anything complex for a while, so I want a simple design that I can make using templates and the hand router, and nothing too complicated. At the same time I want it to be related to my future designs, so I’ve come up with the following design for my next guitar builds once The Clydesdale is done.

Screenshot 2019-01-09 at 16.20.13.png

Whilst I had bigger ambitions for the next set of guitars, I’m still quite pleased with how my more complicated design has made the transition here to something simpler, but still has elements of what I wanted. This particular example is a baritone scale length with a reverse headstock, as that’s what the customer was asking for, but I’ll also do a conventional scale length version to, with a either a reverse or regular headstock depending on the preference of who’s the customer will be.

I did a quick 3D print of this and sent it off to my brother who spec’d the baritone and his feedback was that he’s happy with the shape, which is top news. 


So the plan for this will be to make some MDF templates, then make the body and neck using the hand router and using a rasp for the comfort carves. Fingers crossed that this’ll move me forward more quickly in the first half of 2019 than I did in 2018.

It was quite nice to spend a bit of time in CAD again: it’s a technical challenge that I quite enjoy. I’m not alone in that, as the third Fusion 360 Show & Tell session I’ve organised at Makespace ran this last week, and we had a good turn out yet again with people sharing their tips and tricks on how the designed made made stuff.


It’s been quite a nice community event at Makespace, and we seem to be meeting my aim for this meeting that everyone leave with at least one bit of knowledge they can apply to their next project.

Related to that I’m looking forward to engaging with another community of makers up at DoES Liverpool: I’m going to be back up there for other reasons at the end of the month but am using the opportunity to make good on my promise to run the Fusion 360 training course I run at Makespace for their community. As DoES has a different set of equipment and different kinds of activities going on, I hope for me this is a way that I can find another set of people who’ll in future inspire my design and making process as has happened at Makespace.

The other thing I did was spend a bit more time on the electronics for my pedal switcher design. Slowly managing to remember bits of electronics, but slowly :)


Still, with other things blocked at the moment, it’s been a good mental exercise to keep me moving forward on this guitar side of the fence.

A frustrating start to the new year

I’ve been putting this week notes off as it’s been a frustrating week, but the point of these notes is to show the good with the bad, so let’s get to it.

As discussed last week, having diagnosed the latest fault with the CNC router down to the fact that it didn’t like one particular file I gave it, despite being valid, I generated the same tool path with slightly different layer heights and when I tested that on air, it all worked fine. Thanks CNC Router. 

So, I sat down to route the pick guard once again, but this time I had a more conventional failure, which left me with an almost perfect pick guard but with an annoying ding in it. When cutting out the holes for the pickups in the pick guard the the piece to be removed, which should have been secured by double sided tape on the rear, came loose, which then hooked on the router bit and managed to get in the way:


This caused the bit to deflect, making the edge there slightly less straight than it should be:


The pick guard is otherwise fine, but the professional in me sees this as unacceptable to ship.


So another failure, and I’m beginning to feel very cursed right now. The fix here is to not rely on the tape alone - I should have put a couple of screws in the bit that came loose - but for before now I’d managed to get away with it just using tape. To top it off, before I got time to cut it out again, the CNC Router pendant’s USB port broke, so now I’m without CNC Router until we replace it.

As the kids say, FML.

All in all, a frustrating week, which is why I didn’t really want to write up these notes. I’d set myself the goal to finish The Clydesdale by xmas, then by new year, and each time I hit yet another problem.

However, rather than just stop, I do have a pick guard that is the right shape and size and has the right holes in the right place, so what I’ll do this week is finish the rest of the guitar with this pick guard and then when the CNC Router starts working again I’ll cut another.

And when will the CNC Router be fixed? Well, Makespace has ordered two new controllers to try, and they’ll turn up mid Jan, so fingers crossed one of those will make us happy.

Given the frustration and given I was still trying to spend the new year at home whilst my other half was around more, I did some more playing with electronics. I actually did two years of electronics at university (over 20 years ago for those counting), but have not used it since (whilst I have done on-chip design since then, that’s a very different kettle of fish to playing with capacitors and resistors). 

My little project to help me relearn electronics is to make myself a pedal switcher, as at 6 pedals I have enough in my chain now that with it all “off” I still have a notable change in tone (doesn’t help that three of my pedals are buffered).

Thus, thanks to the advice of my friend Jason (who I helped with his Nixie Clock back in the summer), I got myself an Adafruit HUZZAH32 microcontroller and a bunch of other parts to let me prototype the bits I’d need for a switcher: an I/O expander chip to let me control more things than the HUZZAH32 will control directly, some latching relays, and some transistor arrays to let me power those. 


Early days, bit slowly getting there with the design, here I’ve got everything hooked up bar the relays as they don’t fit onto the breadboard here, I need to make a small board with stripboard with those on. 

A nice distraction from the woes in the workshop.

So, with 2019 arriving and the guitar building in 2018 not having gone anything like as well as I’d hoped, I think I need to reset my expectations. I tried to push along with the guitar building with the assumption that I could do a lot on the CNC Router and that just isn’t the case. Even with a fully working CNC Router I think it’s not a tool designed for quick one-off productions: either it’s something you labour over a lot time wise for a one-off that can be made no other way, or you want to do some test pieces and then never change your design and make multiples of them. But I’m not making enough guitars that I’ve hit that point, so I need to trying to use production techniques that suit the fact I’m making one off guitars for people or one-off test guitars to get me to that small volume production place.

I had designed some ambitious guitars for this year, which used both carved tops and angled headstocks, but I think committing to those would be foolish at this point, they rely too much on the CNC Router as a tool for making the guitar not just templates to let me build guitars by hand. Instead, at this point, I want to be making templates and using a hand router to generate the guitars, and designing to match a more traditional workflow given what I have access to. I just want to make some simple guitars this year that let me regain momentum and having the satisfaction of shipping. 

I’ll finish these two offsets, and then I’ll switch to making a sort of Junior version of the guitars I’d hoped to make for 2019: so flat tops, hand carved contours, and no angled headstocks. On one had I’m again frustrated, as in my head is this guitar I want to make and play, and now that’s a 2020 project, but on the flip side I just want to make guitars and see people play them, and that’s more important. So here’s to more wood being turned into music this year.

An xmas weeknotes

It being xmas this week there wasn’t much time spent in the workshop itself, but I did get some bits done. 

Just before xmas the replacement material to make the pick guard for The Clydesdale guitar turned up (two sheets, just in case), so I nipped into the workshop to have a go at cutting it again. Given I had all the g-code generated for the CNC router to make the pick guard I was going to just re-run it all, on the assumption that the CNC router had just glitched as was demonstrated on previous failures, but paranoia got the better of me and as a precaution I went back to the g-code for that path that failed and re-ran it on the machine with no material (running it “on air” as people like to say) just to convince myself it was a glitch (when the CNC router previously destroyed parts running the same g-code on air had just worked, strongly implying it was a glitch in the machine).

When you cut air on a CNC Router there isn’t much to look at :)

When you cut air on a CNC Router there isn’t much to look at :)

Turns out it was a good job I did this, as the machine did the same erroneous cut as before with this g-code! The g-code itself I’m convinced is valid (and others have reviewed it and said the same). In fact, when the CNC router makes the incorrect cut it is on the final layer of five (the material is 2.5mm thick and I cut it out in five passes, each just taking off 0.5mm), and it fails despite the fact that the exact same g-code is executed for each layer (I took apart the g-code file and compared the section for the failing layer with the one immediately above and it is byte-for-byte the same g-code).

I tried repeating the experiment multiple times, changing the feed speed of the CNC router, the origin offset, and even power cycling it on and off for a period of time, but this valid g-code file will fail 100% of the time.

This is both good and bad, for obvious reasons. Firstly it’s good, when something goes wrong it’s always good to be able to repeat the failure so you have a test case to show that you’ve fixed it. When in the past the machine has failed once and then never shown the error again it leaves you nervous as to when it will next strike. But the bad is that here there is no valid reason the machine should fail like this, so the machine is wrong, consistently. I suspect I can work around this, but it’s proved once and for all that the machine controller is wrong and needs replacing before I do any more expensive work on it.

Unfortunately all this meant I ran out of time to actually cut the scratch plate, so that’ll be this weeks top priority. I’ll regenerate the tool paths from Fusion and run them on air before trying to cut them on the CNC this time.

On the tool replacement front whilst the solutions I was looking at last week seem okay, I had a chat with one of the Makespace directors, and I’m mostly unconvinced any of them will work in the Makespace environment. Here we have a lot of new people come in who haven’t used kit like a CNC router before, and everything I tried just felt too flakey to be put to that purpose. Instead we’re going to investigate more embedded controllers again, such as the Masso CNC Controller.

That’s not to say people shouldn’t use Grbl and tools like Universal G-Code Sender, but just you need to understand what you’re doing with those tools more than you would with an embedded pendant solution like the one we have been using, where there is just less to go wrong, and less to reset when you get confused.

Whilst away from the shop I did sit down and try to get a couple of guitar neck designs done that I’d been meaning to do for a while: one simple and one complicated.

The first was working out how I’m going to make a neck with an angled headstock given the limited set of tools I have at my disposal. The new guitar design I’ve worked on for 2019 is based on a symmetric body design, so I really want to switch to a symmetric neck design, and I’ve gone for something that looks a little more like the Gibson/PRS style, compared to the inline headstocks I’ve been doing until now.

Screenshot 2018-12-31 at 14.01.45.png

Rather than make the neck from a single bit of wood, which would waste a lot of material, I want to make it using a single bit of wood that’s split at an angle and glued back together at the neck/headstock join, what’s called a scarf joint. In theory there’s many ways to do this, but all the obvious ones require tools I don’t currently have. For example, we have a band saw at Makespace, but it’s not that good quality wise - best for roughing up planks not getting precision cuts like I’d need for this joint. We also have a mitre saw jig, but it won’t go to as shallow an angle as I need here (10˚). 

Thus in the end I’ve come up with a slightly harebrained scheme that involves CNCing out the two parts of the neck, and then using a complicated set of jigs to glue them into position. 

Screenshot 2018-12-31 at 14.03.24.png

The positioning is key here, as there angles and shapes make it hard to align without some sort of jigs, and the obvious clamping surfaces are not straight on to the plane of the join, so there’s a risk that the two parts will slide apart once clamped, thus it all gets very complicated. I’ve sat on this problem for several months now, and I finally decided to just ask for help from a couple of more experienced wood workers I know, and they agree that my harebrained scheme probably is the best I can do with the tools at my disposal. Once The Clydesdale is done I’ll probably give this a go as a novelty. 

The other bit I did was playing with new inlay designs, just using my older neck design as a template for it. I do in part still want to replace the neck on The Blues Deluxe t-style so I can feel happy selling it on, as (as I’ve mentioned before here) the neck warped when glued and whilst at the time I managed to save it, I don’t think its up to the standard I’d expect today.

Anyway, I clearly was missing video games, as the design I came up with is a slight mashup based on game controller arrows to complement the chevrons on the strap that guitar has:

Screenshot 2018-12-31 at 14.05.46.png

I have to say, for what should have been a simple design, getting the arrow inlays to work in Fusion 360 was quite a pain. I had to specify lots of constraints so that when I copied and moved the design around the neck each time it’d stay in the right shape, and I could define a proper midpoint for positioning it between each pair of frets.

Do you want some constraints with your constraints?

Do you want some constraints with your constraints?

I’m sure there must be an easier way than what I did, but I couldn’t find one, and at least that got me there in the end. Not sure this is the final version of this look, but it’s nice not to have the same old circles for a change.

And that’s a wrap for 2018! I’ve started a year in review post, but for now let’s just say 2018 did not go the way I planned, but I learned a lot along the way. The trick will be turning that learning into momentum in 2019 and getting more guitars made, rather than just reeling from setbacks as I’ve done this last year. Good luck with whatever you’ve got planned for 2019!

A week in the shop

My aim this week was to get most of the basic bits of The Clydesdale finished, but unfortunately it wasn’t to be, and this the week turned into a lot of trying to work out how to swap the brain on the CNC Router before it crosses me again.

After the success of making a custom pick guard for the prototype offset build last week, my plan this week was to make the pick guard for The Clydesdale, and then I could set about mounting all the rest of the hardware, getting to a point before xmas where all the was left to do would be setup work. So on Monday I sat down at the CNC router and set it to work on pick guard for the Clydesdale. This one is slightly more complicated than the one for the prototype, due to the pickup mounts needed another diameter hole (meaning an extra tool change) and the slot for the three-way selector switch needing a smaller end mill: all in this pick guard required 6 tools rather than 4.


All was going well until it was doing the final pass on the outer contour when the machine decided once again it really didn’t like my design and it would be much better if it just cut the side off :(


I’m not a superstitious person, but I’m finding it hard not to agree with my friend Graeme that this machine knows when I’m working on this customer’s guitar: it’ll be okay when I do the prototype test pieces and then I sit down to work on The Clydesdale and then it decided to play up. The real pain here is that I didn’t have any more cream PVC material to cut another one, and it being the week before Christmas getting replacement was very slow. Although I immediately ordered replacement material, it didn’t arrive until Friday by which time I was getting ready to wrap up for xmas along with juggling some last minute contract work. Obviously this was hugely disappointing.

What do I think went wrong here? My current theory is that the controller being cheap means it’s struggling with floating point numbers occasionally. The g-code language used to describe what the CNC-router should do is full of floating point numbers, and computers always struggle with floating point numbers a little, as they can’t represent them perfectly. For example, you can sometimes get a computer to add 1.0 and 1.0 and it’ll give you 2.00000000001. A friend took a look at my g-code and observed on the g-code for my pick guard had the curved edges represented as a lot of very small lines rather than as a series of larger curves. If you have bug numbers than floating point inaccuracies in computers don’t matter much, but if you have lots of small numbers then these inaccuracies can start to cause real trouble, and it could just be bad luck that the controller (which we know isn’t built to the highest standards) just got its maths wrong at that point. 

I do have more material now, and I will try and get the pick guard cut again when I get back from xmas obligations, so I might yet get this guitar built before the year is out, but it’s going to be tight!

After all the other failures we had slowly been planing to replace the controller on the CNC router at Makespace. I hadn’t rushed this as my assumption was it was the more complex designs I do that were causing it to fail, but given the pick guard here was a relatively simple design I decided to refocus my week (given I was no longer fitting hardware to The Clydesdale) on moving that upgrade process forward.

Adrian, one of the Makespace directors, had built a test rig that let us try out new controllers and related software; in the summer we’d spent a day taking the CNC router apart and putting it back together only to find the replacement controller we had at that point didn’t work, so a test rig like this make finding a new controller much easier.


Adrian’s done a fantastic job with this rig: it features three axis motors and motor controllers, a bunch of physical switches for the various triggers that would be on the actual machine, and a display to show how fast the spindle would be driven. To top it all, he’s mounted it nicely up and labelled it all - a top job.

My task then is to figure out the software side of this. On the board currently is an Arduino running the popular hobbyist controller software Grbl. But Grbl only does half the controller job: it expects to be fed by some other g-code reading tool on a PC or similar device and then it will take commands from that and send them to the motor controllers etc. I tried a lot of bits of software, but the ones showing most promise are Universal G-Code Sender, and Easel by Instructables.

When assessing the software I wear two hats. Firstly, as a guitar builder I’m looking to see whether the software can cope with the complex designs that am generating these days with complex curved surfaces etc. Secondly, as someone who trains people on the CNC Router at Makespace I’m looking for a tool that is easy to get started with and will easily do the simple 2D shape cuts that most new members actually want to make. My research makes me think I can’t solve those two use cases with one bit of software, which is why now I’m looking at two different solutions. 

Easel is quite a nice bit of easy to use software if all you want to do is cut out some simple 2D shapes. It has a nice visualiser of what you’re going to cut, and when you go to machine it has a wizard that takes you step by step through getting everything ready. The only issue I’ve had this week is that although it acknowledges the z-probe on our test rig it doesn’t seem to want to use it, so I need to find a way to get that to work.

Screenshot 2018-12-24 at 12.26.51.png

Universal G-Code Sender has a much more complicated UI and relies on you knowing a lot more, but at the same time you can then do a lot more with the tool.

Screenshot 2018-12-19 at 18.42.00.png

Neither tool is perfect - and for all its faults the current controller we have for the CNC router has a robustness about it that I feel is missing from both these software solutions. If anyone has any recommendations for tools in this space they love please do let me know (Grbl based or otherwise).

That said, I’m getting to the point where I want to try these tools on the actual machine, but there’s still a couple of things I need to get comfortable before we tear the CNC Router apart again.

In unexciting workshop tasks I fixed my headphones in preparation for carrying on my guitar practice when we visit the in-laws for xmas week. As often is the case the wire had broken internally where it entered one of the ear pieces, but thankfully it has a 2 m cable on it, so I just cut it off and soldered a new connection on an unbroken bit of wire. 

Then, to try and give it better chance of surviving I replaced the old strain gauges with bits of orange Sugru; at least now I’ll be able to spot which headphones are mine in the workshop :)

The final act of luthiery before xmas was going to the pub with a couple of other local Luthiers: Matt from Fidelity Guitars and Jamie from Swannell Guitars. It was a nice evening and a reminder that I spent too much time head down in the workshop and not chatting to others to help keep me motivated. Something to work on in 2019!

A busy week in the shop

This week was fairly busy in the shop as I try wrap up two guitars, and had another one come back in briefly for a repair that turned into an upgrade.

The Clydesdale offset build has properly entered its assembly phase. This week I did the cavity EMF insulation, made a prototype pick guard, and finally fitted the neck to the body,.

All the guitars I build have their internal cavities lined with copper tape to provide protection from radio noise for the pickups and wiring. The end result looks stunning, but is never actually seen by the user, as it’s all hidden away inside - a hidden inner beauty. This stage took slightly longer than I expected due to running out of tape halfway through, but thanks to Amazon Prime was completed a day later.


Cavities lined, I then did a quick mockup of the pick guard in cardboard to ensure all the sizing was correct for both placement on the body and that the holes for the pickups are correct.


As I mentioned last week, I’ve not had much luck in the past making custom pick guards on the CNC router, but for this guitar what I have in mind really needs to be done that way, so before I made the one for The Clydesdale I wanted to try cutting another one as an experiment first for a different technique, which I’ll explain below.

Finally, I also mounted the neck to the body this week, taking what has been two parts for so long and making them into one. To simplify the CNC process at the start of a build I don’t cut the neck on both sides using the CNC router, just from the front, which means I have to make the holes for the neck screws by hand. It’s not a major imposition (people have to do this all the time when they buy a new neck for their guitar), but given I do use an initial CNC roughing pass this sees a bit silly, and something I’ll rectify in future builds to save time. Anyway, it took me an hour or so to make sure these were lined up perfectly and straight (at this point the fretboard is rounded remember, so ensuring everything lines up isn’t just a matter of whacking the neck on the drill press and going for it). 


This picture makes me very happy though. The end starts to be in sight!

I mentioned last week that I had finally set up the bridge for the Prototype Offset guitar such that it would stay in tune when you used the vibrato unit. This made me happy, but I decided I wasn’t that pleased with the pick guard I’d made for it previously. Earlier in the year my attempt to made a pick guard on the CNC router had gone a bit wrong: the PVC material they’re typically made from did not sit flat (but to being quite flexible) and thus my attempt to make countersunk holes for the mount screws was a disaster. As an alternative at the time I just made one out of a duo-tone acrylic I had using the laser cutter, but now I looked at the finished guitar it just looked not up to the standard of everything else. 


Thus, I decided now was time to try again, applying all the learning I’ve had on the CNC router since. The problem I faced is that if you make an angled cut on the CNC router, if the material isn’t perfectly flat then you’ll get different amounts of material removed depending on the height of the top of the material from the bed, so I need everything to be as flat as humanly possible before I start cutting.

The first thing I did before doing anything with the pick guard was to level the sacrificial bed on the CNC Router itself to make that perfectly flat again. It needed doing anyway (if you recall I last did it back in late July), and this was a motivator for me to finally get around to it. This time was much quicker, having upgraded the CNC router recently to take 1/2” shank bits we now have a 2” diameter facing off tool, rather than the 1” one I used before.

Once the bed was levelled I took my PVC material and attempted to make sure it was perfectly flat on the bed. Just screwing it in at the corners isn’t enough for a material this flexible, so I took two approaches to this. Firstly, I covered the underside of the PVC with double sided sticky tape, and stuck it to the bed, which gave me overall confidence that things would be flat. However, I didn’t quite trust the double sided tape to hold up to the CNC router’s tug alone, and so the second part of my strategy was I drilled the holes for the screws that would hold the pick guard on the guitar before I did any other cuts, and then I used those to screw the material to the CNC router bed just as I would on the guitar! These screws follow the chamfered edge and gave me as good a guarantee as I can get that this bit of the material will remain flat, and the screws themselves are small enough they’ll be clear of the cutters by a safe margin. As an added bonus, this also meant I didn’t need to use tabs when cutting the material out either, meaning no manual sanding tabs off after cutting.

Plan down, I assembled the bits I needed for the job, can went to work: a 2mm bit for the mount holes for the guard itself, 3.5mm drill bit for the pickup mounts, a 2.5 mm end mill for the contours, and a 30˚ V-cutter to chamfer the edges.


Here you can see the results of the first pass, the material is now properly anchored down by the same screws I’ll use to attach the guard to the guitar.


And the end result was this:


Note the bits inside the pickup holes haven’t come loose despite not being held with tabs: it turned out the double sided tape was quite effective, and the free bits did stay in place. I then had to take the guard off the bed, remove the protective film from the top which was now frayed at the edges, clear up the edges a little and peal off the tape (and wash off the remaining residue. Then it was a little work to swap the old pick guard for the new one, and the guitar now looks so much better!


I’m really pleased with how this has turned out, and also with the fact I’ve got custom pick guards licked after several failed attempts. This coming week I’ll attempt to repeat this success for The Clydesdale.

This guitar here though is now just waiting final setup before I list it for sale. If you fancy a hand built offset guitar, fitted with top quality components (House of Tone pickups, a Mastery Bridge, fitted hard case, etc.) , this get in touch. It’ll be going up on Reverb shortly I expect, priced at £1500 plus postage. 

The very first guitar I built for myself still sees regular usage (indeed last week went to Rotterdam with me), and I was sad to note that the volume pot was starting to fail this week: mostly it’d be fine, but occasionally you’d whack the volume to max and it’d cut out until you rolled it back off and up again. Thankfully it’s a simple enough fix to replace the pot, and was finally the excuse I needed to give this guitar a coil split option on its humbucker.


As I’ve probably mentioned before, a humbucker pickup (the silver rectangle just under the neck in the guitar above) is actually made from two pickup coils, rather than just one as you might expect (and is common for a lot of pickups). The idea is that the two pickup coils are wound in opposite directions, so that radio interference picked up on one is picked up as a negative on the other and when you combine the signal from the two it cancels out (and thus it eliminates the “hum” you get on single coil pickups, thus the name hum-bucker).

In addition to the noise being cancelled out, because you have two coils working together you get a stronger output from a humbucker compared to a regular pickup, which is good if you want heavy rock tones, but there are times when you’d like something less juicy from your guitar. One way to achieve that is to put in a coil split, where you electrically cut out one of the coils in a humbucker and just use it as a single pickup. You risk getting the hum back then (you can see why I do the copper lining above now), but you’ll get a second tone out of your one pickup; and the more tones you can get out of a single instrument the more versatile it is, which in my book is a good thing.

To achieve a coil split you generally need your humbucker pickup to have an extra pair of wires coming out of it already to support this, and the Gibson 490R pickup in my blue tele did indeed have wiring to support this. Thus, all I needed to do was add a switch to let me toggle that extra coil on or off. What’s this got to do with the volume control you ask? Well, the conventional way to add this switch is to have a volume knob that will pull out to switch the second coil out, and push in to return it to its rightful place - so now that I needed to replace the volume control it was time to make this change! I already had the right push/pull switching pot in stock, so all I had to do was remove the old one and install the new one.


You can see here the previous wiring, and the new pot beside it with the extra solder lugs on the bottom of it to let me do the switching. My soldering must be getting better, as doing the swap was a lot quicker than I remember it was putting this guitar together back in mid 2016 when I just got started. I’m really pleased with the results: the new tone on this guitar is still clearly related to the humbucker sound, but if I want to play something a bit more nuanced then I can, but I can still run it as originally intended when I want to crank up the noise. A simple mod that gives you a bunch of flexibility. 

If you’re in the Cambridge area and would like a mod like this for your guitar, do get in touch!

On the topic of circuits, it’s frustrated me that I don’t have a proper way to record existing circuits I’ve made for documentation purposes (like those in the guitar or in the amp I made) or a way to record new circuits that I come up with, so I finally sat down for a little bit and started trying to learn Autodesk Eagle. It’s early days, but I’m starting to get my head around it, despite it’s quite archaic interface. I used to work for Intel a long time ago, and back then hardware design tools always seemed to lag substantially behind other domains in terms of usability: it seems nothing much has changed in the decade and a bit that has passed since then.

Finally a little word of caution if, like me, you find yourself taking your Apple laptop into the workshop. I noticed that my laptop wasn’t charging properly at times, and that the magsafe power adapter wasn’t sitting in it’s socket quite right. I finally had a look at the socket, and it turns out it had picked up a bunch of iron shavings that had stuck to the magsafe port!


Getting them out of their wasn’t easy, as it turns out the magnet in the magsafe port is really quite strong, but once I fished them out the laptop worked a lot better :)

A week in Rotterdam

It was a very short week in the shop this week, as I was mostly in Rotterdam on semi-vacation, but I had two workshop-days in which to get things done, so I tried to make the most of them.

First up, I wrapped up the oiling of The Clydesdale offset. Last week I lamented about how the oiling of the fretboard had gone a bit Pete Tong, enough that I had to just sand it back and start again. Thankfully this time the oiling went better, though only because I kept a watchful eye on it. It was indeed the case that having pedantically wiped off any excess after doing a coat of oil on the fretboard, if I came back ten minutes later I found some oil had seeped out of the fret edge and the deeper grain lines in the wood, and I had to do another round of mopping up excess. After this second pass it seemed okay thankfully.


This let me get both the body and the neck finished before I went away, of this will have been curing in the mean time, so now I can go back and start the assembly this week. Before leaving for Rotterdam I ordered all the remaining parts for the guitar, so they should all have arrived by the start of the week. The only fabrication I need to do now is the cutting to the scratch plate, which I’ll touch on later in these notes.

Given The Clydesdale was at a slow stage where I can’t do much to it, I finally turned my attention back to The Prototype Offset. This guitar was “finished” a little while ago, but I had problems setting up the floating vibrato on it, I just couldn’t get it to stay in tune, and as such I felt I couldn’t ship it. I really should have dealt with this a while back, but with The Clydesdale being behind I wanted to focus on that first, and am only now getting around to it. There are unfortunately only so many hours in the day, and I can’t devote them all to guitars as I’d like.


I sat down for an afternoon with the guitar, armed with some spare strings, some spare bridge springs (some tighter and some loser), and some vaseline to let me make sure everything was moving properly, and I tried to work out what was what. The first thing to eliminate was whether there was an internal interference that was causing the bridge to get stuck or struggle. The problem with the dynamic vibrato design found on the Fender Mustang (similarly the one on the Jaguar and Jazzmasters) is that you can’t see what’s going on inside once mounted (c.f, the unit on a Strat where you could see what’s going on easily). To get around this I constructed a bridge mount made in MDF that was the same thickness as the cavity I’d made in this guitar, which let me check if there was any contact as I operated the vibrato.


This told me quickly that this wasn’t the case: despite being close to the sides, they don’t touch at all, and nothing protruded out the bottom either, which was a relief as it meant I didn’t have to take a router to the guitar to remove some more wood.

But then the question is what was I doing wrong? The short version of a long story, which I will blog as a separate tutorial, is that two things needed correcting. Firstly, I’d not realised that for this vibrato design to work correctly I needed the posts that sit between the strings and the springs to be perpendicular to the body surface when everything is at rest and in tune. Once I realised this, it was much easier to see the impact of tweaks to things like the height of the string mount point on these posts and similarly for the springs, as you could use the angle of the posts as an indicator as how well or badly it was set up.

The all highlighted the second issue, which was that the springs I had were too tight for the strings I was using (11s), and when in tune the bridge was far from it’s optimal position. Thus I swapped the springs for a lighter pair, and switched the strings back to a slightly lighter gauge, and now the guitar is much happier. It’s still not perfect, but definitely stays closer to in-tune after you use the vibrato. Here’s a little vid of me playing the guitar (admittedly not with vibrato, as I was just having fun here and I don’t really know any songs that use the vibrato).

It’s been quite an educational experience working through this, and I’m amazed that floating bridges are as popular as they are given how balanced they need to be. Expect a proper write-up with pictures in the near future.

The one thing I did realise coming back to this guitar after a couple of months away is that I’m not happy with the scratch plate, which was laser cut from acrylic due to the failed attempt at making it in PVC (as documented here). I’ve decided to have another go at making PVC scratch plates on the CNC router, both for this and for The Clydesdale. My plan to overcome the wrapping that PVC naturally does is to use double sided sticky tape (and lots of it) to tape the PVC flat (in addition to screwing it down to stop side motion when being cut). Hopefully this will produce a better result than last time.

After getting more quotes for the metal part for the pedalboard I pulled the trigger on getting a prototype commissioned. In the end I had two quotes at roughly the same price, but I went with the supplier who was more local and would do powder coating as an option, but with a higher lead time. I think right now I just want to optimise for somewhere I can easily go talk to the people if things don’t work out as I hope: this being my first attempt to get metal fabricated I’m sure something will go wrong as I’ve not designed it right or communicated it correctly. 

Screenshot 2018-12-09 at 18.56.13.png

I suspect I’ll get the first prototype back in January, which is quite exciting.

For most of this week my other half was attending a conference in Rotterdam, and having never been to that part of the world I tagged along. Given I’d be at a loose end for much of the time, and the weather forecast was poor, I took advantage of the fact we were travelling on the Eurostar rather than flying and took my guitar with me, so I could keep up my regular practice routine in the AirBNB each day.


I was using Garageband on my Mac via an iRig adapter to let me get sound without annoying the neighbours, and it was fun to try and actually record some songs along to a virtual drummer - mostly a reminder that I don’t do that often and as such my timing needs a lot of work :) But that’s inspired me to do more of that, and ideally do so with my mic’d up amp and pedal board.

A week in the shop

At the end of last week I’d finished levelling and polishing the frets on The Clydesdale offset, leaving me to “just” etch the headstock and then oil the neck.

For the headstock rather than just have the normal Electric Flapjack logo, this guitar is also having its name etched their. When the customer ordered this guitar I’d told him I’d be happy to laser etch something custom on the headstock (or body), and when he decided to name the guitar “The Clydesdale” it seemed a natural addition to the headstock (the name stems from the customer’s hometown of Glasgow being on the River Clyde, and the Clydesdale is a type of horse, fitting for a guitar derived from a Fender Mustang).

I sent the customer some mock ups of how it might look with various combinations, and he came back with a rough idea of what he wanted along these lines:

clydesdale headstock.png

Nothing too complicated, but it had the one downside that it doesn’t leave much room for a string tree. But rather than change the customer’s contribution, I simply made a tweak to move the Electric Flapjack logo around to match the name text which frees up some space around the nut end of the headstock for a string tree. Having done that I then sent a rendering of what that might look like back to the customer for approval: once the etching is done that’s it, so I just wanted to be sure he was happy with how it would all look.


He was indeed happy with this, so then I went and did the scary bit and set about etching the design into the headstock. I also had the neck from the recovery offset that needed a logo adding to it’s headstock, so I decided to do the same angled Electric Flapjack logo on that one, which gave me an excuse to double check alignment before doing it on the customer’s neck.

As ever, each bit of wood is different, and the strength of the laser cutters over time will vary, when I cut the neck I made sure to save the offcuts for this moment. Using these offcuts lets me calibrate the design and check how it’ll look before committing (measure, measure, cut!). One thing I realised on my test cuts was that the logo didn’t quite look right when rendered at an angle relative to the laser bed, but looked nicer when the laser cutter’s natural motion aligned with the text. Normally I’d just lay the neck so it matched the direction of the cutter, but given the results seemed slightly less good when text is drawn at an angle, I simply rotated everything around and put the neck on the bed at an angle instead.


Once I was happy with both these things, there was a nervous few minutes, and voila, one nicely etched headstock.


The etching of the headstock marked the end of physical production for The Clydesdale (modulo drill holes for later assembly). The body had already had its coats of crimson stain the week before, so it was starting to look like a guitar ready for finishing.


As I discovered when making the prototype offset, which also had a wenge neck, wenge works best when finished more like maple (oiled for coating) compared to rosewood (oiled for absorption): if you apply a fingerboard oil as you would with rosewood you lose all the two tone colour of the wenge.

The process for oiling is simple enough: apply a coat all over each part, then you go back after a few minutes and wipe off any excess that the wood hasn’t sucked in, leave it to dry a day, and then give it a quick rub down with a high grit sandpaper (I’m using 2000 grit for this) to get a nice smooth finish before applying again. 

The maple neck has blown me away now that oil has been applied. Before oiling it was nice, but a fairly flat looking bit of wood; after oiling the wood came to life with a lot of depth depending on how the light plays in it. I posted a video to instagram (which I can’t embed here, but you can go see it here), but for comparison compare the headstock here to the photo above.


It has so much more definition now, and the light reflection move beautifully as you move around the neck relative to the light source.

Whilst I was bowled over by the maple, once again the wenge provide to be my nemesis. The first couple of coats went on okay, but the third coat seemed to dry with blotches on it, despite my wiping away the excess. It did look like some had seeped out from under the frets, or it could have come out of the natural figuring in the wood. Whatever the cause, it wasn’t a finish I found acceptable. Thus, my Sunday was lost to sanding down the entire fretboard again (between the now installed frets) to get it back to an even state all over, followed by working my way back up through the sandpaper grits to get a smooth finish once more, re-polishing the frets as they’d been inevitably scuffed during the sanding process, and then finally another coat of oil applied to the fretboard. 


This time I was very fastidious about removing excess oil, and checking back after a short while to remove any that had seeped out from behind the frets (which it had done again, but this time I caught it). The first wenge fretboard wasn’t so much trouble, so I’m not sure why this one has played up so, but it is just adding to my general acceptance that wenge is a better wood than I am a woodworker, and for now I need to find alternatives for future builds.

I’m going to be taking a short trip to Rotterdam this week (my other half is at a conference there, and having never been before I’m going to tag along), so my aim is to get the oil finished before then so it can have a week to cure before I move on to assembly.

Last week I talked a bit about my pedalboard design, and how I was looking for quotes for fabrication of the metal parts. I’ve had one quote back, one request for more detailed drawings (which I generated and sent - I knew that O-Grade in Technical Drawing would come in handy one day…), and one not respond. The quote I had back was acceptable, but that place doesn’t do painting/powder coating, so I’m just trying to get one more quote from somewhere that does before I pull the trigger.

In the mean time I laser cut a plywood mock up of the top surface to check that my modelling of the pedals in CAD matched reality:


Turns out 3mm ply is quite flexible when you put pedals on it like this, so not an ideal test material, but it was enough to convince me that my tweaked design makes cable routing much easier than on my current board.

Final bit for the week was I learned how to turn wood on a lathe, which was interesting (and fun). Wood turning is much more free form than metal turning was (at least on the kit that Makespace has), being more hands on material, so it’s clearly a skill to be mastered. But if anyone needs plectrum bowls, let me know: I’m sure that’ll be the hot product for 2019 ;)

A week in the shop

I was under the weather for quite a bit of last week, so not a huge amount of progress to report on the guitar build front. The Clydesdale offset continues to move forward slowly: I’ve now finished the fretwork, and the next step is laser etching the logo oiling it. Fretwork is definitely something I need to find a way to improve on: although I felt the early stages went well, getting the frets buffed out was quite a lot slower than I expected. But it is there now:


For all the effort and my desire to avoid wenge for a few builds, it is looking lovely. Headstock etching, final sanding, and then oiling are the next stages.

After upgrading the CNC router last week I made a quick collet holder, as the old one no longer worked due to the new collet size. It was quite a quick job in Fusion 360 to design: it actually is pretty close to what I teach when I do my introduction to Fusion training course. Not long after I had the parts cut out on the CNC router and glued up, and the following day it was up and in use:


It’s little jobs like this that I really love about having all the tools in Makespace: you have an idea of something that’ll make your life easier, and a day later there it is in use. There’s a lot of focus in hack spaces on the esoteric things you can make, but I find this every day problem solving much more interesting.

On a related topic of making things that scratch a particular itch, I’ve been looking a little at sheet metal fabrication this week. With my new pedal setup I realised that although I have quite a nice pedal board, the way it does cable routing wasn’t quite right given some of the bigger pedals are larger than what the board was designed for, and perhaps I could design one that was better suited for my needs. The pedal board I have is made from a bit of folded metal with some nice wooden end-caps: the wooden end-caps I can easily make, but sheet metal is not something I’ve played with much before, and I felt it’s been a while since I pushed myself design ways, so I thought I’d give it a go.

Fusion 360 has a mode for sheet metal folding, which I had played with before a year ago when I tried modelling the pre-built chassis I bought for the amp I made. This time though I wanted to design something that didn’t quest exist yet, which was more fun, and I need to actually get it made rather than just worry about something already made, which is a different kettle of fish, as I have to think about construction practicality as well.

My initial mockup of the metal part for my new pedal board looks a little like this:

Screenshot 2018-11-26 at 22.50.56.png

The main change between the pedalboard I have and the design I have here is just a tweak to how the cable slots are cut. I actually modelled it with rough replicas of my particular set of pedals in Fusion 360, so I know that they’ll fit perfectly. Beyond that it’s mostly what you’d expect to see in terms of a raised sloped surface with space underneath for cables and a power brick. In the back I found a socket and switch unit that’ll take a standard kettle lead which I can then forward to the Walrus Audio power supply I have that’ll sit under the board, making it easy to plug and unplug and turn on and off.

When folded out the sheet metal looks like this, which is what I’ll have to get someone to cut out and fold for me:

Screenshot 2018-11-26 at 22.52.06.png

It’s really quite nice how Fusion lets you design in this way, making quite a complex task relatively simple for someone inexperienced in metalwork like myself. It has a stock set of bend radiuses etc for different metals, but it’ll be interesting to see how different fabricators will deal with my design.

I’ve sent it off to three different manufacturers for quotes, noting in the request what I thought might be problem areas with the design for fabrication just to enable them to complain if I’ve done something silly sooner rather than after I’ve spent a lot of money on getting it made. It’ll be interesting to see both how much it’ll cost to make and whether this easy to generate design actually cuts the mustard when given to a professional metal work person.

Music Memos

A friend recently pointed me to an iOS app that Apple make but don’t seem to sing about much, but has really changed the way I practice guitar, and seems to be a great potential song writing tool too: Music Memos.


The app starts from a simple premise: it lets you record little bits of music as you play, just as you would with a voice memo app. That in itself is useful enough when practicing - it’s always good to listen back to yourself and hear what you sounded like not what you think you sounded like - but once you have recorded your little clip, it does all kinds of useful other bits. It’ll do an automatic transcription of what you played, putting it on a bar based view with the chords for each bit annotated over the track

You can also add a backing band to your recording, adding various drummer styles and an automatic bassist (which I don’t think is even available in GarageBand, Apple’s normal go to software for amateur musicians). The range of drummers isn’t as comprehensive as say GarageBand’s, but it’s enough to let you get an idea fleshed out quickly without making the UI overly complicated. It also will sync my recordings via iCloud so I can listen to them later on my iPad, or open them up in GarageBand directly if I then want to take the moment of inspiration further, or I can post them to SoundCloud etc. to share them with people.

But for all its features, one of the best things is that for the most part the UI when you’re playing is uncluttered and simple letting you concentrate on playing guitar, not the app.


The main UI just has one button to let you start and stop recording, and there’s even an automatic mode that I assume will track when you start and stop, but I’ve not yet played with that (when practicing I don’t need it to automatically record my scales :). At a lot of guitar apps I’ve tried have very cumbersome user interfaces and when I’m playing guitar I don’t want to be thinking about the app, it’s enough for my brain to cope with trying to play new things my guitar teacher has set me.

The number of times I practice and think “oh, I wish I ‘d recorded that” is reasonably high, so having an app I can just have to one side and turn on quickly to capture something fills a niche nicely. I’ve also taken to using it at my lessons: I have a terrible memory for rhythm patterns and timing, and so I’ll get a new riff or sequence down whilst at my lesson, but the next day looking at the sheet music have no idea how it was meant to flow. With Music Memos I now just record that riff at my lesson before we move on, so for the rest of the week I have that reference to hand.

If you’re a guitarist with an iPhone I really recommend you give it go, and a big thanks to Jason for pointing me to it!