A brief interlude in the shop

I managed a couple of days in the workshop this last week, as pressure on contracting work let up briefly, however having just limited time in the workshop is highlighting that just getting a day in the workshop doesn’t mean I get a day to work the workshop, and I need to better manage my time whilst there. I do most of my guitar building in Makespace, which is a community workshop, and in general that community aspect is a wonderful thing. But being in that community means I have certain responsibilities and commitments to help the workshop run smoothly, and whilst none of these are particularly onerous, they do break up the day in a way that makes it hard to do long jobs if I’m not careful. For example, this Friday I went in and I had one client meeting to squeeze in for half an hour in the morning and a training session on workshop kit to run in the afternoon for a couple of hours, this fragmented the day enough that I didn’t get time to do the fretwork I’d intended to do that day, which means things spill into next week or consume my weekend.

In computing there’s a term called “context switch overhead”, which means that you can’t just swap from doing task A to task B without some level of wasted time as you pack away task A and get started for task B, which is just as important to humans as it is for computers. Even just someone coming and asking you a question when you’re concentrating can cause you to have to do two context switches and spoil your flow, which is why in the community workshop we have a orange lanyard system as a way that politely says “do not disturb”. Whilst I like to try and help people at Makespace as much as I can, I suspect at times I should learn to pick up that orange lanyard.

Most of my time in the workshop has been on the next for The Clydesdale, the first of my commissioned offsets I’m building. The last time I was in the workshop I managed to get the neck carved and sanded, but I wasn’t quite happy with the feel in my hand, so I took a second pass at it and now it feels just as I want. The birds eye maple is also coming up pretty nice: at times I think I prefer a simpler look to maple, but this one I’m really pleased with now.


With the neck carve done I did a final pass on the fretboard to sand it smooth and ensure all the slots were to the correct level ready for the frets to go on.


Fretting is appropriately named, given it is one of the more stressful parts of a build, at least for a new builder like me. This is my sixth fret job so far, and I’ve not yet found my routine. It’s in part here where doing guitars at low volumes is not good, as by the time the next one comes around you have to relearn all that you did last time. I actually did my 5th fret job a couple of months ago on the neck for the recovery offset, but that feels like an age ago. Perhaps I should just make a batch of necks at some point in the near future just to do it repeatedly and build up some sort of muscle memory for the process.

Frets in a guitar, much to most people’s surprise comes as a length of wire, which you cut into short strips. The cross section of the wire has a sort of mushroom shape: there’s the bit you normally see on top of the fretboard, called the crown, which has a half circle shape, and then a bit sticking down, called the tang, which will go into the slot on the fretboard. The tang has some teeth on it to help it grip in the slot. Here’s a random picture of some fretwire from the people I buy mine from to help make it more clear:


Depending on the width of your slot vs the width of the tang you may or may not need to use glue to help the fret stay home. My aim is to avoid using glue if I can, as using glue will make life harder for the person who in future years needs to replace the frets as they wear down. But it’s not a hard rule: when working with wood you can’t be sure of how it’s going to behave and as such hard rules need to be more flexible, so if necessary I will use a little glue to make sure the frets stay home, but only if required.

Before you cut the fret wire into strips, you want to put a slight over-radius on it. Your fretboard itself isn’t flat, the ones I make have a 9.5” radius on them for instance, and to make it easier for the frets to be hammered in without jumping out you want to give your fret wire a radius that is slightly tighter than the board. If you have it with a bigger radius then as you hit it in at one end then the other end will see-saw out and it’ll be much harder to get home. If you over radius it slightly you can get the two sides home before securing the middle of the fret, which makes things easier (note easier, not easy :). To put the radius onto the fret wire you need one of these fret bending tools:


The main thing that I struggled with this time was I over bent the fret wire when I did the previous fret job. All my fretboards at the moment are the same 9.5” radius, so to save time and storage space, I did the radiusing of the fret wire in a single go. I didn’t struggle with that fret job much, but there the fretboard was made of maple rather than wenge, and maple is a much easier wood to work with. I’m really growing to dislike wenge as a wood to work with, due to the fact thet the density of the two colours in it are very different, so you have some very hard bits of wood and very soft bits of wood mixed in. When cutting the fret slots this can cause your saw to jump out the slot if you’re not careful (which risks scratching the fretboard surface), and here when putting the frets in you have bits which accept the fret readily and bits that don’t want to take it at all.

As I mentioned, when I did the maple fretboard last time, the fact I’d made the fret over-radiused by quite a bit didn’t matter too much, as the wood made the fret insertion relatively easy anyway. But here where the darker bits of wenge are much less forgiving it was taking much more work to get the frets home anyway, and the fact I put in more curve to defeat than I needed just compounded the issue. In the end I just threw out the already radiuses fretwire and got a new batch (which luckily I had to hand) and made it only just more than the fretboard radius.


Here you can see that it’s just in at both sides and only just out of the slot in the middle. Before I had a millimetre or so of daylight there. 

Once I’d made the switch, things progressed more smoothly, though I’d not say easy, but certainly easier, and after about three hours I had all the frets home:


One other tip I’d pass on here - this time I trimmed the edges as I went along rather than doing them all at the end, and that is worth doing. If you do them all at the end it’s hard to get your cutters in on the more closely spaced frets, and it also makes it easier to do the final tapping home of the frets on teh edge without risk of hitting the extruded bit and bending it.

I still unfortunately haven’t finished here. One of the frets in the middle where the wood is softer on the left side isn’t home right - down to the fact it was one of the overly radiused set and the soft wood isn’t gripping it, so I’m going to remove that fret and put in a new one, and potentially add glue to secure it properly.

Like I say, you can see why it’s called fret work.

The other thing I did was prep the recovery offset’s body for the first coat of grain filler, sanding it up to 400 grit all over. I ran out of time to actually apply any filler last week, but hopefully I’ll get time in the near future to do that. I should actually start doing the stain on The Clydesdale too, which is going to be a dark deep red similar to The Red Rocker tele I made last year.

A week out the shop

As has been the theme of the last few weeks, I’ve been flat out with contracting work again, and failed to get into the shop at all during the week, and at the weekend I went to the Mozilla Festival, a sort of technology festival ran by the Mozilla Foundation (the people who make the Firefox web browser). I imagine it’ll be another busy week coming, and then I hope to get back into the shop again.

The one thing I did manage to make was a small set of clips to try keep the microphone I bought for my amp straight. I have a low profile Sennheiser e609 microphone, and like a lot of people I simply drape it over my amp, looping the cable around the amp’s handle to keep it from slipping. This works fine in theory, but in practice you’ll find your mic cable has a natural twist in it, and getting the microphone to sit perfectly parallel to the amp face is a very frustrating experience. To try and help keep things aligned I made a simple pair of clips that hold onto the cable and provide a flat surface to rest on the top of the amp, helping me align it better.

Screenshot 2018-10-30 at 15.20.08.png

I made a front and a rear clip with slightly different designs: the rear one has an extra section to hook on the rear lip of the amp to stop the weight of the microphone dragging the cable forward, and the front one just has the plate to stop the cable twisting. It took me a couple of goes to get the sizing right, but here’s the clips in situ:


And as you can see, it’s kinda working. They’re not a great solution - you can see teh front clip is at an angle due to the way the cable bends over the front lip, and I’ve not got the balance quite right between the tightness of the clip on the cable and the ability to squeeze the cable into the clip. But it’s good enough for now, and better than the previous solution of just wrapping it around the handle and hoping it’ll sit in the right place.

To me the best thing was that it took me 30 minutes to go from idea to having a parameterised design in CAD that I can readily tweak to having the first printed prototype: all this practice is paying off.

The only other thing I’ve done guitar wise is order some grain filler for the Recovery Offset’s body. Whilst normally I like to stain my guitars, the recovery offset is made from a body and a neck that the CNC router tried to ruin, and thus they have obvious patches on them. A friend has a spray booth setup for painting, so I’m going to use this as an excuse to learn how to paint a guitar properly. To ensure a proper finish though I’ll need to apply grain filler to the ash body (if I want a flat finish, which I do), and thus I’ve got some of that on order, which hopefully will turn up before the end of the week.

A week (mostly not) in the shop

My run of having to do contract work that keeps me out the shop for the majority of the week continues, as I happen to have a couple of clients both with near term deadlines, but I managed to safeguard a day and a half this week to be in the shop and move things forward on The Clydesdale (the commissioned offset). To start with I finished getting the fretboard just right. and sanding smooth the transition between the headstock and the fretboard. Wenge is a particularly tough wood to work with I find, so getting the finish right on the radius and removing all the machine tool marks from first facing it off and then adding the rough radius took quite a bit of elbow grease, but the results are worth it:


The frontside done, then it was onto the rear of the neck to carve the bulk of the neck, which I do by hand using a Shinto hand rasp. I don’t just go at it blind, I set up a bunch of pencil guidelines to help me get the initial facets from which I then just freehand, taking frequent pauses to check how it feels when you make various chord shapes and using a straight edge to check the taper is right.


This is the firs time I’ve made a neck with birds eye maple, and the back has come up wonderfully textured as you can see.


The neck is now roughly sanded smooth and I’m happy with the feel, so the next step is to get the frets in.

My new audio recording gear arrived this week. I’ve been shooting little demo videos with my iPhone, but the audio quality isn’t that great through its microphone (the microphone is fine, just it’s not pointed at the amp where the sound is coming from :). Whilst in general I’m quite content writing this blog over making videos, there are times with musical instruments when a short video with sound will capture something much more succinctly than I could do so in prose, so I’ve been trying to solve the audio quality issue. My desired solution was to mic up my amp properly, and get a thing that I could plug straight into my phone so I could record things straight to instagram or youtube without any additional steps: to this end I ordered a Sennheiser e609 microphone and a Zoom U-22 portable audio interface.


The Sennheiser microphone appealed to me because of its low physical profile and it has a good reputation (along with the higher end version the e906). Where my amp is set up I don’t have the free space required for a regularly shaped microphone on a mic-stand, so the arrangement with the e609 draped over the amp as in the picture above is perfect, as I can leave it in place all the time without having to worry about getting it all setup when inspiration to record something strikes. The U22 is a simple microphone pre-amp that’ll take in a standard microphone XLR connection and output it to USB audio, and is supported to work with iOS devices through Apple’s USB/Lighting adapter (not all USB audio devices will work this way, particularly if they need power over USB). Thus I just need to plug the U-22 into my iPhone before firing up the camera app and everything Just Works™. 

Here’s a little video I recorded this week to test it all works:

For the most part it did just work, but there is one small snag in that it seems if you give iOS mono input it treats this as stereo audio with just the left channel when it’s recording video. I did some hunting around for solutions to this, and it seems a general problem with some of Apple’s video tools when presented with mono input. Unfortunately I’ve not yet found a neat way to solve this. For the above video I had to import the final movie into the old Quicktime Pro 7 tool, then export the audio as a mono track and import that back into iMovie where it did the correct thing. Whilst this works, it’s no good for quick off the cuff record and post to web scenarios I hoped for. If you have a better solution please let me know, or I’ll post one here when I find it.

I’m also interested to note the Zoom U-22 claims to take guitar direct in, which means in theory I could use it as a replacement for my IK Multimedia iRig 2 adapter that I cart around with me when I know I won’t have an amp to practice with. I’ll report back on how that goes when I get time to try it.

As mentioned a while back, I have this finished offset guitar where I’m not yet happy with the dynamic vibrato bridge setup, and I’ve not had time to investigate it. The first step to understanding why the dynamic vibrato isn’t happy is understanding the mechanics of how it works at all, and to that end I wanted to build an experimental jig where I can attach the bridge unit but see what’s going on inside. To this end, I knocked up this jig whilst feeling a bit too tired to work on fine detail work like neck carving:


The bridge cavities you can see in the centre go all the way through to the back and are the same depth as the cavities on the actual guitar, just open on the back. I’ll attach some clear perspex to this and then I’ll be able to see how the bridge works, whilst at the same time checking it’s not rubbing on the sides or the bottom. I also have a bunch of springs of different tension I can try, and this jig lets me change those and see how good they are without risk of scratching the actual guitar. Some fun experiments ahead, which I’ll try to record, as there’s very little info out there that I can find on setting up this type of vibrato system. I did chat to an owner of some vintage mustangs this week and he says both his stay in tune fine, so I know it’s an achievable aim.

Finally, a bit of dull but important admin: I’ve started implementing better backups for my laptop. Whilst I do back up my Mac laptop using Apple’s Time Capsule device, that device is starting to creak at the seams a bit, and at the same time backups are becoming all the more vital for me. When I wrote software I’d push everything to a cloud server designed to store software code, but for a lot of the guitar work I do there is no nice easy service, so more of value resides on my local disk. Also, doing more rich media content like videos means I need more reliable storage than I can fit in my laptop too.

The solution to this has been I’ve got a Synology NAS storage system, to which I’m going to be both backing up my laptop, and using it as a holding place for large video and audio files that I generate. The NAS has three drives in it, one of which is just used to double check on the other two (it’s a RAID array if you’re technically minded), so not only does it mean I have things off my laptop which can get lost/destoryed etc., but the NAS itself is tolerant of a disk failing too. 

Anyway, this is mostly tedious admin, but let this be a reminder if you don’t have backups for your work then you should do. I’ve had several hard disks die on me over the years, and backups are vital to getting back to business quickly when such things happen.

Rediscovering that first guitar

Whilst I’ve no intention to switching to vlogging on a regular basis over blogging here, working with musical instruments means there are times when words aren’t enough: I want to let you hear things not just read about them. To help with that I just recently got a proper amp mic to help me capture better the sounds I want to share at times, and this video is a quick first test of that new kit. I got into guitar building because the guitar I’ve owned since university wasn’t making me happy, but recently I found a niche where it does a particular sound better than anything else I have, which means I’ve got back into playing it, and I wanted to share that.

Between dropping the guitar to Eb and pairing it with a much better amp than it’s ever been put through before, the guitar is now getting regular play for the first time since I started building my own guitars. It’s not good for everything I want to play, but it’s nice to have a spot in the range of things I play where it gets some use rather than being a bulky memento to my university days.

A week barely in the shop

This last week I got to spend very little time in the workshop, which was frustrating, but I’ve had one contracting job enter a crunch phase and another just kick off, so last week was hectic, and the weekend was spent visiting family. I hope this coming week to carve out at least one weekday and some of the weekend to be in the workshop. The two offsets are nearly over the hill in terms of the more risky bits of assembly.

I mentioned the recently that the client who commissioned the first offset, now called The Clydesdale, wanted to do a custom headstock design, and he’s now provided me with the artwork for that. I’ll use this to laser etch the name alongside the regular Electric Flapjack logo; it’s nice to be able to offer personal touches like this to the guitars.

The one bit of time I did get to do in the workshop was helping out Jamie from Swannell Guitars, an amazing acoustic guitar builder also based in Cambridge.

Jamie needed some holes made in a bit of wood: sounds easy enough, but he needed 649 holes, each about half a millimetre in diameter, all into a piece of wood about 2mm thick. Whilst in theory he could do this by hand, it’d be very tedious very quickly, and that’s normally (at least for me) when mistakes happen, so this seems a good excuse to break out one of the more automated bits of workshop equipment that are found in the Makespace workshop.

My first instinct was to use the laser cutter for this, as whilst I do have some very small end mill bits I can use on the router (0.6mm diameter), I was worried about such small bits breaking as you cut, and that’s never a worry with the laser, and at 2mm thick the laser cutter will have no issue getting all the way through. I did a test run on paper with all 649 dots first to convince myself that the 0.5mm pitch wouldn’t be an issue on the holes, and then tried a smaller sample on some scape wood.

It cut the holes fine, but the results were not satisfactory: the main problem we hit was the laser doesn’t cut a perfectly straight hole, but more of a v-shape, with more of the upper layer of wood being burned away than towards the bottom of the hole. These holes would be used to hold a wire in place, and so the fact that they were quite tapered let the wire wobble around too much. Instead we turned to the CNC router and those tiny end mill bits I have:


The bits have a diameter of 0.6mm, so just about the right size for what we need, but we didn’t want it much wider. The first problem was just making a 0.6mm hole with a 0.6mm bit like this isn’t going to work: router bits are generally as a rule side cutting tools, not end cutting (as with drill bits), and in my experience if you get a CNC router to drill it moves the head down quite rapidly, which in our case would just break the bit anyway given how thin it is. To counter this I imported the design into Fusion 360 and told it instead to bore a 0.6mm hole and claimed the bit was 0.59mm (Fusion sensibly won’t bore a hole if the bit’s diameter is as large or larger than the hole’s diameter). I gave this a quick test run on some scrap, and much to my amazement the fragile looking bits did exactly what I asked without complaint.

We put in a sample bit of wire, and it held nicely, so in the end we cut all 649 holes using this technique. This did require us manually selecting all 649 holes in the CAM part of Fusion 360, which was quite tedious, but I couldn’t see a nicer way of achieving this quickly, but it got the job done. If anyone knows a better way of doing this then answers on a postcard to the usual address.

Having been on the receiving end of so much help and support from other luthiers and woodworkers, it was nice for once to be able to give back and help someone else out. Hopefully ta some point you’ll get to see what Jamie is working on: it’s going to be pretty special based on what I’ve seen!

Despite my non-guitar workload this last week I did find time to  continue to tweak the guitar designs for 2019 whilst away from he workshop. I’m living the rapid prototyping dream by updating my design in CAD, 3D printing the model at a quarter size to get a sense for how the lines and curves look in reality, then feeding that back into CAD when I have a spare 15 minutes, and kicking off another print again.


At the end I plan on making a poster of how the design evolved from all of these!

One thing that’s important to me is the ergonomics of the top, so I’ve been experimenting with different top carves on designs. I was not very happy with how the initial carve designs looked, when Jonathan Woolf (Makespace general awesome person) pointed out that I was using low res to save time, but for things where you have a non-flat surface like this you want to accept the slow print time and do it at fine resolution. On reflection, he’s clearly right: the top carve is around 5mm in depth of the guitar, and I’m printing them at 25% scale, so I get around 1mm of difference between the top and bottom of the carve and going from the “normal” fast print 0.15mm resolution to the extra fine 0.06mm resolution is the difference between 6 and 16 layers across the carve area, and you indeed get a sense of the subtle curves much more when you use the extra fine setting.

Still some more prints to do, but it’s been a very useful process, and I feel I’m slowly getting there with the design. Of course, I need to clear my backlog before I get to build any guitars based off this work, but it’s the find of thing I want to be ready to go once my backlog is cleared.

A week in the shop

This week has mostly just been a week of graft, getting the two necks moved along: the maple on maple one for the recovery offset build (or painters tape on maple as it has looked most of the week), and the wenge on birds-eye maple for the commissioned offset build. 


The recovery offset neck is slightly ahead of the commission offset neck, deliberately so given it’s been a few months since I did the fiddly bits of making a neck and I might as well get practice on the less pressured build first. On the recovery neck I just did the first pass of the fret work all the way through to, but not including, the polishing off the frets. They’re all level now, and the ends just need a little tweaking before I do the polishing, but overall this one is looking good.


I wrote a few weeks ago about how I wasn’t happy with the finish of using clay inlays, and so I decided at the last minute to just go for pearl effect inlay dots on the commission neck. However, I had to do a sort of hybrid of both, as I’d drilled the dot holes on the assumption I would be using clay inlays, which require more depth, so they were a mm or so deeper than I needed to glue in the pearl plastic inlays (3mm deep vs the 2mm deep inlays). So to get that mm back I used a little clay and made the worlds smallest coffee tamper using a spare inlay dot and a cocktail stick to let me press the clay down properly :)


I let the clay set for a day and then glued in the pearl inlay dots and they sat perfectly That done I’ve then been sanding the radius in and making sure the fret slots are kept clean. The wenge has needed much more sanding to get the machine marks out than the maple did: I think the more brittle darker bits of wenge took more marks than the maple did, so that’s a learning for next time I machine a fretboard blank. But now it’s radiused and ready for fretting in the next week.


The top of the hill is almost insight for these: after fretwork it’s finishes and electronics, none of which are quite so much stress and graft as fretwork.

t’s nice to find oneself in a little community, even if I’m a total amateur compared to the other two luthiers I know in Cambridge. My friend Matt has a spray booth now, and so I’m going to have a go at spraying the recovery offset’s body in the near future. Normally I’d stain and oil the body, but given the repair I had to make on the recovery body is visible on the surface, but totally smooth to touch, I figured this was a good time to try out painting. I have been asked in the past to do painted guitars and have said no, so it’s a useful learning step to enable me to say yes to that in the future.  I have some prep work to do, notably using grain filler which I’d not normally bother with, so that’s on my todo list for the next couple of weeks.

Jamie, who builds acoustics, got in touch asking if there was anything at Makespace where we might make some 0.5 mm holes, to help with a build he has on at the moment, and so I can do my bit to help for a change, and this coming week I’ve arrange with Jamie that we’ll see if we can use the laser cutters to solve his problem.

A week in the shop

This week my first goal was to glue up the fretboard for the commissioned offset to its respective neck, but I wanted to avoid the issue I’d had when I’d done the same bit on the recovery offset guitar that is also being worked on. If you look back at my notes from that week, I had the issue that although the fretboard seemed to dry clamp to the neck fine, after I added glue the join between the fretboard and the neck didn’t close as well as I’d expected, and I had to fill the resulting seam with glue so there wasn’t a gap along the join. This worked and doesn’t look too bad (and indeed, I think with a contrasting fretboard wouldn’t really be visible), but it was stressful to deal with, and I ended up with a lot of excess glue to get rid of which also was extra unnecessary work.

Having reflected on what I did that time versus what I’ve done in the past, I observed that in I’ve clamped an extra bit of wood on top of the fretboard when gluing it, to spread the pressure from the clamps more evenly over the fretboard. However, because with these latest fretboards I’ve radiused them before gluing, that wasn’t possible, as there is not flat surface to have the extra bit of wood sit on. So, I decided before gluing this latest fretboard onto the neck I’d make some radiused clamp blocks to solve this.

Screenshot 2018-10-01 at 11.33.53.png

To do this I simply modelled up a block with a concaved surface on once side to match the fretboard radius, and flat on the other side to camp onto, which I then machined out of some left over swamp ash I had from a previous guitar build.


Once cut out, I then laser etched a description of the radius on the back of them, as I suspect over time I’ll end up with a series of these for different fretboard radiuses, and I’m not that good at eyeballing subtle curve radiuses :)


Blocks made, I then put them to immediate use, gluing the wenge fretboard to the birds-eye maple neck for the commission offset. 


The blocks worked a treat: they spread the load from the clamp across the width of the fretboard nicely so that the edges were just as tightly clamped down as the centre. As a result I had a good tight join between the fretboard and the neck when I removed the clamps. Success!

If you’d like your own set of blocks like this, then I’ve added the Fusion 360 design to my open source page so you can use that if you have your own CNC router, or if you’d like me to make you a set for a given radius give me a shout: I’ll send you 10 blocks of the same radius for £25 plus postage.

Having glued the fretboard on, I’ve since tided the neck up with hand router using a following bit to trim the fretboard to be perfectly flush with the neck, and I’ve started cutting the fret slots. 


Whilst wenge is lovely once finished, cutting the slots is a lot more demanding than it is on something like rosewood or maple. The two coloured woods have quite different toughnesses, so you’ll find your saw blade will be stuck one moment and then running smoothly the next, and if you’re not careful with how much pressure you apply you can end up with the blade jumping out the slot and cutting wood where you’d rather it didn’t. Even though I give myself a head start by laser etching a guide slot on the surface of the wood, it’s still slow work and not something it’s worth trying to rush given the cost of mistakes!

I also made a start on the fretwork for the other offset I’m working on which is slightly ahead in the pipeline.


As I said I would last week, following on from attending the Festival of Maintenance, I created a page describing the setup and repair work on existing guitars I’m happy to take on and guide prices for that work. If you’re in Cambridge and would like me to give your guitar a bit of TLC then get in touch!

Looking to the future, I’ve also been doing a little more design work in spare moments on my own designs. Whilst it’s been fun building models based on existing guitars, it has always been the aim one day to have my own designs so the reflect both a mix of my own preferences and those of the people who commission them. To help me visualise those designs, I 3D printed some scaled down mockups to help me get a sense of how the designs are physically, rather than just on screen.


First up, it’s quite amazing how small a quarter scale guitar feels. I guess it’s 1/16th by volume, but they really do feel quite tiny. If there’s a market for guitars for dolls and action figures then I guess I’m ready to corner it :)

That aside, it really was useful having the scale model to get a better sense of how your design looks in reality. This even applies to other guitars that I’ve seen in screen or in pictures: I saw the St Vincent guitar at the weekend up close, and in person it looks very different from how it does in pictures, with the angles being more more striking that it appears on screen or page. The same was true of the curves on my initial prototype here, some of which worked better than I expected and some worse. But the nice thing about using a 3D printer to make the mockups rather than carving them in wood is that it was then but a few hours after I updated my design before I had the next prototype printed and in my hands for evaluation.

It helps that I have someone lined up for the first one of these (a baritone version no less) once I’ve cleared the current backlog with the offsets I’m building, and we’ve been exchanging ideas for the looks and features of that guitar to make it match their tastes, and it helps me evaluate the flexibility of having a guitar design that I can tweak to be flexible enough to meet different needs. 

At the end of the week it was nice to head over to the UK Guitar Show being held at London Olympia.

IMG_1712 2.jpeg

The show was quite a bit bigger than the other shows I’ve been to in the last year, with more mainstream brands being on display and less of the independent guitar builds and sellers. Whilst I do enjoy seeing the smaller builders (and there were some of those here), it meant overall I felt less bad about having a go on everything even though I didn’t intend to buy; at the smaller shows I’d feel bad about wasting someone’s time.

Some random highlights:

  • I got to hold a St Vincent guitar, though not play as the Music Man stand didn’t have an amp. It is even more lovely in the flesh than in pictures. The sad thing I noted is that there is no left hand version. I’m not a left handed player myself, but I know a few people who are, one of whom would have liked this, but alas that’s ruled out for them, which was sad.

  • I had a go on the Fender player series Strat and Tele on their stands. Nothing remarkable about these, other than to congratulate Fender for having a left handed guitar up front and centre on their stand ready for people to play.

  • I had a go of a Line 6 Variax guitar plugged into a Line 6 Helix effects board, and my mind was blown by the flexibility of this combo. The Variax works like a standard guitar, but you can switch it into a mode where it ignores the pickups and just runs a microphone under the bridge into a DSP unit that can then emulate all different types of guitars: strats, teles, Les Pauls, and even acoustics. It also lets you change tuning without adjusting the strings whilst in this mode. At the turn of a single control you can do from standard tuning to drop-D, to Eb, to DADGAD. Switching between tunings is always a pain, so that’s quite a game changer. When plugged into the Helix you can then adjust each tuning (e.g., drop the high E as well as the low E when in drop-D). It was hard to tell how good the sound was given the general cacophony of the show floor, but it was certainly a fun ten minutes I had messing around with this.

  • I tried the Vox Starsteam guitar, which like the Variax is meant to emulate a large range of different guitars, and has quite a nice unique design. It was a bit of a let down for two reasons: firstly, the stand had no batteries so all of the fancy features didn’t work, nicely highlighting the downside of fancy digital electronics in your guitar, and secondly whilst I love the looks of the Starstream, it isn’t that comfortable if you play it sat down, with the edge digging into you. It’s clearly designed to be played stood up, but there wasn’t a strap on the guitar. All in all, a bit of a let down, but still I find the general ideal inspirational.

  • I bumped into Jamie of Swannell Guitars, a Cambridge maker of amazing acoustic guitars. Jamie let me have a little jam on his first electric guitar, which was fun to play, and quite interesting as you could feel the acoustic heritage in the otherwise familiar shape of an electric guitar - I hope to catch up with him in the near future and find out more about his design.


A week in the shop

Brief notes this week, as I’m keen to press on with things!

To start the week I made the fretboard on the commission offset, just as I did two weeks previous on the recovery offset, the only difference being that this fretboard is made of wenge rather than maple. I did both the cutting out on the CNC router and the laser etching of the fret slot guides.


Thankfully for once I was just repeating what I did before, and it all went predictably well. I find one of the things I’ve yet to do is repeat any designs, so most things are always new which I think is part of why I’ve been so slow to make guitars this year. My original intent with the commissions I took in February was to just make some more t-style guitars, to leverage the fact I was building the same again, but in the end there’s been enough change I’m still slowed down by each piece being new to me. Which is why it’s nice I essentially have two guitars being built right now following the same path.

The inlays for the dark fretboards I like to make with clay. I get some air drying clay, set it into the fretboard holes, leave it to dry, and then sand it down. One drawback of this approach with something as textured as wenge is that the clay gets into the natural figuring in the wood. It does sand out eventually, but it’d be nice not to have to do that. So as an experiment I tried putting down masking take before putting the clay in:


This worked quite well, although you do need to be careful that there is a nice clean edge around the hole with no frayed edges on the masking tape, as otherwise when you remove the tape later there isn’t a perfect join between the wood and they clay.


The only thing I’m on the fence about is whether I’m happy with the final finish on the clay. I like clay because it is a traditional finish - it’s now Fender used to make the dots on the guitars originally, and I prefer it’s natural finish to plastic “pearl” inlays. The one downside though is that as you sand the fretboard it will take on some amount of wood dust, which even with a light wood is visible. Here’s a close up of the last neck I made on the prototype offset:


It’s not quite the quality I want when examined up close, though as a whole it’s not that noticeable, and it does give it a more natural finish, which is the aesthetic in general I want to go for.

On the neck that was already in progress I did do the fret insertion, which went smoothly.


Certainly the more I do things like this the better I get; it is simply grinding like in WoW :) And this is in part why I’ve persisting in the finishing off of this recovery guitar made from bits that were damaged: it just gives me another chance to practice bits like fretwork to help me get slowly better. This neck is now taped up and ready for fret levelling, rounding, etc.

I’ve recently been trying to record a few videos of my guitars (and other lesser known brands, like Fender) through my amp, and I realised that the sound quality recording through my phone is good enough for a “what is he playing” but really doesn’t pick up the tone that I’m nearing in the room very well. So I decided it was finally time to get a proper microphone for such videos.

After much research I ended up ordering a Sennheiser e609 microphone, which is a dedicated amp microphone, and has a low profile so won’t get in the way and be something I trip over all the time. I then needed something to let me plug the microphone into my phone, and for that I’ve ordered a ZOOM U22 interface, which will take both an XLR microphone or a guitar direct in (assuming I’ve understood the manual), and has a USB out that is compatible with the iPhone USB converter dongle, or can be plugged into my laptop.

It’ll be a week or so before they turn up, but I’m looking forward to trying to capture some better tonal examples of my guitars and amp in action.

The spindle sander I ordered last week turned up, and I’ve installed it at Makespace. Many thanks to Matt for the recommendation.


I’m actually quite impressed with the finish this thing gets with the stock 80 grit inserts, much more smooth than I’d expected, which I guess is down to the oscillating motion the spindle does whilst sanding, but whatever the reason the results are great, and it’s going to make sanding guitar bodies so much easier.

At the weekend I went to the Festival of Maintenance, a one day event in London celebrating the people who do maintenance as opposed to create new things, and a look at the issues they face.


The talks covered a wide range of domains, from social housing, to documenting the HTML standards, to gorilla maintenance of your local area when no one else will. It was a very good day, with lots to think about.

I did remind me that I’ve been meaning for a while to push the fact that I’m happy to repair and setup of existing guitars, and that’s something I need to update this website for to make it more prominent. Both it’s a good thing to do environmentally in terms of keeping existing instruments in circulation, but also when I do guitar repair work it’s very educational seeing how other guitar manufacturers have approached things (something that Juha Lottenen was advocating also when I visited him).

A week in the shop

Last week I was working on the recovery offset’s neck: the recovery offset is a guitar being made from the parts that the CNC router tried to ruin that I’ve plugged and am going to finish off as a sort of factory seconds. The motivation for this is to get as much value out of the bits I start as possible, even if they’re not something I could ship to a paying customer (at least claiming it was in tip top condition), but the more I make things the more I learn.

I’ve wanted to try a revision to my workflow for building necks, and rather than try that first on the commission that I have part way through at the moment, I can try it here with less risk, so it’s convenient timing. I guess the thing I want to encourage is that when things go wrong, don’t see that as final for that thing: try to work out what other learning value you can get out of it before you part ways with it.

At the end of last week I’d glued on the fretboard, and I was lamenting that as I’d glued the maple fretboard on, the seam between it and the neck which previously had been a nice tight fit had expanded the moment I added glue (presumably down to moisture?) leaving a small but visible gap. I’d then had to layer on more glue than I would otherwise have used to fill the seam.

The positive is that this application of excess glue did work: the seam did fill and now that the neck has been carved on the back and sanded down, the glue line isn’t too bad. It’s distinct, but there’s no large bits, and it’s fairly uniform. Had I been using two different shades of wood, you’d probably not notice it much, but as I’m using two bits of maple it is visible, but it’s more of a feature.


It looks a little worse there than it does in real life where you can rotate it a little to see that it’s a nice flat transition between the two pieces.

The negative was that removing the excess glue I’d added was a real pain in the neck. In the end it took a lot of sanding and a light amount of chiselling to remove it, and I wished I’d not sanded down the neck so much before doing this stage, as it took out a little more than I’d have ideally removed. So whilst the excess glue did the job, it wasn’t without consequences. In future I’ll do less sanding before I put the pieces together.

The reason this neck is a recovery pieces is that the CNC router got confused and tried to ramp down at a rate into the piece when on the final contouring pass which cased the bit to shatter taking quite a bit of wood with it:


This I patched a while ago using wood glue and sawdust from the same piece of maple. Unfortunately it’s quite a dark patch: I don’t think using this approach works very well on light woods. My hope was that as I carved the neck most of the patched wood would be removed. Alas, not as much came out as I’d hoped:


Still, it’s no worse than a knot in the wood I guess, but you can definitely tell that it’s been patched, and annoyingly, as this is a right handed guitar, it’ll be visible to the player that its been patched. But, all that said, it’s totally playable now it’s been smoothed down, and thus definitely worth taking through the rest of the steps to complete, along with the recovery body.

Carving it went well enough, though amusingly there’s a kink in the grain line that made it hard to see the middle, as the kink was just off to one side, so I had to be careful to make sure I carved it symmetrically. Once carved there was a bunch of sanding to do, and the next stage is to fret it, which I’ll do next week.


The person who commissioned the first offset got in touch asking if he could have a custom logo etched on the headstock. The nice thing about having access to a laser cutter is that I can readily say yes to this. I made a few mock ups in Illustrator of what we might do, and I’m now awaiting a design back of what he wants.

Timing is good here, as I should make the fretboard for his neck this week, so in a week or so I’ll be ready to etch the headstock on this one.

The sanding of both the neck above and the two bodies that I have on the go reminded me how much I hate sanding. With the guitar bodies you have the awkward curves around the horns that you can’t get the orbital sander I have into, so I have to do them by hand. For finish this is fine, but for trying to remove machine lines this is quite tedious. What I really need for this is a spindle sander. 

Encouraged by the experience of such machines by some fellow luthiers I put together a proposal for Makespace to get one for the workshop (basically pricing it, explaining the use case, how it’d be stored and maintained), which was approved, so next week one should turn up for the woodwork shop, and I’ll have to work out how to plumb it into the dust extraction system. But again, anything that saves me manual sanding time is worth it!

A week back in the shop

This week I was properly back in the shop, determined to move guitars forward. Although I am doing some contracting at the moment, I managed a good 4 days in the workshop last week, and it felt good to be back, even if it wasn’t all plain sailing, as we’ll see.

Given it’s been a while since we talked properly about progress on guitars, let’s do a quick recap of what I’ve got on the go right now:

  1. I have the first commissioned offset that is my top priority to get shipped. The body is being sanded smooth in my spare moments, and the birds eye maple neck has been routed out, and next needs to be mated with a fretboard.

  2. I have the factory seconds offset that is made from the body and neck that the Makespace CNC router tried to ruin that I’ve patched up to let me continue to learn/experiment on. This is now acting as a sort of advance party for any new approaches I want to try whilst building the commission.

  3. I have the completed prototype offset where I’m not quite happy with the vibrato. I have all the bits to experiment with the mustang dynamic vibrato (both softer and stiffer sprints and infant a whole second bridge), but whilst it’s tempting to try get this done, I really feel I need to get this first commission off my shoulders. The commission will be a fixed bridge, so it isn’t blocked by me understanding vibrato bridges.

So that’s the three guitars I have in various states of progress right now. I do have a second commission for a friend that is not yet started, and a potential fifth that I’m ignoring right now until I make progress on the first. To say this year hasn’t gone to plan is an understatement.

The next step on the commission offset is the fretboard. In the past I’ve radiused the fretboards using a radius sanding block, but as I mentioned a few of weeks ago, that’s very hard to get spot on, and so I’ve been wondering how I might improve that process to make things more reliable/repeatable. There’s two paths I could follow here: I can try using a hand plane to put in the radius (which is how Juha Lottonen said he makes his when I visited last week, and how Ben Crowe at Crimson Guitars makes their hand built guitars), or I can try CNC routing in the radius when I cut the general shape out of the blank. This CNC router approach being the one I’d already started testing with before going away (when I messed it up due to a combination of rushing and being tired), so I had another go at making a fretboard this way.


I used Fusion 360 to map out a roughing tool path on the CNC router that used a larger 6mm bit to remove the bulk material over the fretboard, and then a second pass that used a much smaller 2.5mm end mill to carve the radius into the fretboard, using 1mm steps. 


This wasn’t as slow as I’d feared it might be, and gave a wonderful finish better than I’d expected: it only took a light sanding to get that looking close to perfect. Given I was doing the radiusing I also got the CNC router to make the inlay holes and the nut slot.


Having got the rough shape done, the next step was to laser etch the guides for the fret slots, so I can then saw them in properly in the correct position. Doing any job on multiple machines means you need to have a strategy for aligning the piece on the second machine so that the cuts you did on the first machine all line up. My technique for doing this on the laser cutters is to tape paper to the bed and score the outline of the piece using the laser into the paper first to give me a reference. When doing this I try to use the most distant pair of corners I can to minimise any error due to angles being wrong, and the score marks in the paper are very fine so you don’t get much margin for misalignment from them either, so all in all it’s a reasonably accurate way to position things. Once the piece is aligned on the paper I then adjust the laser cutter height to have the focus plane on the top of the fretboard rather than the bed, and away we go with the etching.


It all came out well as far as I can tell at this stage; I’ll not be able to properly tell until I string it up if I got it spot on. However, guitar frets aren’t perfect anyway: if they were mathematically prefect they’d not be straight, so there’s some scope for a tiny amount of error in there.

Slot guides etched, the next step is to get the fretboard inlays made. For this maple fretboard I did what I did for the Blues Deluxe guitar I made a while back, where I laser cut the dots out of a sheet of Rocklite that is about 3mm thick. The fretboard inlay holes are notionally 6mm in diameter, so you might expect I can just laser cut a 6mm diameter circle, but it’s not as easy as that. The router bit will not be perfect, nor will the CNC router, nor will the laser cutter, which will burn off some edge material as it cuts etc. In the end after a bunch of test pieces I found that I had to make 6.5mm circles of rocklite to fit snuggly the holes in the fretboard. These I then super-glued home.


The next stage is to glue the fretboard to the neck itself, and this is where I struggled a little bit, and I’m not sure why. I first clamped the fretboard to the neck without glue, and all the seams looked nice and flush with no gaps. But once I added glue, the fretboard and neck joined less well, with a small gap at points along the joint line. I combatted this with more glue along the seam and more clamps, but it wasn’t an ideal situation, and I wish I understood why it looked okay before glue and not after - please do drop me a line if you have any thoughts on what I might do better.

I’ve had one other failure like this before: that neck had quite large gaps I couldn’t fill with just glue, so it’s now sat on a shelf waiting for me to decide its fate. This one mostly seems to have come out okay, but it was a fretful 24 hours whilst I waited for the glue to dry and I could remove the clamps. Even though the gaps seem to have been mostly filled up, I was left with more excess glue than I’d have liked over the outside the I had to sand down. Still, it’s not ended up on the scarp pile, so that’s okay.

The observant of you (that can see the pictures that is) may have noticed a little square of material near the headstock transition, and wondered why that was there. 


The reason is that the last time I cut out a fretboard on the CNC router, although I didn’t radius it, I did do the slope down to the headstock to get that nicely aligned with the neck. However, you can see it goes to quite a fine point on the left hand side there, and this became a weak spot and as such I lost the tip of the point to the CNC router, and then hand to sand in a new transition later on in that build. So this time I left some excess material so there was no weak point when I cut the neck out on the CNC router, and now that the fretboard is glued to the neck it’s supported by the neck material underneath, so I could use the hand router to cut off the excess with no risk of tearing the tip point. This worked a charm, and after hand routing to smooth the transition between the neck and the fretboard it looked perfect.

In fact, it worked doubly well, as I aligned the jutting out edge with the nut slot so I had a nice clear reference point for fret 0 when trying to align the fretboard on the laser cutter for etching the fret slots: a double design win!

Speaking of the slots, the next step was to cut the slots with my trusty Japanese fretsaw, which went smoothly (and slowly, but that’s also why it went smoothly).


Slots cut, it was then time to do the side dots, as this is the last point where there is a nice flat side on the neck for its entire length.


I started doing the carve on the back, but didn’t get to do more than start that process, so I’ll have to finish it in the coming days.

Overall this new fretboard process went well, but I am a little concerned about that gluing incident. The birds-eye maple neck I need to tackle next is not cheap (£50 quid in material) and I didn’t think about that cost when I was working out the spec on the commission. It’s not that the customer didn’t mind paying a little extra to get a nicer guitar at the end, but whilst I’m still a little unsure at how well a neck will glue (I’ve had 1.5 failures out of 5 necks I’ve made, this being a half failure in my book) I’m now very nervous about the next stage I have to do as the cost of failure is high. But do it I must: we’ll just have to see how it goes, and hope the fact that mostly it’s gone right will carry us through, so my next task is to repeat all this again with some wenge and get the commission neck through the same steps I’ve just described.

One thing I realised early in the week was that my fretboard gauges were not in the workshop, but at home. However, rather than head home to get them, or indeed order a second pair for the workshop from Amazon Prime, I realised I could make my own! Sometimes I forget that I’ve levelled up in manufacturing in the last year. So in half an hour I’d designed and printed my own radius gauge:


Better yet, to save others the hassle, I’ve shared the design I made, both as a parameterised Fusion 360 file so you can make one of any radius you like, and as a series of STL files for common radius sizes you can just send to your 3D printer. My first bit of open source hardware :)

As I mentioned at the outset, hand planes have been on my mind this week, so it was fortuitous timing that fellow Makespace member Graeme who is an experienced woodworker spotted me with my hand plane and offered to each me how to set it up properly and get the blade nicely sharpened. Although I’d had a go at this myself in the past, I figured getting a lesson from an expert wouldn’t go amiss, and boy was that the right choice.


Graeme walked me through both how to sharpen and correctly bevel the blade using a veritas guide, and he also taught me how to quickly get the plane set up to the correct blade height and cutting properly. In the end taking my blade from not properly maintained to correctly sharpened took an hour, but the effort was worth it: my hand plane is now a joy to use, and I’ve started practicing trying to radius bits of wood with it. Whilst the CNC is accurate, it’s not very satisfying, so eventually I’d like to have another go at hand radiusing them.


I spent a lot of time this week also trying to help others with weird behaviour they were having on the CNC router; I’m now considered a local expert, which is a little unnerving, but I’m definitely happy to try share what I know. Eventually we got to the bottom of the issues, but all of it just left me convinced that the sooner we get off our current CNC router controller at Makespace to a better understood unit the better.

One person was struggling due to the fact the Fusion 360 post-processor we use generates g-code seems to work for the common case with our router, but in his case didn’t. Given our CNC router controller is a random unit from China with little documentation that’s hardly surprising, so a move to a GRBL or Linux-CNC based controller would certainly help there. We didn’t find a solution, but at least we know how to work around the issue now.

The second person’s issue turned out to be one we diagnosed quickly as a mis-setting in software, but it took us an age to then get things to work, as the CNC router controller cached that old mis-setting even though we fixed it, so until we tried in desperation turning it off and on again it didn’t take effect. And people wonder why I find the idea of using a hand-plane more appealing than the CNC router…

Finally, out of the shop, I got a new toy to go on my pedalboard: the Hall of Fame 2 reverb by TC Electronic. 


I had found that I was struggling with motivation of late playing guitar, but this has been a much needed injection on inspiration. I’ve written some more about it here, and I’ve posted a few videos to youtube of playing with it on.