curtisa

Not a Steinberger type. It's more akin to the Floyd Rose in construction - two knife edges to pivot on and a tremolo block where springs attach Strat-style within the cavity under the pickups. You can treat it the same as a FR in terms of cavity routing, making appropriate changes to the dimensions to suit the bridge. Floyd Rose have routing diagrams on their website you can use as a basis.

Which is why I didn't mention scale length. I was more highlighting the commonality between the two guitars I built that exhibited the same phenomena as something to explore, that being steep string angle and/or tight bending radii behind the bridge.

IIRC you asked a similar query a couple of years back about this issue. I don't recall if there were any good ideas that came out of that discussion. I have built two guitars that exhibited the same issue you have noticed - the lower wound strings would buzz very easily even with excessively high action or exaggerated neck relief. The neck itself had no issues with fret levelling or warps, yet the buzz persisted. The only way to tame it was to change my playing style to deliberately use a softer picking action, which was not really an acceptable solution. I would say, however, that in both those guitars the bass strings behind each of the saddles was kinked over quite severely as it passed over the crown of the saddle. In both cases this was unavoidable and a quirk of their design. I do wonder if the thicker wound strings vibrate differently if the amount of downward force at the saddle is extreme or the amount of stress placed on the string near the point at which they bend over the crown of the saddle is steeper. The first guitar I found this issue was on the low F# of an 8-string (.74 over 26.5" scale length) with a standard Hipshot hardtail bridge. By necessity the intonation of the lowest saddle had to be wound back pretty much as far as it could go, which made the string pass through a right angle bend over a very short distance as it went over the crown of the saddle and through the body. The second guitar that exhibited this issue, the saddles used were the Technology For Musicians headless tuners for multiscale guitars which are quite tall units, and again the string (a low B, a .59 over 26.5" scale length on a 7-stringer) had to make a very sharp bend over the crown on its way to the tuner. There was also some neck angle built into the guitar to account for the high saddles, which would have increased the break angle on the low B string on the saddle even further. Maybe you could experiment with the bridges a bit? If you have used a Tune-o-matic style bridge you could change the amount of downforce/break angle at the saddle by reducing the angle at which the strings pass over (either raise the stop tail higher or place some kind of spacer between the strings and the body if you're used through-body ferrules). If you've used a hardtail bridge, maybe try installing a normal D string in the low-E position, and tune and intonate it accordingly. Doing so would move the saddle closer to the nut, consequently reducing the break angle behind the saddle (not really a solution, but might go some way to proving/disproving the saddle string angle theory).

You'll probably need to do some jumpering and bypassing in order for the piezo signal to make it through the preamp minus a volume pot, but it should be doable.

The Ghost saddles are designed to be paired up with a preamp of some kind. Doesn't the Ghost come with the Acousti-phonic preamp as a complete kit? No need to ground the bridge/strings if you're only installing piezo pickups.

Well played. I'd 'like' your comments, but it would eat into my daily quota.

The forum software uses 'reactions per day', which I take to mean 'the number of like/laugh/win clicks in a single calendar day per user'. It's not clear if that calendar day is in a particular time zone or your local time zone, so perhaps you simply used up most of your your quota within the first part of a 'day' (somewhere in the world) and the remainder of your likes within that same day, even though it was overnight within your own timezone? Dunno. I'd have to request more information from higher up to give you a better answer. Or, y'know. You could just comment that you liked someone's work in your reply to them

You sure you didn't just use up your quota of likes over a 24 hour period, rather than over a single calendar day? I can check up with the forum software to see how it's calculated and defined if you like?

Hi @Urumiko, I've just bumped you up to a Patreon member. Just check to see if you have the necessary accesses.

Correct, and the issue I have with that listing (among others) is that EMC2 does not support USB connectivity, and is unlikely to ever do so. This is more to do with the fundamental operating model of EMC2 requiring a dedicated real-time, latency-free interface to the CNC machine, which USB currently cannot do. Under EMC2 your only connection options are parallel port or ethernet. (BTW, EMC2 has long since abandoned that name and now operates under 'LinuxCNC') Correct. Fusion360 is the software you use to create your 3D model and generate the G-code, which contains all the necessary movement commands that are understood by the CNC machine. You then need a second piece of software to send that G-code to the machine and make it move. There is no direct link between Fusion360 and the CNC.

There appear to be a few variants of the CNC6090 floating around at the moment: https://www.ebay.com.au/sch/i.html?_from=R40&_trksid=p2380057.m570.l1313.TR3.TRC0.A0.H0.Xcnc6090.TRS0&_nkw=cnc6090&_sacat=0 The version that retails for less than $2000AU is usually fitted with a parallel port interface. The 3020 CNC I started out with was similar to this. The one that costs about $2500AU has the USB interface. There is another set of variants which seem to cost $3000AU - $4000 which are a different design and weigh nearly 150kG, so they must have more substantial frames for more heavy duty milling work. If you buy one with the 4th 'A' axis that adds a bit more to the price. If you buy one with a 2.2kW spindle instead of the 1.5kW spindle it costs a bit more again. There also seems to be another version for over $5000 that I haven't seen before, but it supposedly has an all-steel frame and weighs over 200 kilos. It's probably a bit of overkill for your situation.

Yes, the 'A' axis is normally the rotary axis. I think when @MiKro said 'poor rotary' he was more referring to devising ways of holding the work statically on an angle and machining it in increments of 45 degrees (or 22.5 degrees, or 10 degrees etc), rather than on a mechanised continuously variable rotary spindle like in your video, above. Not sure. The CNC you have linked to seems to be similar most of the other ones I've seen for the price. One thing to be careful of is what software is meant to be installed. It's not clear in the listing what software is being installed and what computer it is for. The listing above details multiple conflicting things about the software - that a 32 bit version of Windows is required, that a 32 or 64 bit version is required, that EMC2 works (impossible with USB), disabling antivirus software to install it, using unlock codes to defeat the trial period (sounds a lot like cracked software)... That seems to be a common risk with the Chinese CNC mills; that they are so poorly documented that you just don't really know what you're going to get. If you're adept at tinkering and circuit modification it's possible to work around these hurdles, but if all you want is a plug-and-play CNC machine you'll need to be prepared for the possibility that it will not work straight away without a lot of investigation and researching.

Sounds like neither unit would have worked for you with the limited cutting height, so perhaps it was a good thing that the Amazon seller cancelled your order. 7.5cm of Z height isn't much to work with for a machine large enough to carve a guitar neck. The eBay CNC6090 unit has a maximum Z working height of 140mm. Maybe have a look at that one? The main drawback with the generic Chinese machines on eBay is that the software will be somewhat restrictive as to what you can use and how you use it. After sales support and documentation will also be very poor.

Sorry to hear that. As you've probably already found, this is a particularly expensive hobby for most people. My first CNC machine was something called a 3020, which just refers to the maximum travel for the X and Y axes (Y=30cm, X=20cm). They were very common on eBay at the time. It actually wasn't all that bad provided you only worked within its limitations. There were other models available that were bigger - 3040, 6040, 9060. I think the biggest models were around the $3000 mark.

I applaud your keeness and tenacity to jump in head first into such a project. And welcome to the world of CNC with your new purchase. I would advise caution, however, to not get too carried away too quickly if this big step also involves a lot of first-time experiences for you. Have you built a guitar before using more conventional methods? Are you familiar with the fundamentals and concepts employed in 2D drafting and 3D modeling? Purchasing a CNC with the intention of building a guitar with it is a bit like buying a Steinway piano and immediately booking a concert recital at Royal Albert Hall. I would highly recommend you start with small projects on the CNC and work your way up to a guitar. It will be a long journey and you will make mistakes, but you will learn a lot along the way. Personally I would use a thickness (drum) sanding machine to get the top from 4mm down to your required 2.6mm, and then use the CNC to work with the piece at the correct thickness. While the CNC could do it for you, the main complication will be trying to secure a wide, flat, thin, flexible piece of wood to the CNC table so that there are no bulges or warps that would prevent the top being milled down uniformally by the CNC. The only way I could see that being achievable would be the use of a vacuum table, which is not a cheap or easy exercise to add to a CNC, let alone the X-carve. Assuming you could hold the wood down perfectly flat and securely, the CNC would likely take a long time to zigzag back and forth to remove the excess material from the wood. Ideally you would use the biggest cutter you couldfind to minimise the number of back-and-forth motions and reduce the amount of time it takes, but I believe the X-carve uses a regular router with a 1/4" collet, which limits the size of the biggest bit you can install to perhaps a 1/2" diameter cutter. I would expect such a machine to take over 30 minutes to complete the task, which is probably about 20 minutes longer than it would take using a thickness sander. Granted it's not a particularly excessive amount of time, but there is a lot of complicated work to plan and execute for both yourself and the machine.

There is no distinction between 2.5D and 3D with regards to the machine itself. A 3 axis CNC will move exactly where it's told to move by the milling code fed to it from the controlling software and/or hardware. The details of how it makes those moves is governed by whatever software you use to generate the milling code in the first place, which is entirely independent of the CNC machine. Think of the CNC more as a car, and the milling code is the driver. The driver of the car determines how to accelerate, brake, change gears, approach corners etc. But without the driver the car goes nowhere. While you could thickness an acoustic top using the CNC, it would be extremely time consuming and wasteful. A more logical use of the CNC for an acoustic guitar would be to cut the sound hole in the top, cut the channels for the sound hole rosette and purfling, rough shape the bridge from a block of wood, rough shape the neck profile, slot and radius a fret board, add inlays etc. The CNC will not take away the requirement to bend and glue parts together on an acoustic guitar, so there will still be a significant manual labour component to building, even when applying all the practical uses of the CNC to the construction of an acoustic instrument. You also need to be aware that using a CNC does not automatically result in a group of guitar components that just magically snap together and make a guitar, nor is it a case of pressing a button to generate an unlimited supply of guitar parts in a matter of minutes. There is a significant investment in time required in designing the guitar parts, experimenting with cutting speeds and feeds, optimising tool paths, reiteration of work, problem solving with regards to failed cutting operations etc that can easily overtake the time it would require to manually build the equivalent guitar using more conventional hand and power tools. It can take days just to design and generate the milling code for one guitar bridge that comes out exactly like you intended it to; you could easily make several in the same period of time if you do it by hand.

Huzzah! I've been knighted! Turd polishing rarely results in a glossy shine Very soon. There'll be a photo dump coming your way in the next week or two.

Very much a worthy win, Scott. Well done! Now start sucking more. PS - I'm just fishing for you to call me 'Sir' as well.

Oooerrrr. Behave, Mrs Slocombe! *Ahem* Nice carving work BTW @Bizman62

Great to see something a little left of field show up around these parts You certainly aren't afraid of getting your hands dirty in a variety of different instrument construction projects. What's next - bagpipes pehaps?

I wrote up a post detailing the routing strategy I employed for shaping bodies and tops with reference to a template which you may find useful. The idea is to only route certain curved shapes halfway from the underneath on the router table, and then flip the body over and route the remainder from the top.

Awl see what I can do...

'Awl take two, thanks' 'Is that awl you've got?' 'He's as wise as an awl' 'We could awl play a game of darts, but I've just turned them in to something else'

I'd personally reverse your order of assembly - shape headstock, attach cap, trim edges. You could have the headstock shaped, glue the cap to it cut a mm or two oversize, and then use the headstock itself as the template. The cap timber would then be largely reinforced by the substrate it was permanently attached to, and the tiny amount of overhang left to remove would further minimise the risk of a piece disappearing into thin air. The most fragile component undergoing the most destructive operation needs the most support. Shaping the cap first is pretty risky, as you've discovered. Doing it last, however, once it's got something to 'hang on to' will give you a better chance at success. Looking at it another way from your cabinet workshop experience, you wouldn't pre-trim a Laminex/Formica sheet before you glued it down to a bench top. You'd glue it first then run the laminate flush trimmer around all the edges to get rid of the overhang.

I generally wouldn't bother with adjusting the speed of the router, but that's just me. I have three routers in the workshop, and the only one that is variable speed is the little Makita trimmer. I've left it set at '5' since I got it and have never adjusted it. According to the manual that equates to about 27K RPM. The other two routers I have just run flat out at 22K RPM. The smaller router gets used on smaller jobs. That also equates to using smaller bits and taking smaller bites, so I guess the higher RPM makes more sense to keep the chip load up on the smaller diameter bits, where the surface speed will be much lower than on a larger bit. But I've also never got too hung up on fiddling with variable speed, primarily because until I got the trimmer I never had the option, and secondly because when I did try playing with it it didn't make such a ground breaking difference to my work that I felt I should be adjusting it all the time.