Making a Baritone Harp Guitar

[Dave White]

Acoustic baritone guitars are mysterious beasts indeed – so far I’ve made three. They are usually designed to be tuned five semitones down from a “normal” guitar and five of their six strings share the same range – E to B or the bottom five strings of an EADGBE instrument. The baritone low string is what it’s all about and there are similar problems to acoustic bass guitars here. To go down to the low A you need a long scale length, around 29” or more, a big guitar body to help support the low frequency notes and the choice of back and side woods that support the low frequencies but give clarity and separation between the strings. A reverb rich Rosewood may well not be your friend here. You also hit the problem of string gauge – beyond 0.070” and you are into double wound strings which sound very different purely acoustically. Some builders believe that the physics of longer strings and heavier gauge mean that they need a much higher string tension than on say a 25.5” scale length instrument even in the same note range to function efficiently. I had a long e-mail discussion with David Berkowitz about this – the string gauges he uses on his baritones have an overall tension of well over 200lbs compared with a more normal 160-170lbs and he uses a bridge doctor to “shape” his sound. I currently think that what is happening here is that these high tensions are giving a power to the more normal mid-bass notes that makes the ear think that it hears the really low notes more strongly – with these string tension the baritone will certainly “roar” but, you have to brace the top much more heavily for this and I personally don’t think this to be a very good trade off overall. I brace mine as for a normal Jumbo sized guitar and aim for around 170lb string tension. Purely acoustically I have yet to hear a baritone that works really well on a low A or G note, plug it in and that’s a different story, and purely acoustically I personally think that baritones work best in the C-C range.

Then I had a “eureka” moment about other ways to fool the ears for these low notes – on my hollow arm harp guitars and hollow necked acoustic lap-slide guitars the bass notes are really deep and resonant, more so than on normal guitars. The extra length and body volume of the hollow neck/arm adds resonance and reverberation. So I got to thinking – what would happen if you made a baritone with a hollow arm? Prior to the server crash Pete Howlet posted a video here of a guitar with a hollow arm (though not a baritone) that he had made to enhance bass response so great minds think alike - honest guv, I had been thinking about a baritone with hollow arm for many months before I saw that video. I also started thinking about the long scale length of baritones and that this was really there for the lowest note so wouldn’t a multi-scale make great sense here? This has been done on baritones by many makers. Having a hollow arm without any sub-bass strings seems a bit of a waste and I noticed on my other hollow arm harp guitars that the addition of sub bass strings helps the fretted bass notes even more via the sympathetic vibrations – another trick on the ear. The multi-scale approach now makes even more sense as the transition to lower sub-bass strings calls for longer scale length string by string and with a multi-scale this means that the bridge design and the way it drives the top is much more efficient as the slanted bridge stays more on the central active area of the lower bout.

OK – I was sold on the idea and it’s time to see if it works. First I need a name – easy, my baritones are called “Treebeard” so this has to be “Fangorn”, older and deeper! Next is scale lengths – I’ve found that a 30mm difference between scales with the ninth fret as the orthogonal one works well so 705-735mm it will be. Then it’s the playability of such a large beast – the biggest influence on my harp guitar designs has been Stephen Sedgwick whose instruments are smaller and more compact than most, so applying this to my “Treebeard” model results in a guitar that is probably still more compact than a Dyer, and in addition I will use a Manzer/Cumpiano type wedge with the treble side deeper than the bass side. I thought about a cutaway but I want maximum body depth here (widdley notes above the 15th fret are probably not going to feature much in the repertoire of this instrument) and beveled arm rests are out for the same reason as well as that to my eyes they look butt ugly!! So this just leaves the question of how may sub-bass strings and the wood choices. After considering tunings, playability and string load on the top I’ve settled on four sub bass strings. The main strings are probably going to be in AEADEA or BbFBbEbFBb tunings and this would give the option of something like EFGD with the first and EFGA or FGAEb with the second that will hopefully work wit 0.070” as the fattest string. For woods I have a nice Lutz spruce Weissenborn top from Shane Neifer that can be made to work with some creative wood inlay on the hollow arm, and I have a large slab of African wood that was bought over 30 years ago by a local wood turner that I think will work - I think it might be Ovangkol.

Baritone harp guitars aren’t that common and I haven’t managed to find any with a hollow arm so this could be new territory. So it’s on with the build!! First it’s the plan – just the outline and main string positions for now, the bracing and sub-bass string details will go on soon – and templates:



Next I made a dowel mould. I recently bought a bigger bandsaw that I can use to re-saw and this is part of the packing case - I never throw anything useful away:



This time I’ve drilled the dowel holes all of the way through the base so the mold is “reversible” – I just flip it over and push the dowels through when I want to work on the back. I can also set the dowels at different heights which will be useful when it comes to doing the “wedge”:



I used the new bandsaw to slice the “African wood" and here are the back pieces having been thicknessed joined using hot hide glue and the “tent” method:



Here’s the back after the glue has dried and clean-up:

[Kim]

MrBurns.jpg (33.04 KiB) Viewed 26517 times

Excellent!

Looking forward to this one Dave

Cheers

Kim

[DarwinStrings]

Always a pleasure to see something a bit different Dave, will watch with baited breath.

Jim

[P Bill]

I'll be watchting this one as well. You're lucky to be able extend your opperations indoors. That's a no go area for me.

[Dave White]

Kim - D'oh!!

Jim - thanks.

Bill - I try and operate a luthiers version of Parkinson's Law

Now for the sides - the treble side is a single piece including the hollow arm side and is bent in the “fox-bender” leaving the hollow arm side protruding:



This is removed from the bender and tested in the mold:



The rest of the bend is completed by hand on the hot-pipe and the fit tested in the mold:




The bass side is then bent in the “fox-bender” bending only the waist and lower bout:



The rest of the bend is done by hand on the hot-pipe and here are both of the sides in the mold:

[Dave White]

Some more progress:

The hollow arm peg-head is to be made from mahogany “scraps” from the Edwardian bureau that has produced wood for a number of my instruments. Firstly four pieces are glued together using fish glue to form the block that goes inside the hollow arm:



When dry they are shaped into a sloped “bull nose” at the front end and then glued to the peg-head using fish glue again:



The peg-head is then shaped, the recesses for the sides cut and a piece of English walnut binding glued at the end of each side. The Madagascan Rosewood overlay is then glued on using fish glue:



While the glue dries I move on to the end-block and end graft. The Madagascan Rosewood centre strip is glued onto the English lime end-block using hot hide glue, followed by two pieces of the English walnut binding:




The sides are then cut to the right length and glued using hot hide glue to the end-block with a piece of bwb purfling in between them and the end graft to be mitred into the side purflings:




The hollow arm peg-head veneer is trimmed to shape and the holes for the tuners marked and drilled and the tuners checked for fit::




Four 6mm holes are then drilled in the back on the peg-head block for carbon fibre flying buttress braces that will be fitted later:

[liam_fnq]

excellent.

I haven't seen the butt strip done that way before.

keep the updates coming.

[Allen]

Dave is always thinking outside the box. Stay tuned and your sure to learn something.

[ozziebluesman]

Yep, this is such an inspirational build.

Thanks Dave, I will enjoy the journey to the finished instrument.

Cheers

Alan

[Dave White]

Alan - thanks I hope you find it a useful trip.

Allen - the end-graft is outside the box, or at least at the edge

Liam - I've done this a few times now on harp guitars and lap-slides as turning these things on end to cut end grafts has it's challenges. Also if your sides are a little short you can do it this way.

Some more progress:

The hollow arm peg-head is then glued onto the sides using fish glue:




Now it’s time to make the neck-block. The 7mm bolt holes are first drilled in the mahogany block allowing a little wiggle room for the 6mm bolts:



The hole for the fingerboard extension support yo go in is cut and then the facets for where the carbon fibre flying buttress brace rods will go are cut and the 6mm holes drilled:



The sapele fingerboard extension support is then glued in with hot hide glue:



The angle on the front face of the neck-block is sanded in and then the tricky gluing and clamping procedure starts – this would have been even trickier with a three piece rim-set. This takes time to set up and clamp so it has to be a fish glue joint. First the middle and bottom clamp is put in place:



Then the rim-set is flipped over and the top clamp put in place:




Finally a small mahogany wedge is glued in place (an off-cut from the faceting) to allow for the geometry of the hollow arm and neck-block:

[Nick]

What a great way to kick off the "New & Improved" Forum Dave. Thanks for taking the time to document & show us this build.

[kiwigeo]

Looking good there Dave.....ya make it look soooooo easy

[Stephen Kinnaird]

Totally cool Dave.
Really looking forward to watching this unfold.

Steve

[Dave White]

Nick. Martin and Stephen - Thanks.

Here’s the rim-set:



Now for the top – I’m using a Lutz spruce Weissenborn set that I got from Shane Neifer of High Mountain Tonewoods and the edges have been prepared and joined using hot hide glue and the “tent” method. They are weighted down using some Eitimoe slabs (that will get re-sawn eventually into back and side sets):



When the glue has gelled the excess is removed and tapes put on:




It’s then left for the glue to dry overnight – next to it is another Lutz spruce Weiss set that I joined at the same time that will be for the Parlour harp guitar Fimbrethil (the lost wife of Fangorn) that you’ll hear more about later:



Next day the top is thickness to just over 3mm and cut closer to shape – the harp arm will be “added to” later:



Now I’m doing some “odds and sods” stuff that will be used later. The fingerboard will be a nice Madagascan Rosewood one that Colin Symonds kindly gave me. As some of you know I believe passionately in using every scrap of tonewood that I can (this stuff is a scarce resource, precious and should be revered) and I want to get two fretboards out of this using a method I’ve used before – one for Fangorn and the other for Fimbrethil. First of all I cut two binding slices from each side of the board – about 2mm thick:



Then the central portion is cut into two:



These two pieces will be glued to some sapele pieces to make full thickness fingerboards and as they will be bound, the sapele won’t show. This makes sense to me as a better use of scarce wood, and with heavier woods like ebony you end up with a lighter more resonant composite fingerboard with enough depth of the main wood to hold the frets as if it was not a composite:



Fangorn’s fretboard is glued to the sapele backing strip using fish glue:


For the next “odds and sods” the mahogany neck-block scarfe joint is cut and glued using hot hide glue:

[Dave White]

Now Colin Symonds is a good friend and a man of great generosity. When he was making himself a Wells-Karol circle cutting jig he made one for me too and this is the first time I’ve got to play with it – it’s the dog’s b’s !! First task is to route the rosette channels for a circle of “African Wood” that I made using the new cutter and then checking the fit and gluing it in:




Next I set the jig to cut the channel for the outer ring of rope purfling using a 2.5mm bit and checking the fit:




Then the inner ring channel is routed and the purfling glued in:




When the glue has dried the rosette is scraped flush:



Now it’s time to decide how to extend the harp hollow arm. I thought long and hard and came up with many different ways involving intricate mitred purflings but decided that all of these would be too distracting on the eye. So for simplicity of visual impact the join needs to be part of the hollow arm soundhole and here’s how it goes. The soundhole is going to be a “dark hole” similar to the one I did on my last harp guitar but using Madagascan Rosewood, so using the new circle jig I cut a curve into the harp arm and a matching curve in a spruce piece from an off-cut of the top with matching grain. The main part of the top is then glued to a piece of Madagascan Rosewood using fish glue:



When this was dry, the extension piece was glued with a piece of rope purfling in between:




Next I experimented with the jig until I could cut the right sized 2.5mm wide circular channel that joined up the rope purfling as a match. The jig was used at this setting with the 2.5mm bit to route the channel on the hollow arm to the depth of the spruce top:




Using the same setting and bit a circle of Madagascan Rosewood was cut – the external circumference matches the inside edge of the rope purfling ring:



The spruce on the inside of the hole in the hollow arm was then routed away carefully leaving the backing Madagascan Rosewood strip exposed:



Using the jig pivot pin the rope purfling and centre insert are glued in using fish glue and then the pivot taken out. There is a nice grain line in the Madagascan Rosewood that “joins” the crossing purfling and although it is straight, when the centre hole is routed out the eye should follow this curve – well at least that’s the plan:

[Dave White]

Time to cut the centre out:





Now onto the back – I’ve decided to put the centre strips in first this time. In the past I’ve done this after bracing but the hot hide glue seems to give the back a slight dip along the centre line so I’m doing it first on the assumption that gluing the bracing on will bring back the curved surface. I’m using some cross grain Colombian Rosewood strips that I had in the workshop:



Next I shape the bottom of the European spruce X braces to a 10’ radius, notch and join them and check the fit:



Then the braces are glued in using hot hide glue on at a time:




Then I did some work on the top. The soundhole patch is made from a top off-cut and is glued on using fish glue – the circle cutter jig pivot pin is used to line things up together wit a brace clamped along the top. These are then removed and the glue left to dry:




The main soundhole can now be routed out:




Now I can draw in the rest of the top bracing:



The European spruce X braces are profiled to a 13’ radius, notched and then glued in one at a time:

[ckngumbo]

What a great thread!
Thanks

Rob

[Dave White]

Rob - thanks.

Next the Rio Rosewood bridge plate is glued on using hot hide glue and then the spruce X brace cap is glued on as a rub joint using the same glue:




The ends of the X braces are trimmed so that the top fits the sides and the sides are profiled for the top curvature. The position of the X braces where they meet the sides are marked and the top linings glued in using fish glue:





Then I did an initial carving of the top X braces and notched the linings for the brace ends. The top was attached to the rim using brown tape and the rim-set removed from the mold. The rim-set is then profiled to accept the back using a chisel, cabinet scraper and belt-sander. The slope for the Manzer style wedge is a gentle one with a 10mm height difference between the sides, and the hollow arm has been tapered. A mahogany wedge needs to be glued to the bottom of the peg-head block later:



Next the backs’ X braces have a Madagascar Rosewood cap glued on and are rough carved. The ladder braces are profiled, notched and glued on using hot hide glue:




The ends of the back braces are trimmed so that the top fits the sides and the sides are profiled for the back curvature. The positions of the braces where they meet the sides are marked and the first of the back linings glued in using fish glue:



The mahogany wedge is glued to the bottom of the peg-head block – this will be profiled later:

[Kim]

Great work Dave and a most enjoyable thread. Looking forward to the next installment.

Cheers

Kim

[Dave White]

Kim - thanks.

The last of the linings are glued on:




Now it’s time to make the side soundport. First a 2.5mm circle is routed in a piece of Madagascan Rosewood for a rope purfling ring which is glued in with ca glue and put through the thickness sander:




The outer ring is then cut with the circle cutting jig and this is then glued to a matching backing strip of Madagascan Rosewood using fish glue:




The inner ring is then routed out:



The position of the soundport is marked on the rim-set and a hole drilled for the 6mm pivot pin and a backing mdf piece made and stuck with double sided tape inside the rim-set with the pivot pin going through it:



Then taking a deep breath and clenching all buttock muscles the circle cutting jig was used to cut a hole in the rim-set matching the outside diameter of the insert:




The soundport insert was tested for fit and then glued in using fish glue:

[Nick]

Then taking a deep breath and clenching all buttock muscles

Dave, it's good to see you are another person that employs my technique when operating power tools on on near wood that has had a 'few' hours of my time spent on it. Think I do this everytime I bring a closed box up next to the rapidly spinning binding channel cutter . MInd you, I have got a "tight set of buns" (according to my partner) as a result
Thanks again for the pictorial on this build, it's going to be one sexy instrument when it's all done.

[Allen]

Was there some sort of backing to the sound port binding as you are gluing it in? or will you add it latter. Don't imagine it was the MDF.

[Dave White]

Nick - I'm afraid that my buns are more like jam donuts in their relaxed state

Allen - I probably didn't explain it too well. If you look back at the pictures there is a "disc" of Mad R/w with the rope circle inlaid in it that fits exactly into the side "port hole". This disc was glued onto another 3mm thick matching piece of Mad r/w and then the sound-port hole routed out of both pieces. So the disc is proud of the base by the side thickness. The second Mad R/w strip is on the inside and glued to the rimset with the disc pushed through the "port hole". Does this make more sense? Here's a picture from the inside

[Allen]

Heaps more sense. And a bloody good solution. Thanks for the clarification.

[Dave White]

Next job is to glue in the 6mm carbon fibre rod flying buttress braces and side braces for the hollow arm peghead, and the carbon bar struts – hopefully this will keep the hollow arm rigid under the sub-bass string pull:





Then the main 6mm carbon fibre rod flying buttress braces and side braces are glued in using fish glue:



The back braces are fully carved and the back glued on using fish glue. When the glue has dried the back is routed flush to the sides:



The rest of the top bracing is glued on using hot hide glue and carved:









The top is then “voiced” and when I’m happy I sign and date it:





Then the top is glued on using fish glue – there are a few places where I trimmed the top a little too much but this will all be taken care of when the binding is routed and installed: