Sunday, April 18, 2010

Segovia Syndrome and the Hyper-Tension

Nigel's guitar is in for a quick fix. The instrument in question is a 1991 Ramirez SP semi professional model. Cedar Top, solid Indian, 650 mm scale, with all the usual Ramirez specs.
Nigel is a strong student with well developed, mature technique. He's asked if I can do something about the feel of the strings in the first position. His instructor noted how difficult it was to depress the strings, and giving it a go myself, I have to concur that a barred Fmaj chord is a real chore!

Looks like this isn't the first time the instrument needed nut work. A nicely made shim of laminated cedrella has been glued to the base of the nut. There's evidence of the removal process where the rock-hard Ramirez urethane finish has chipped off.

A look at the bridge. Not a pretty picture. There's little room left to reduce the saddle, the bridge itself is tipping forward, and one can see that there's another shim down in the saddle slot.

Despite this, the instrument produces an agreeable Spanish tone with good volume. This guitar displays the hallmarks of Madrid-style building geometry. The top is essentially flat, with characteristic deformation pulling up behind the bridge, sinking in front. The neck shaft is pitched forward from the plane of the body to a remarkable degree, rising almost 5 mm at the nut. (This is more than twice the rise I build into my instruments.) Coupled with the tapering fretboard, this forward tilt means that a straightedge laid across the frets will meet the bridge at a point only 1mm above the surface of the top. Remarkably, this isn't the most "extreme" Ramirez setup. Richard Brune reports having seen necks with as much as 10mm (!) forward tilt.

The action? Getting up there towards 5mm. That's about the upper limit I'd consider normal. Not much to be done with it, considering the bridge's condition. I suppose one could make a new, lower bridge. But why? This instrument is designed around having the strings 11 -12 mm above the soundboard. It would lose sustain and the powerful open vowel sound. You'd be left with a mediocre flamenco guitar.

Can I get on my high horse for just a minute? I'm not the first luthier to bemoan the Segovia influence on young musicians. It's been noted that with the Maestro's passing guitarists were suddenly free to pick other instruments with less macho setups. Why is then that in the past two years I've met three young guys (probably born after Segovia's death),who've bought Ramirez cedar tops? Not that I have anything against the company mind you. It's just that these instruments are designed like race cars. They're tailored to all-out volume in big halls, to be played aggressively with strong hands. Also like race cars, they're essentially impossible to repair. The flat top, high bridge, forward neck bow, heavily tapered fretboard with lots of relief, cedar top - all conspire to start driving up the action from the moment the instrument is first strung. Twenty years is the lifespan. Because of the integral-neck aspect of the Spanish system, in another year when Nigel's action is becomes unbearable, it will be necessary to peel the back binding, work the back free from the interior portion of the neck block, tip and pare off the excess, etc, etc. Forget about planing the neck. The extreme forward tilt and tapered board mean I'd have to remove almost 2mm at the nut end. It'd look freakish. The other option would be to make a completely new fretboard with a reverse taper, from 5 mm at the nut to 7 at the sound hole. Again, not ideal.

I guess what I'm saying is that Segovia didn't have to worry about this. Mr Ramirez was always ready to trade in for a new instrument whenever it was necessary. And it happened frequently. There are quite a number of Ramirez 1A's out there with Jose III's label certifying them as having been played by the Maestro. Contrast that with the 25 years of service that Segovia got from his original domed-top, lightly built Santos/Manuel I, or the almost 30 years from the Hauser.

I'd like to see young players educated about this situation. If you're going to lay out the serious dough for an guitar to last your lifetime, you deserve one that will make elegant repair possible. Don't try to drive the F1 car when the BMW, Mercedes or Saab will give you the street-level performance you need.

Let's do what we can. A stack of feeler gauges at the 1st fret tops out at .035". Not so bad actually, about .008-.010 heavy.

Wow. High nut equals big fret wear. More pressure needed to note means more friction. This looks more like steel-string type wear. The green tape is there to keep strings from rubbing the finish. The urethane on this guitar seems almost crispy. It's flaked away on the binding in several areas leaving voids with sharp borders.

I've popped out the nut and we can see the wood sandwich. I've stroked some pencil lines to verify where I'm removing material.

Here's a teensie-tiny jointer system. I really don't like to see nuts that have been sanded out-of-square. Worse yet is a nut with a radiused bottom from rocking one's grip. They've got to make good contact all the way across the neck. This is a mill file clamped against an oak strip in my mini mechanic's vise. With this setup I can very accurately taper things (to the bass side, in this case).

After several alternate checking/shaving cycles I've got things down to where I'm satisfied. I'm pressing at the 1st fret. You can see how much airspace is left over the 2nd fret. At first I thought the neck was bowed. There is some considerable relief through frets 3-9, but it actually has more to do with the fretboard taper and forward set of the neck. I did do a little filing on the fifth and sixth string grooves, mostly to improve the line.

One other side-effect of a hard action at the nut is the tendency to negate the compensation factor built into the string length at the saddle. The string is stretched on both sides, pulling it seriously sharp. Did I mention that Nigel has been experimenting with extra-high tension strings?!!! The forward tipping at the bridge has eaten up about .5mm of the compensation so far. I'll recommend he switch to as low a tension as he can stomach.

One positive is that the lowered nut has actually brought the action down to 4.5mm on the bass, 3mm on the treble.

Wednesday, April 14, 2010

Ashwood Staves

I've been assembling another lute bowl this week using this material. It's flat-sawn ash. Over the weekend I had a great educational experience at the A&M Wood Specialty warehouse in Cambridge, Ontario. It was a musical instrument makers open house with the guest speaker being noted Toronto luthier Michael Schreiner. Musical accompaniment was provided by lutenist Terry McKenna. Mr. Schreiner offered a slide presentation detailing his endeavours in various museum collections to document historical lutes and early guitars. One example of interest to me was the extremely early Laux Maler lute in the Cite-Musique, Paris. The wide ribs of flatsawn ash reminded me of Kenneth Be's six course instrument constructed by Grant Tomlinson. I had the pleasure of listening to a recital of Spinacino played on this lute in Vancouver a couple of years back, and I was extremely impressed by the tonal quality.

As luck would have it, I have a supply of (nearly) flatsawn ash offcuts from a project I undertook a few years ago. The wood was salvaged from a pew removed from the upper loft at Melrose United Church in Hamilton. I used it to construct a scale model of the church organ to present as a retirement gift for music director Faye Grinberg. I also fabricated some Arts and Crafts style furniture for our living room. At the time I had resawn some 1/16" strips to reduce the amount of hand planing necessary. This material has been seasoning for more than eighty years, and I have just enough for one lute bowl. It's not sequentially cut, but I hope that the haphazard grain patterns will add an air of jaunty playfulness. At least that's what I'm telling myself.

I'm scraping the roughsawn staves to near near finished thickness. I wouldn't try to do the whole batch of 15 at one go, as this would lead to crippled thumbs and boredom. Scrape one, bend it, glue it, and repeat. I didn't photograph it, but I have a little digital thickness gauge I use to ensure uniformity. It surprises me how many woodworkers don't have a steel cabinet scraper in their arsenal. It's so ubiquitous in luthiery that it always takes me by surprise when someone asks me how to use or sharpen one. Still wishing I had a third hand for moments like these, the picture shows me pushing the scraper away from me, pitched slightly forward into the direction of cut. Of course, I'd usually have both hands on the blade. For aggressive scraping I flex the scraper with my thumbs. As I want to keep things quite flat here, I'm not imparting much camber. Like the scraping jig? I got tired of messing with F-clamps. This lets me take advantage of my exceptionally long arms.

I enjoy scrapers so much, I've developed quite a
collection! They're made of springy carbon steel. Some are ground into curves for specific hollow areas. I've radiused the corners on others to make them "safe" in areas I don't want to accidentally leave scratches on adjacent surfaces.

The scrapers rely on a very fine hook on the square cutting edge that acts like a microscopic plane. The hook is produced my carefully deforming the sharpened corner.

I start by removing the old hook using a smooth mill file. The object is to produce a nice hard 90 degree square. I file both the sides and the top. I've demonstrated the various Veritas scraper jigs at numerous seminars and wood shows, but in my own shop I've always relied on my hands. Hmm. That might be the reason my knuckles always look so chewed up. (You can see them of course only when they're not covered with dried glue.) But seriously folks, note how I use the knuckle on my index finger as a fence to keep the file square to the edge.

For really rough scraping I'll use the surface produced by the file. Most of the time however, I'll polish things up with my 800x ceramic stone. I like these stones because they don't get scored with grooves, I can use them dry, and they only require the occasional cleaning with soap and water. Again, I hone both sides and top edge. If I'm using the scraper in a very refined way- perhaps to level a finishing flaw, I will forgo the hook and use it with a square corner.

My burnishing tool. So unlike the fancy expensive chrome-polished beauties available for $30 or more. I made it from the shank of a 3/16" drill bit, heated to straw-brown to harden it. Yes, I should buy a better one but I've had this for twelve years now, and I'm used to it.

I lubricate it with with a little paraffin wax. Mr. Lee always suggested using some grease from the side of one's nose. I could never bring myself to do that, especially in front of a crowd of people.
Note that very little pressure is necessary. Most novices really crank down on 'er. Note also that I'm again using my finger tip as a fence. This needs to be a controlled maneuver. Too much misplaced force could require a tetanus shot.

That's better. Little wispy shavings - not granular dust. I have to be gentle with this wood, as I don't have extra to play with. It's special to me as it's absorbed the musical energy of eighty years of Christmas candlelight concerts, countless performances of the Messiah, and my own voice.

Monday, April 5, 2010

Carving a Rose

My tool kit for rose carving is pretty simple.
Frequent honing is important, so my 8000 grit ceramic stone is close by. To outline the borders of the rose I use an inexpensive Olfa compass cutter. I find this tool gets the job done, but there's more flex and play in it than I'd prefer. One day I'll make a more substantial tool. To cut through the soundboard I use two home-made punches, 1.5 and 2 mm. These started life as high speed steel drill bits which I ground down and fixed in wood handles. The 1.5mm is wrapped in cushioned grip tape which makes it easy to hold and prevents wrist strain. I have a Japanese skew chisel for the chip carving used in borders, and a Veritas Carving knife. I hesitated before buying this as it's a bit expensive but I've found it far superior to the X-acto handles. It's loaded with a #11 surgical scalpel blade.

A close-up shot. The cutting mat is an important part of the set-up too. I've screwed mine to a piece of plywood that I can carry around the most comfortable location I can find. Comfort is necessary due to the amount of body control required. I try not to work longer than an hour at a time on this stuff in order to prevent cramping and loss of dexterity. It's engrossing though, and time goes by pretty fast due to the focused nature of the work.

Here's a rose pattern I designed for my current batch of four 7 course lutes. I laid it out on 1mm graph paper and colored in the negative space. On one side I sketched in the over/under cuts to get a better look at what the weaving effect would do. It's glued in place on the inside of the lute.
Photocopying the pattern can distort things a bit, as can the uptake of moisture from the glue, as can the stretching effect of pressing it in place on the sound board. I suppose the smart thing to do would be to make up a CAD pattern and take it to a laser cutter. But where's the fun (and eyestrain) in that?! Seriously though. What's the point in making a lute at all, when a computer generated midi synth can do the playing for you? In the words of Cosmo Kramer, "Why fly a kite when you can just...pop a pill? Ee-yayuh!"

The scalpel is used to initiate the outline of the cuts. Its most important function is to provide a little compressed area for wood to move into when the piercing cuts take place. Think of digging in clay with a flat spade. Because of the linear nature of the grain , if any pressure is placed against an unsupported segment the connective structure will fail and (pop!) out breaks a little sliver. Then I've got to glue it back in place. Bother.

I was surprised to hear Grant Tomlinson describe his technique as "punching out a rose". It's an accurate description though. The cutting takes place by chopping down vertically with the tiny chisel which is beveled at an extremely low angle, probably 10 degrees or so. Care must still be taken to provide relief cuts and to respect the grain direction. Just a note on scale, I used the macro function on the camera -these little squares are only 2mm wide.
Eventually all sides are loosened and the chip pops free. By keeping the cuts vertical, there's little deviation in the pattern when I turn it over to complete the visible side. I use the scalpel to clean up any fuzzy cuts. I then use the skew chisel to make the over and under pattern, cutting half way (1/2 mm) through the thickness of the board. The back side of the rose is bolstered with a number of 2mm square spruce splints as well as a portion of the bracing structure. I dye these supports black to make them less visible.
The surface of the rose is painted with a thin spirit varnish to seal things up a little bit.

Here's a finished rose of the classic "Knot of Leonardo" variety. The pattern came from Lundberg's Historical Lute Construction. I can pick out the problematical areas, and it's not quite as crisp as the laser would cut. I'm still proud of it though. I recall bringing in one of my early efforts to work. One of my co-workers was a proficient carver of decoys, caricatures and the like. "You're nuts", he said. He's perfectly right of course. Few other crafts push the limits of wood as a structural medium to such outlandish extremes.