Getting To Know Roubo -- Leg Vise Tweak

When I made my Roubo bench I used a roughly 5" section of 3/8" steel rod as a pivot pin for the parrallel guides on both leg vises. I don't move the pin often, but when I do it's a bit of a pain to grab that little pin, especially when the vise is almost closed.

I'm pretty sure I got this idea from a picture of Lie-Nielsen's new Roubo bench, posted on Woodworking Magazine's weblog. The area of the picture was a little blurry, but it planted the seed, and I went with it. So credit to Lie-Nielsen Toolworks for this idea.

So I lengthened the pins, then inserted them into a turned handle. The pins are now much easier and quicker to grab and reposition.



The pin for the sliding vise is a bit longer since the rear chop on the slider is wider.



After using the handled pin a few times I realized that some thought and effort was required to avoid over-inserting the pin in the parallel guide's hole. Pushing the pin in too far places the handle between the jaws and prevents the pin from seating flatly against the leg. If part of the handle goes past the edge of the leg, the parallel guide would mash that part of the handle into the leg when clamping.



The solution was to stop the pin at a fixed location without interfering with the function of the pin. Anything larger than the pin's diameter would't work. My solution was to drill through each pin and drive in a 1/8" roll pin.



This makes changing holes truly brainless. It's impossible to over-insert the pin.

Changing pins in the parallel guide is quicker and easier than ever.

Tradition meets technology

A few years ago John Herin invented a new type of geared tuner for stringed instruments that he calls the "Peghed". It's unlike any other geared tuning peg. It actually looks and works just like a real wood peg, without all the real wood peg problems. The device uses a system of planetary gears housed inside the peg's aluminum shaft. With the 4:1 gear reduction that these provide, fine tuning is effortless. And the pegs stay put without all the idiosyncratic movement required with wood pegs. When I first saw these pegs in a local music shop I was really impressed by the function of the pegs, but not so much by the looks. The shaft looked fine, but the head was a rather crudely fabricated black plastic material (pictured above) that didn't look or feel like ebony, let alone any lesser wood. It reminded me more of phony leather than wood. So I wrote them off, thinking that I would wait until another option came along. I should clarify that the pegs I saw in the local shop are actually Perfection Pegs, which use Herin's mechanism, but marketed under a different name. I recently became aware of a major change in Herin's product through my friend and fellow oud enthusiast Mike Malek. During a recent visit with Mike, he showed me an entire box of these new tuners headed to a luthier in Egypt. I was really impressed. Here's the difference. The heads of these pegs are actual wood. Ebony, in this case. When a recent customer asked if I could install these new pegs on his instrument, I jumped at the chance. I sent a set of my rosewood Nahat-style pegs to Herin, and he worked his magic.



Herin separates the head of the peg from the shaft, and uses a CNC lathe and mill to excavate a complex mortise in the end of each peg head.



Then part of the mechanism, which matches the shape of the mortise, is glued into the head.





And here's the interesting part. The shaft of the peg which is visible outside the pegbox walls, is fabricated from anodized aluminum to mimic the color of the rosewood heads. The Pegheds are also quite light for a geared tuner. 12 of the rosewood Pegheds weigh 108 grams in their raw state (the extra long black section of the shafts will be trimmed shorter when installed), while 12 rosewood pegs weigh 91 grams. That's about 1.5 grams difference, per peg. In other words, 12 Pegheds are about equivalent to 14 rosewood pegs. Not a huge issue. Up close, the pegs won't fool the discerning eye. But these Pegheds are a far cry from the plastic-banana pegs that I saw in the music shop.

Inlay and Rosette Tricks

I'm always looking for a quicker and more accurate way to inlay without visible gaps. Inlaying a straight-edged design does help, but as always, the room for error is basically a couple thousandths. In other words, not much. The inlay either fills the mortise right up to the edge or it doesn't. Lots of fancy inlay is done on darker fingerboards like rosewood or ebony. Filling gaps in these woods is easy. Just overfill with epoxy and let the dark wood hide the errors. Inlay in lighter woods like walnut poses a challenge, and inlaying soundboards of soft spruce calls for a light touch and extreme accuracy. My technique isn't unique. However, I do have a little trick I use that improves my results. I always tack glue my inlay to the substrate with a tiny drop of CA glue. It won't budge while I scribe with a #11 Xacto blade. And it pops off quite easily afterwards.



I route the majority of the mortise with a 3/32" spiral bit, then sneak into the corners with a 1/32" bit. These little buggers are really fragile, but they cut very smoothly and allow me to get into tight corners. I clean up the very tip of the corner with the #11 Xacto.



And here's where my little trick comes in. I use a highly polished jeweler's burnisher to compress and spread the aris of the mortise wall. 99% of the time this takes care of any iffy spots in the fit. If a particular area is off by more, I'll use some extra pressure and try to smoosh the wood fibers a tad more.



The burnished areas are visible around the perimeter of the mortise. These will get scraped and sanded flat after the inlay is glued.



A matching caul puts pressure evenly on the inlay.



This is ivory in walnut.



For rosettes with inlaid central calligraphy, getting the pattern on the blank in the correct position can be tricky. There is no room for error when the border of the inlay is a 2mm circle surrounded by 2mm walnut. Any deviation from the pattern will be obvious.



I cut some tiny windows into the pattern to line up exactly where the pattern should lay. Once I position it exactly I tape one edge to the blank, flip it over, spray with 3m Super 77, flip it back and press it down. The tape acts like a hinge, returning the pattern to exact spot.



This dark walnut rosette is made from several plies of alternating grain veneer, epoxied together and cured in a press. This plywood makes for easy, predictable cutting since the "grain" of the wood is the same in every direction. And obviously it makes the rosette very strong.





After cutting the pattern, but before removing the paper.



The finished rosette. Walnut and ivory.

Pneumatic Circle Clamp

Today some friend were visiting the shop and I had just finished clamping veneer lines onto the inner edge of the soundhole with a balloon. This technique is so simple, but it's also totally ideal for this situation. The pressure is finely adjustable, the "clamp" is extremely lightweight, the rubber is soft, it won't mar the veneer or distort the shape, the pressure is perfectly distributed around the circle, the clamp is almost free and disposable. I get a kick out of this technique every time I use it. But today I didn't get the last laugh. While my friends were in the shop I stepped out for a few minutes, but it was long enough for them to get into some mischief while I was gone. And that's what worried me. I never keep soundboards clamped in the open air like this when someone else is in the shop (or even when I'm in the shop, for that matter), so naturally I was thinking about the safety of the soundboard for those few moments, especially with half a dozen people milling around. So as I step back into the shop, I hear "what's happening with this balloon setup" followed by a loud pop and shreds of yellow rubber flying through the air. My heart skipped a beat, and an entire day of carefully fitting the rosette flashed in my mind. Once the raucous laughter settled down, I realized I'd been fooled, and then I started cracking up too. I'm still trying to figure out how they found another yellow balloon so fast.

Assymetrical Oud Bowls

Finishing the construction of an oud bowl is a milestone. For me, the bowl is the most technically demanding aspect of oud construction. After my second instrument I decided that I wanted to duplicate the bowl shape of older ouds, specifically the shape most commonly used by the Nahat dyanasty of luthiers. My first two bowls were symmetrically shaped. In other words, I took half of the face profile and rotated it 180 degrees around the centerline of the face. Theoretically, this meant that each rib was exactly the same size and shape.



Like slicing a perfect orange in half, then separating the sections. All the sections would be identical.



My second oud had a symmetrical shape. The bowl profile at the apex was taken directly from the face profile.


The shape of Nahat bowls have two distinct differences over symmetrical bowls. The first is the shape of the tail end. It doesn't terminate (meet the face) at the apex of the curve, but rather continues back a bit past the apex to meet the face at a greater than 90 degree angle. This makes for a more aesthetically pleasing look, and more importantly it softens the bottom corner of the face.



This makes the oud more comfortable to hold, as the players arm curves around the bottom of the bowl.

The other difference occurs towards the neck end of the bowl, at the apex of the bowl. The bowl is made deeper in this area. This also gives the bowl a nicely curved shape, but the real effect here is the increase of the air volume of the bowl, and combined with the position of the deeper section itself (which is directly behind the main soundhole) the projection (volume) of the oud is increased. The greater air volume probably affects the openess and fullness of the sound as well.

There have been a lot comparisons between the shape of the oud and the shape of the human body. The closest natural shape I've found that mimics the Nahat shape is actually a kalamata olive. Can you spot the oud in the following picture?



Departing from a symmetrical shape presents all kinds of challenges. Each rib is now a different shape, and with the curved tail end, the ribs closer to the face must twist at each end. No longer is the rib simply bent in one plane, but it is now a bent and twisted rib. In order for the ribs to join at their edges seamlessly, one must accurately accomplish three tasks. 1. Bend and twist the rib to match the profile of the bowl (while matching as closely as possible the curve of the previous rib) and land flatly on the the tail end and neck end gluing blocks. 2. Joint the curved and twisted rib to the previous rib (I shoot for a light-tight joint) without altering the way the rib lands on the mould. 3. Joint the opposite edge of the rib in preparation for the next rib. For me, this involves a stepped approach of gradually tweaking the bend and twist of a square-sectioned rib until I'm close, sawing the taper in one side, jointing the edge on an inverted jointer plane, further tweaking the bend and twist, final fitting of the edge, and finally, sawing and jointing the opposite edge.



Sometimes during this process, the taper of the rib itself will be off, causing the widest part of the rib to depart from the mould, instead of hugging the shape. More material must be removed from the ends of the rib in order to pivot the rib into correct position. Of course, this can drastically change the twist of the rib, which must then be corrected on the hot bending iron.



If the rib continues to need twisting, I'll heat the rib in the area I want to twist, place the rib on the mould, then gently but firmly twist the rib by hand while it's still pliable. Alternating jointing the edge and bending/twisting is necessary to avoid overcompensation of either tasks. I prefer to fit all ribs with oversized (by at least 1cm) square sectioned blanks in order to leave the most room for fitting. On good days, I can fit a rib in a little over an hour. On bad days, I just chuck the problematic rib across the shop and call it a day.





When things go well, I end up with a bowl of the intended shape and size, and a feeling of relief after accomplishing the most difficult part of oud construction. My latest oud bowl is taken directly from the shape of the old Nahats.

The Benchcrafted Tail Vise-Now Available

I'm excited to announce that the Benchcrafted Tail Vise is now available!

Introductory pricing is only $315 for the complete vise. All you have to provide is the wooden dog block to fit your bench.

Details can be found at the Benchcrafted Website. Click on the Ordering tab to place an order.

Vises are built on a made-to-order basis, and turnaround time should only be 2-3 weeks. Eventually the vises will be a stock item, and shipment time will only be a day or two.

As always, email with any questions.

The Benchcrafted Tail Vise

I've received numerous inquiries about this tail vise (also called a wagon vise), and whether or not it will be available for purchase. I'm glad to say that this vise will be available for sale soon.

This vise is extremely robust, precise, and operates smoothly and effortlessly thanks to a precision acme screw and nut. The precision 1-1/4" diameter left-hand screw features a pitch of 4 tpi, which strikes a balance between fairly rapid movement and controlled, gradual clamping pressure.The vise moves the same distance per turn as typical tail vise hardware. The 5" dia. handwheel is chromed cast iron, polished to a high luster. The rim is comfortable to grip when approaching final position or realeasing the vise. The cold-rolled steel outer flange is blued for a subtle look and rust protection. A coat of paste wax will further protect the parts. All other parts are cold-rolled steel. Moving parts and surfaces are machined for smooth and precise action.

The vise is designed around a Roubo-style bench (it can be retro-fit to some benches) with a top thickness of 4" or greater, although it could be used in thinner benches with wider aprons. All hardware is included. The wooden dog block will be customer-made to conform to individual dog hole row and bench thickness. The vise is set up for a maximum dog block width of 1-3/4", although wider blocks could be used. Vise capacity is variable depending on installation. On my bench there is about 7-1/2" max. between the jaws, but this can be increased by lengthening the slot and/or decreasing the width of the end cap. Maximum travel with a 2-3/4" wide end cap is 10-1/4". Instructions and templates will be included. This vise will be produced on a made-to-order basis, although turn around time shouldn't be longer than a week or two. Introductory price for the vise will be $325, not including shipping. The vise will be available with two options. A larger turned handle (in various woods) to replace the handwheel's steel handle will be available for an added cost of $25. Substituting the handwheel for a traditional iron T-handle (seen below) will be an extra $15. Please feel free to drop me an email if you are interested in ordering a vise, or have any questions.









Handwheel knob in cocobolo.



Iron T-handle (wood not included)