Sunday, April 23, 2017

Day 147: Soundboard Fun and a Day at the Beech

Now that the soundboard is completely jointed, I can start preparing it for the bridges and bracing. The first step in this lengthy process is cutting it to size so I can eventually glue it to the liners that run along the cheek, bentside, tail, and spine. First, I laid the soundboard on the case with the spine side lined up along the inside of the case. I did this because, as you may recall, I only had 1/8" to spare on the width of the soundboard once it was glued up.

I then drew a line on the bottom of the soundboard using the outside of the case as a guide. Eventually, I will need to trim the cheek, bentside, and tail sides down by 1/2" to make it match the inside of the case, but, for now, it gets an outside line.

Next, I'll use a jig saw to cut along the line and work on the 1/2" line after that because I'd like to clear some material away before making that final cut.

Along with the soundboard work, I made the management decision to go ahead and replace the upper and lower registers with beech versions. The poplar registers just weren't working for me. They were pretty ragged and flimsy and, after seeing some of Bill Jurgenson's beautiful beech register work on Facebook, I thought, "Hey, I can do that!" So, I started by cutting a 6" x 32" inch piece of beech from a 3/4" x 8" x 96" board and resawing it down to 1/2". I then ran it through the planer to smooth it and get it down to a true 3/8" piece.

What I neglected to mention above was how my decision to make my own jacks also affected the decision to make new registers. Rather than use the plastic Hubbard jacks recommended by Mr. Miller in his eBook Most Excellent, I decided to make them 1/2" wide, yet the same thickness as the previous ones because I'm still using Owen Daly's custom table saw blade for the jack thickness cuts. In the end, I also needed to attend to the width of the registers themselves because there's not much wiggle room between the two based on parts I've already installed.

I also decided to cut a slant into the upper registers to accommodate jack movement when turning the registers to the "off" position. This essentially moves the jacks for that register to a slanted position, thereby pulling the plectra (the things mounted in the jacks that pluck the strings) away from the strings; they still pop up and down, they just don't produce string vibration and, hence, sound. Rather than build a custom jig or chisel them out by hand later, I decided to use the Incra miter gauge I picked up a while back.

First, I lined out the cuts using a marking tool and an "old" register that I sawed apart to re-open the comb. I made a 10-degree cut for each register slot. I then lined each slot up with the saw blade (yes, the saw was shut off each time) and clamped the register to the miter gauge, turned the saw back on, and made the cut. Contrary to most of my great ideas, this worked remarkably well.

Before proceeding too far, I checked the cuts and was fairly pleased with what I observed.

I then finished up the angled and straight cuts and found that they still needed to be cleaned up a little with a razor knife blade. I only sliced about 1/2 mm into my left thumb, so no blood, but plenty of excitement.

Once I completed the cleanup, I put the piece I will use to "close the comb" close to the register to gauge what it would look like and I was quite pleased that it not only provides plenty of stability, it is also perfectly matches the width and height of the "old" registers.

Not too shabby at all. One concern I had was with a miniscule amount of tearout because I had neglected to use a backing board when cutting the slots. Such a board is simply used behind whatever one is cutting on a table saw to prevent "tearout" - the detritis left on the back side of a cut. The absence of a backing board was also mentioned to me on Facebook by Michael Johnson, a master harpsichord builder from London. So, I taped a backing board onto register number two and went to work.

Now, I had virtually no discernible tearout on register number one. And, wouldn't you know it, guess what I observed on register number two. Yep, a little tearout between two of the slots.

This is pretty wild stuff. The backing board's sole purpose was to prevent this and, well, there you go. Overall, the cuts were cleaner, it was only this one that felt the need to remind me that I'm still just an amateur. And, speaking of being an amateur, as I was cutting the piece to close the combs of the registers, I experience my first bona fide kickback. And it hurt. I was using a push stick that kicked out of the blade and into the palm of my hand.

It doesn't look like much, but it hurt like a mutha, yet just for a little while. I decided at that point it was probably time to call it a night.

On a completely unrelated note, I received a book a couple of days ago that is profoundly important to me. I've connected with Bartolomeo Cristofori in a way I cannot yet fully articulate, or even understand, and will, someday, build a copy of his 1693 spinetta ovale instrument. Owen has recommended I not do that and concentrate on a "little Italian" for my next build, so I'll be following Owen's recommendation, yet Bartolo and his fantastic oval spinets (virginals, really) will always remain in the back of my mind.

The book actually covers the 1690 instrument and its copy made by Tony Chinnery and Kerstin Schwarz. In the unlikely event I do end up building a 1693 copy, this book will go a long way toward helping me make that a reality. In the meantime, I have plenty to do finishing this instrument and looking forward to Owen's guidance on the little Italian.

Until next time...

Monday, April 10, 2017

Day 146: Finishing the Soundboard Joinery

While I've decided to make new upper and lower registers (aka jack guides), I didn't have room to run the beech board through the table saw because the soundboard glue-up was in the way. That's okay because I'm slowly learning that concentrating on just one thing practically ensures success. Okay, well, it prevents screwups. Mostly. So, it was more soundboard joinery this past weekend. Rather than comment on each photo, I'll just run them below with a few comments here and there.

Progress so far below.


And done:

Never fear: the darker wood will get lighter when I run it through the thickness sander.

One thing I did that was most helpful was clean up all of the joints before gluing them by creating an impromptu (notice I didn't say "makeshift") shooting board. This allowed me to turn my beloved Lie-Nielsen low-angle 62 hand plane on its side and run it along the edge, creating a perfectly smooth interface for each of the boards.

It's a really great way to clean up the edges and leaves a nice curl if your plane blade is razor sharp.

Orthodox Plane
Mixed media, 2017

I'm quite concerned about the thickness of the soundboard at this point and also about its width. It's generally better to joint thicker boards, but mine were planed down nearly to their final thickness of 1/8". This made the glue-up nerve-wracking, to say the least, and I'm hoping additional cleanup of the front and back surfaces doesn't result in a soundboard that's too thin to be useful. If this happens, it's back to the drawing board.

I was also left with a scant 1/8" overhang on the width - that's just 1/16" on each side at the keyboard end. This is not the best situtation, either. I would have preferred sufficient width that overhangs the outside of the case. That would have allowed me to cut the soundboard down along the outside edges and then trim it back to fit the inside walls of the case. Now, I'll need to be quite creative with how I approach the final fitting. It's not like I haven't been in similar situations with this instrument before, right?

EDIT: On further reflection, I realized I could simply line the spine side of the soundboard up with the inside of the case, which will give me more overhang on the cheek side. Crisis averted. I think.

Once the soundboard was glued up, I did finally turn my attention back to making new upper and lower registers out of steamed Eurpoean beech. I cut a 6" x 32" piece from an 8' board and did the impromptu shooting board thing again, this time on the Roubo.

Because I decided to not go with the Hubbard jacks specified in Mr. Miller's eBook Most Excellent and make my own, the dimensions of the jacks are different. This means I need to cut new upper and lower registers to accommodate the new jack dimensions, which is not such a bad thing because I'm now using the beech, rather than poplar, which I found to be a bit flimsy for the uppers. I'm looking forward to working with the beech - it cuts and cleans up like a dream and makes me feel European, so there you go.

Until next time...

Tuesday, April 4, 2017

Day 145: Here Fishy, Fishy

It's been quite a week at Tortuga Early Instruments Worldwide Headquarters. After looking at the hot hide glue seam in the soundboard for the better part of a week, I decided it simply could not stand. So, I chopped it off and reglued it using the trusty, old fish glue. All was going swimmingly well (pun intended) until I pulled the clamps off to discover the distal end (away from the keys) somehow came out of alignment so that one of the boards was higher than the other.

Rather than panic, I knew the glue was not completely set (heck, the open time for the fish glue is an hour), so I went ahead and, like my dear mother when I was a sick little monkey, provided a moist, hot towel and applied some pressure.

I left the it sit for about an hour and same back to find that it had realigned itself, so I went ahead and shot a little more glue in and reclamped it for the final push.

The next day, things looked pretty great.

I'm pleased with the joint and even more pleased with the ease of use and performance of the fish glue. Now, my beloved mentor, Owen Daly, tells me a story about receiving a clavichord for repairs some time back that was constructed with fish glue that had fallen completely apart. This is, of course, not good. On the other hand, Mark Roberts tells me about a fellow luthier who built a multitude of classical guitars over a 25-year career with instruments now living on every habitable continent on the planet in every conceivable environment and he has yet to receive one report of failure.

I rather enjoy the fish glue. Its open time is phenomenal, it cleans up nicely, and once set, it's hard as a rock, unless sufficient heat is applied. Owen explained to me that simply rubbing a fish glue joint between the thumb and forefinger will loosen the joint, but I've not found this to be the case. In fact, I've found that moisture will not cause a failure - I soaked a test joint in water overnight with no failure - and it takes quite a bit of heat from a heatgun to melt the glue to the point where I can pull the pieces apart. In a strength test, I applied forty pounds to a joint and the wood broke before the glue joint failed.

As you can imagine, I defer to Owen with regard to just about everything I do related to harpsichord building, yet I can't help but wonder if, like hide glue, there are different strengths of fish glue. I have no idea why the open time of the fish glue is so extensive and I plan to call the supplier to inquire about any chemical preparation they might be doing to the stuff, as well as ask them about strength over time, etc. I suspect there is more to the fish glue I'm using than meets the eye and I intend to get to the bottom of it very soon.

Until next time...