First Cello Build

[Karl Wicklund]

Thanks for posting this! How does the tool work to draw the topo lines?

[Bob Francis]

These posts are great lessons Tom!
Thanks

[Tom Griffin]

Thanks for posting this! How does the tool work to draw the topo lines?

There is a fixed stylus in lower position of the fixture and a pencil that can be set at different distances from it. By dragging the fixture around the edge of the cello with the pencil set at various heights, a series of topographic lines are created. They make any variations in the profile very apparent.

Tom

[Tom Griffin]

These posts are great lessons Tom!
Thanks

Thanks Bob,

I just started cutting the grooves for the purfling, so I will be posting some content on that procedure shortly.

Tom

[Bryan Bear]

I agree, I like this thread a lot!

[Eric Knapp]

Wow, when machinists retire...

Great work.

-Eric

[Tom Griffin]

I lost my 28 year old son August 12th., so life has pretty much been on hold. I'm just starting to get my head screwed back on straight enough that I trust myself to get back on the cello project. We'll see. All you parents out there give your kids a hug, because life is fragile and comes with no guarantees.

The first, somewhat mindless task was to make up the linings, glue them in and whittle them to shape. They taper from full thickness at the glue surface to very thin on the inside edge to save weight. Obviously the entire body has to vibrate to produce sound, so weight is a factor, although the edges of the plates probably move the least. I also got the corner and end blocks shaped. Next I may create the tapered hole in the lower block for the endpin (the telescoping metal pin that supports the cello when it is played). It seems like it would be easier to do it now, before the front and back are attached. I also have to scrape the inside to remove the excess glue that I neglected to wipe off when it was first applied and easy to remove. The outside needs scraping as well, especially in the area of the corners to make them perfect. I think that would be easier now as well.

Tom

[Tom Griffin]

The reaming of the tapered hole for the end pin went without a hitch. The tapered reamer was one of the necessary special tools that I'll never use again, unless I build another cello. The purfling on the front proved challenging, but I think a lot of it had to do with my state of mind. Hopefully it will go more smoothly on the back. It was a bit of a challenge to get it all in before the hide glue set up. I heated the edges of the front with a heat gun to buy as much time as possible. Once it was all glued in, I planed and scraped it flush with the surface.

Tom

[Tom Griffin]

The scariest part of the project so far was thicknessing the Spruce front. Most of it all needs to be about 4mm thick, a little less where the bass bar mounts. It was first roughed out with a Forstner bit, set 4.5mm above a plastic pad. I think I checked that dimension at least a doezen times before getting up the courage to cut into the $200 piece of Sitka Spruce. The process removed a lot of material and provided a depth guide for the final gouging. It was also a hell of a lot easier than gouging it out by hand. The final thickness will be arrived at using a gouge, plane and scraper. The front is becoming light and flexible even at this stage and emits a pleasing tone when thumped. Final graduation will be done by exciting the front with various frequencies and observing patterns produced on it with a fine powder (ground tea leaves seems to be a favorite). I'm still trying to wrap my head around all that.

Tom

[Barry Daniels]

Tom, I am so sorry to hear about the loss of your son.

The cello work is very impressive.

[Tom Griffin]

So I jumped in with both feet and started the tuning process on the belly, or front plate. When played, a cello acts as an air pump to produce sound, and the front and back plates must be tuned to the same frequency so they vibrate at the same rate and don't fight each other.

The tuning process consists of sweeping the plate from roughly 20 to 180Hz and observing the points at which it resonates. For a cello, there are five points, or modes in this range, with two of interest, mode 2 and mode 5. Generally mode 2 should occur around 60-65Hz and mode 5, an octave higher or around 120-130Hz. These modes can be observed by sprinkling particles such as spices, sawdust or glitter on the plates and watching various patterns form (I used Basil ). The particles tend to gather where the plate is vibrating the least. When I first checked the plate, it was quite thick, around 5mm, and the two modes were at 160 and 75Hz., which made sense because the thicker wood should have vibrated at a higher frequency. As I reduced the thickness of the plate, the modes fell to 135Hz for mode 5 and 62Hz for mode 2. The plate thickness now ranges from 3.5-4mm. I would like to bring the mode 5 frequency down a bit more, but want the mode 2 frequency to stay the same, and I'm not really sure how to do that. I think that thinning the outside edges will help, but I'm not totally sure. Guess it will be lots of small changes and checks to keep an eye on what is going on. Once the plate is tuned, I'll cut the sound holes and mount the bass bar, which will raise the resonant frequency again. Then it will need to be re-tuned by refining the shape of the bass bar until the resonant frequencies of the plate fall back within range.

I understand that the plate can also be tapped and the resulting impulse recorded and then analyzed in Audacity to determine the modes, but this technique seemed way more fun. Also, from what I understand, the early makers used the same theory, but used a bow on the edge of the plate to excite it in lieu of a yet to be invented stereo.

Equipment used was a cheap audio oscillator to generate the sine waves, a stereo amplifier (one channel), a hi-fi speaker (JBL L-112), some foam blocks to hold the plate in close proximity to the speaker, and a tuner app on my cell phone to read the frequencies. Fun with physics!



Mode 2 (two opposing C shapes)



Mode 5 (a closed pattern roughly following the edges of the plate) The lack of particles near the middle of the plate is due to the steepness of the slope.

[Barry Daniels]

That is interesting. With guitars many builders try to tune the back around a third interval higher than the top. This way the back will be activated by the top's vibrations. If the back is tuned at the same note it tends to act as a damper to the top and will rob power from it.

[Tom Griffin]

I've seen it mentioned both ways for the cello (depending on the source) and will definitely look into it further before tuning the back.

Tom

[Tom Griffin]

I've seen it mentioned both ways for the cello (depending on the source) and will definitely look into it further before tuning the back.

Tom

So I've read three different articles on plate tuning for the violin and cello, one by Allan Carruth and two by Carleen Hutchins (one online and one from an excellent book entitled The Art of Violin Making), and all three recommend that the modes for both front and back match as closely as possible. Guess that's the way I'm going to go. If it sucks, thanks to hide glue, I can always open it up and work on the plates some more.

One thing I did see mentioned is that if mode 5 and 6 are too close, it can cause wolf tones. Not sure I have the skillset to control them all that much so if wolf tones are an issue, I'll have to deal with them with dampers between the bridge and tailpiece (wolf tone eliminators).

Tom

[Tom Griffin]

I got the belly tuned to 62Hz. for Mode 2 and 130 Hz. for mode 5. That's pretty close to an octave apart and in the recommended range so I called it good. The addition of the sound holes and bass bar will change the tuning, so it will have to be revisited anyway.

I ordered spiral blades for the jewelers saw, thinking they would make cutting the sound holes pretty easy, but they didn't work well in the Spruce. All they did was follow the grain, so I couldn't cut very close to the line, which made for a lot of carving. They also left too wide a kerf for the narrower gap in the sound hole. I left the sound holes a bit on the small side so they could be opened if needed during the final tuning and voicing.

I also steamed out a few dents in the belly left from the button during the thicknessing process. I noticed them pretty quickly so there weren't too many to remove. A new button with a larger radius solved the problem.

Gonna do the bass bar next, but am heading North for the week to get away from all the craziness around here. We actually had a truck hijacked at a local market for it's paper goods contents. For whatever reason, toilet paper is in huge demand.

Tom

[Randy Roberts]

Tom,
Wonderful thread. Thanks for sharing this with us.

Have never done a bowed instrument, but would a jeweler's saw using regular blades like you would use for cutting pearl work better for the sound holes? I've used them on wood for a lot of different things and they work well. I would think a 3-0 blade would work well for your sound holes.

A trick with your top and back tuning that works well with guitars is to use poster putty or what I find even better is a couple of rare earth magnets on either side of the top.
By moving them to different places on your top and then running your resonances, they will give you a preview of what removing wood at that spot will do to your Chaladni patterns. The increased mass of the magnets will have the same effect as the decreased stiffness of shaving off some of the wood.
It lets you move them around easily ( the magnet on the back will slide with the magnet on top when you move the top one) to see where shaving wood will end up giving you the effect on your patterns and frequencies that you want before you commit to actually shaving more off.

[Tom Griffin]

Tom,
Wonderful thread. Thanks for sharing this with us.

Have never done a bowed instrument, but would a jeweler's saw using regular blades like you would use for cutting pearl work better for the sound holes? I've used them on wood for a lot of different things and they work well. I would think a 3-0 blade would work well for your sound holes.

A trick with your top and back tuning that works well with guitars is to use poster putty or what I find even better is a couple of rare earth magnets on either side of the top.
By moving them to different places on your top and then running your resonances, they will give you a preview of what removing wood at that spot will do to your Chaladni patterns. The increased mass of the magnets will have the same effect as the decreased stiffness of shaving off some of the wood.
It lets you move them around easily ( the magnet on the back will slide with the magnet on top when you move the top one) to see where shaving wood will end up giving you the effect on your patterns and frequencies that you want before you commit to actually shaving more off.

Thanks Randy.

I did use a standard blade to saw the narrow gap end on the soundhole and it was much more controlable. I suspect the spiral blades would work well in a wood with finer grain.

The magnets are an interesting idea, I just happen to have a few laying around. I'll try them out when I tune the back.

Tom

[Randy Roberts]

Tom,

A couple other things:
It may not be a factor with your project, but with the guitars, it's best to support the plate at the node lines, and drive the top with the speaker over an area of maximum movement. It looks from the picture that the plate is laying on top of the frame of the speaker which, if it is resting on areas that want to be moving, might alter your patterns and frequencies. Maybe it doesn't affect a plate as big as yours, or the power of your speaker may over power any damping from the edges of the speaker case, but I've found where the plate is supported can have an effect on a guitar plate.

I made 3 support posts from dowels stuck in blocks of wood with a pad of foam rubber on top of the dowels to rest the plate on, and move these supports to where node lines form. My speaker is just a bare speaker held over the center of the main area of movement for a given mode. (I use glitter from a hobby store, but it does make me look like I went to the Prom dance after a session of tuning.)

[Tom Griffin]

Thanks Randy.

You can't see them in the pics, but the plate is resting on small blocks of foam, located on the node lines. I did notice a difference (mostly in intensity) in where the plate is supported for the various modes. The speaker should also be positioned under areas of movement for the same reason.

Tom

[Tom Griffin]

I reached a milestone in the project and got the body closed up. It was a shame to cover up the work that went on inside, but it's good to see the hard part done. The thing emits a nice clear tone when tapped on the back, so it seems to be working as it should. I also got started on the neck by sawing out the rough outline. Now it's time to get to work and start carving the scroll. I think I can see light at then end of the tunnel.

Tom