41 replies to this topic

[Addie]

What profile of bass bar do you use? I am seeing a long ridge, a long hill shape, and a hunchback that dips down rapidly.

I had made one of the latter type, but got a tubby sounding G. The top grain is fairly wide at the edges, close in the center. “Germanic” arching. Pretty average graduations.

[L.D.L.]

I asked this question on an other forum,and got this anser from Michael Darnton.

"I use a simple plan that is based on something I got from the Carl Becker shop, who get full credit for it. The basic idea is to measure the bar height including the top thickness, with a graduating caliper. Divide the bar into eighths (nine pencil marks, including the ends) and the thickness of top plus bar is 5.0mm, 7.6, 10.2, 12.8, 14, 12.8, 10.2, 7.6, 5.0, from one end to the other at each mark.

It's simple, gives exactly the right profile, and compensates for top thickness. I start the bar 40mm from both ends, and it's 5.5mm wide, 1mm inside the outside foot of the bridge, and 12mm from the center of the upper bout and 15mm from the center of the lower bout.

That's all there is to it, and for smaller instruments multiply by .925 for each successive size downward."

I have been using this way of doing it for all my violins ever since. Worked good for me.

Hope that helps , Larry Lewis

[Brad Dorsey]

...I have been using this way of doing it for all my violins ever since...

Me, too.

[Bob Spear]

There are a few general plans that are followed for bass bar shape, plus a lot of individual adaptations. One idea is that the shape of the bass bar should mirror the shape of the outside arch. I think this is similar to what Michael is describing. This and similar plans make the length of the bar critical. If the bar is too long, it will tend to be too heavy.

Another shape I see quite a bit is similar to this, but the bar tapers down more quickly and then rises at each end. This bar has an appealing shape, but the extra mass at the ends of the bar tend to be detrimental acoustically.

Bars not only taper from end to end, but the spring can be controlled by making the top of the bar thinner than the bottom part.

I use a shape that puts the highest part of the bar directly beneath the bridge foot because the bar also serves a structural function. I try to keep the bar as high as I can until it is past the upper eyes of the f-hole because it is at this point that the top is weakest. Especially critical in larger instruments like cellos and basses. The bar can taper down pretty fast after that. If you are using free-plate tuning methods and have access to a speaker and a sine wave generator, looking at the patterns of the modes while working down the bar at its 2/5ths and 4/5ths points can be very instructive in showing the acoustic effects of the bar.

[robertdo]

the C&J gives some indications about what the shape of the bar should be, mirroring more or less the outside long arch shape (for example if the arch rises quickly and then has a long central flat area, so will be the bar). But in the book the bar ends up not being symmetric since the highest point is located at midpoint between the middle of the bar and the bridge location.
So it looks more like having a feeling of how the bar would best match the shape of the top than an exact mathematical measurement.

Edit. Bob had posted his answer while I was typing.

[colledge]

I asked this question on an other forum,and got this anser from Michael Darnton.

"I use a simple plan that is based on something I got from the Carl Becker shop, who get full credit for it. The basic idea is to measure the bar height including the top thickness, with a graduating caliper. Divide the bar into eighths (nine pencil marks, including the ends) and the thickness of top plus bar is 5.0mm, 7.6, 10.2, 12.8, 14, 12.8, 10.2, 7.6, 5.0, from one end to the other at each mark.

I use this method as well, and find that dividing up the bar into visual areas also helps with fitting. There are respective measurements for viola and cello as well. The think the 10.2 - 7.6 area of the bar is important because, without measuring, one might leave this area thicker. You can also measure things out .75 mm higher and make a template to get you in the ballpark very quickly

[baroquecello]

I see that there are many preferred differences. I also see that nobody, apart from the person asking the question, has mentioned a concrete connection between the (kind of) sound of an instrument and the shape of the bass bar. (apart from a noted negative acoustic effect when the ends are increased in height). This makes me wonder about the practise of making shorter bass bars in baroque instruments. Does anyone know how well documented this practise is? How big are the differences in terms of sound and/or response? And how important is, structurally, the length of the bass bar compared to the height? I am asking as I have seen a couple of (relatively new) baroque cellos with a short and "undeep" bass bars with sunken tops (clearly visible at the bass f hole!), and wonder with what parameter this has to do. I presume Bob Spears comment on the f hole might have something to do with it?

[Don Noon]

I have experimented with various bass bar heights, and found that it had far less impact to tone and stiffness than I would have expected. So I more-or-less just go with something similar to the Darnton profile. I think one mm or so taller or shorter and you'd be hard-pressed to tell the difference by sound, although I don't know how the shorter one would hold up over time.

[David Burgess]

I try to trim bars based on what the strength of the bar/top combination is, rather than by making assumptions based on dimensions. It's the same reason why not every plate can be graduated the same.

[colledge]

I try to trim bars based on what the strength of the bar/top combination is, rather than by making assumptions based on dimensions. It's the same reason why not every plate can be graduated the same.

I learned that the numbers cited are on the conservative side, leaving room to check this, but this is another good reason for a profile template that brings you close, but heavy.

David, in an effort to not make assumptions, are you saying that you just weigh and flex your way to a final bass bar shape?
Is there any restoration shop in the world that does not give their people a number to go to?

[carl stross]

I try to trim bars based on what the strength of the bar/top combination is, rather than by making assumptions based on dimensions. It's the same reason why not every plate can be graduated the same.

Is there any evidence that stiffness is the main factor in a bass bar ?

[DarylG]

I'm with David and adjust the thicknesses depending on the top. I like to keep the triangle shape that the Becker system uses so if I need a taller bar I alter the other thicknesses accordingly. Right now I flex and tap, but I've had an idea floating around in my head for a jig that could also be used to measure the deflection at the bridge foot when a load is applied. I generally make my bars 5.5mm thick but sometimes I wonder if that's just because that's what I was taught. Gram for gram, is it better to add height or thickness to increase stiffness? What about tapering the bar before fitting so that once fit it's thickest in the center and progessively thins towards the ends? Cheers,

[David Burgess]

Is there any evidence that stiffness is the main factor in a bass bar ?

I'm sure mass is a factor too, but I have less experience with that. I've tried increasing mass without changing stiffness (easy to do experimentally), but have never tried starting with a bar with extreme low mass/high stiffness, which would probably be something like a ribbon of carbon fiber separated from the top with balsa or foam, and worked my way up in mass from there.

I think both mass and stiffness are important throughout the violin, but probably most important on the top.

One thing I can say is that reducing the height of a bar just .5mm will produce a measurable change in stiffness, and to get the same stiffness out of two bars installed on the same top, one may need to be 4mm higher than the other. I don't think it's anything that can be ignored, any more than it would be wise to ignore stiffness of the top (when making) before the bar is installed.

[Don Noon]

Gram for gram, is it better to add height or thickness to increase stiffness? What about tapering the bar before fitting so that once fit it's thickest in the center and progessively thins towards the ends? Cheers,

Neglecting the top for the moment, the bass bar stiffness varies with the cube of the height, and only directly with the thickness. So if it's stiffness/weight you want, taller and thinner is better. Or, better yet: an I-beam profile, thicker at the top and thinner in the middle. But I'm definitely NOT advocating such a thing, as I don't believe that gross beam stiffness/weight is the only thing going on here.

[Addie]

I’m still digesting all of this info, so forgive me if I missed something.

Highest point of the bar, on the Darnton method, would coincide with the lowest point of the inside arch, which would generally be where the bridge sits? Is that right? Other highest point ideas are the center of the bar, and between the bridge location and the center of the bar.

5.5 mm wide. I have also read 6 for "low arch” which would not apply to “Germanic,” or what I would describe as a narrow tranverse arch.

True or false? Weak bar is responsible for tubby G.

How does this relate to the Baroque bar question?

Thanks!

[DarylG]

as I don't believe that gross beam stiffness/weight is the only thing going on here.

Hi Don,

Care to expand on this? I've been working under the assumption that stiffness and weight were the main factors. Cheers,

[robertdo]

I’m still digesting all of this info, so forgive me if I missed something.

Highest point of the bar, on the Darnton method, would coincide with the lowest point of the inside arch, which would generally be where the bridge sits? Is that right? Other highest point ideas are the center of the bar, and between the bridge location and the center of the bar.

5.5 mm wide. I have also read 6 for "low arch” which would not apply to “Germanic,” or what I would describe as a narrow tranverse arch.

True or false? Weak bar is responsible for tubby G.

How does this relate to the Baroque bar question?

Thanks!

If you divide the bar in equal segments (8) and put the highest point in the middle I don't think it can coincide with the bridge location because the bridge is at about 195, so below the middle of the plate (about 177mm, where the highest point would be if it was located just in middle of the bar).
I guess this is why C&J suggests to put the highest point between the middle of the bar and the bridge location. But there is only 10mm difference, and unless the shape of the was to sharply go down at 90degres, the bar is still at its high point in this area.
I did ask once if the bar was responsible for the strong and deep G strings on the violin I had made, and the general answer was that it was not the main responsible. Arching and graduations coming long ahead as factor..

[David Burgess]

True or false? Weak bar is responsible for tubby G.

I think it's true, but hard to separate from a bar/top combination which is weak. Some tops are weak enough that there's not enough you can do with a bar to compensate.

Well, maybe that's not entirely accurate, since I haven't tried the really weird stuff, like running a bar all the way to the upper and lower blocks, or putting in a violin bar more than 16mm high.

[DarylG]

One thing I can say is that reducing the height of a bar just .5mm will produce a measurable change in stiffness, and to get the same stiffness out of two bars installed on the same top, one may need to be 4mm higher than the other.

Hi David,

Do you make your bars all the same thickness? What about the shape of cross-sectional profile? I make it the way I was taught, which I'm sure is similar to what some others here do, but why? If you think about what Don said about the stiffness being the cube of the height than why not make the bars thinner and square cross-sectionally? It seems to me that shaping the cross-sectional profile only serves to increases the mass side of the mass:stiffness ratio. Cheers,

[David Burgess]

Hi David,

Do you make your bars all the same thickness?

I do, pretty much, within .5mm.

What about the cross-sectional profile? I make it the way I was taught, which I'm sure is similar to what some others here do, but why? If you think about what Don said about the stiffness being the cube of the height than why not make the bars thinner and square cross-sectionally? It seems to me that shaping the profile only serves to increases the mass side of the mass:stiffness ratio. Cheers,

The reason I don't make bars thinner is that the thinner the bar (at the base), the more it seems to imprint on the outside surface. So I agree with what Don said, and also try to merge this with practical considerations involving distortion observations over long periods of time.