Bass bar springing?

[polkat]

Another recent post here about the bass bar got me to thinking about springing the bar. I don't do it, but I was wondering when this technique got started, and what is the intention of it? I assume it is to help the top plate retain it's shape against the tension of the strings, but are there other reasons? Thanks!

[Michael_Molnar]

This is a topic of contention for sure.

I do not put mine under tension. In fact, I know how to make a perfect fit without tension. This is probably the one (only?) thing I do best in my violins.

I remember reading in Weisshaar that a bar under tension could pull the plate out of shape. He also wrote that this theory of bassbar tension probably originated with a person who did not know how to make a good fit without tension. (Ouch!) Read it for yourself and have a good laugh or cry.

Some people argue that a bar under tension will produce a higher ring tone (mode 5). That is true only if the bar's spring constant is non-linear (read up on Hooke's Law) which would be true if you bent the bar to the point of breaking. Anyhow, I find that a properly fitted bar without tension gives me all the ring tone and damping I need.

[Andres Sender]

Hmm I wonder if Weisshaar was just not familiar with historical texts on violin making. Fitting bars with tension is mentioned by Bagatella (1782) and Count Cozio (b. 1755 - d. 1840).

As to which is more difficult, I'd suggest anyone who thinks properly fitting a tensioned bar is easier give it a try.

[MANFIO]

This is a highly controversial issue, with good professionals in both sides. I am not a "springer".

[David Tseng]

I remember reading in Weisshaar that a bar under tension could pull the plate out of shape. He also wrote that this theory of bassbar tension probably originated with a person who did not know how to make a good fit without tension. (Ouch!) Read it for yourself and have a good laugh or cry.

I heard Hans Weisshaar said that back in the mid 80's. His reasoning was that wood would creep and eliminate the tension. Michael Darnton countered that the bar tension remained after he removed the old bar. I used to leave 0.5mm at the end of the bar, but now I fit the bar as close as I could and stop using the aluminum fitting jig.

[Magnus Nedregard]

It is not a about "tension" really, it is about obtaining the right shape. You want to avoid a flattened bass side, and bulging arch over the bar ends.

I:

Imagine that you mount the plate on a frame, and fit a perfect bar to it. No flexing anywhere. This should be what people call "without tension" then.

Now, when you put that plate on the frame, why did you have to clamp it? Wasn't the plate actually under "tension" when you had it in the frame? So have you now fitted a bar with "tension or not?

II:

Imagine that you fit bar to a loose table, still without any flexing or bending. You then glue the plate to the ribs, and you realise the table has to change shape, and actually is put under tension as you clamp it down on the ribs. "Tensioned" or not?

I might go on with examples like this for a while, but I think that will do. My point is that I think the discussion about "tensioned" bars derive from (failed) attempts to understand what is going on in the shaping oft the violin table - in it's various states: free plate without bar, free plate with bar and plate as glued to the ribs, or better; the variety of rib shapes out there!

In most cases, a table fixed to a frame, with a perfectly fitted bar, gives approximately the same result as a bar fitted to a loos table with some "tension". But the frame method gives a more predictable result.

If what you want to do is to raise the bass arch a tiny bit to make room for future sinking the most logical method would still be to put the table on a frame. Then I would let it hang upside down in the frame with a slightly heated sandbag for a while, to shape the bass side a little, and then fit the bar (perfect) to that. (But I see no reason to make a pushed-up bass side.)

[upnorth]

I wonder what the statistics would be on this? There are a lot of variables involved other than fit.

I buy violins from a supplier that has refitted them with a new bar, they sell quickly and usually before ones that were not re-barred.

I have concluded that the bar is critical.

I hear a lot of silence on this matter. Silence from those who sell violins for large sums.

I have concluded that the bar is critical.

[Magnus Nedregard]

Rather a mess, my last post. My point is just that you have to manipulate the shape of the table a little, and the best way to do that is by means of a frame.

Otherwise you are fitting to something that has the wrong shape.

The free-hand bending idea is just a rather unsophisticated way of dealing with that problem, the concept of "tension" is a false problem, and I think it is misleading.

[Mark Sullivan]

As a starting point for someone with no experience, should the frame be shaped, bent or shimmed to represent the shape/tapering of the ribs before fitting the bass bar?

[FiddleDoug]

I think, and was taught, that the main reason for keeping the plate clamped to a frame was to keep it from changing shape while it's off the ribs. Fitting a bar to a twisted, or warped plate, could cause problems down the road. I don't put any spring in my bars. I fit as closely as possible to the plate (with chalk), and then glue. I imagine that trying to chalk fit a sprung bar would be a bit tricky.

[Adam Edwards]

Non springer

[Ken_N]

I glued my last one in after the belly was glued on the ribs, and before I set the neck and glued the back on. I fit it with chalk as close as I could and just glued it with a rub joint. I would guess there is not much tension. Seemed to work fine, but what do I know! I just finished a belly today and plan to glue the bass bar on it the same way. I have to finish the back and make a neck first because I'd rather do all the gluing and not leave it halfway glued up. But then again, how much tension are we talking about anyway? It seems like the belly is so flexible, 1mm here or there doesn't make it that hard to push down.

[Tim McTigue]

As a starting point for someone with no experience, should the frame be shaped, bent or shimmed to represent the shape/tapering of the ribs before fitting the bass bar?

I feel qualified to answer this, since I have only slightly more than no experience...

I worried about this quite a bit before I tried it. I made my violin with all of the taper in the upper bout (i.e. from the upper corners to the neck block), as I had read a comment by Michael Darnton to the effect that this was how Stradivari did it. Therefore, the belly is somewhat tensioned just for that reason. Or so one would think. I fit my bass bar on a simple plywood frame, which was flat, not shaped or bent in any way. So the bar was fitted to the belly while the belly was held flat on the frame, and no spring was introduced (not intentionally, anyway). When I glued the belly to the ribs, I was worried it might be difficult, but my fears proved groundless. It was a comment from Jeffrey that convinced me it would work out, and it did. There is really very little tension in doing it this way, I discovered, other than the normal tension one feels in doing something like this for the first time...

[stradofear]

I used to think that only springing was right, but I've come around to a mellower view: I think it's a lot like with posts. If you're making a violin for a pusher, someone who plays hard and rarely tests the delicate end of things, they'll be happier with a bar and post that push back. If you are making for a light, sensitive player who looks first to the gentle end of the spectrum, and rarely taps out the extreme pressure end, that calls for less spring and less post tension. In between are people who are in between. Most players fall on the gentle end, or as betweeners, and most or many of the virtuoso types, real or imagined (say about 30 players, world-wide, and everyone in New York City) are at the other. Bar for your customer.

That's the working theory at this time, anyway.

I shim so that the curve of the top matches the general direction of what it will have to join up with relative to the ribs. If you don't do that, what you're doing is one sidedly random, relative to spring, but I don't do it as much as to match the ribs exactly, though. and do like to see the curve of the top match the ribs when I glue up. There's a lot to this issue that doesn't make a whole lot of sense because it's hard to control everything, and I find myself trying more to shoot down the middle on the issue of matching the ribs. I do think that if you spring a bar and don't spring the top when you fit it, with a bent rib plane you run the risk of putting an effectively negative spring in the bar, which doesn't sound right at all. Right?

If you don't spring the bar, and you don't spring the top on the frame, but the ribs are arched, then you end up with a double negative unsprung bar, and that must be worse than bad. Double worse, in fact.

Somewhere in there is the absolutely correct answer. :-)

[David Burgess]

Nice post, Strado.

I've probably done some triple reverse negative springing on old instruments in my day, each step unintentional.

[Janito]

I shim so that the curve of the top matches the general direction of what it will have to join up with relative to the ribs. If you don't do that, what you're doing is one sidedly random, relative to spring, but I don't do it as much as to match the ribs exactly, though. and do like to see the curve of the top match the ribs when I glue up.

I find this very difficult to follow.

Would you please expand.

[Tim McTigue]

Would you please expand.

Oh, great, now he's going to be even bigger...

... ah, it must be spring...

[FiddleDoug]

"If you don't spring the bar, and you don't spring the top on the frame, but the ribs are arched, then you end up with a double negative unsprung bar, and that must be worse than bad. Double worse, in fact."

I don't think that the ribs are supposed to be "arched". Sure, on many instruments, the rib heights are different at the neck and tail, but, as far as I know, this is a straight taper, not an arch. I've also never heard of springing the top on the frame. Do you have any references to support this?

[stradofear]

Sure, Sacconi and Hargrave, and general knowledge. Sacconi says the taper is from the upper block up, on the back, sort of like on a bass, which in real life does affect the plane of the front side of the ribs, and Hargraves, who wrote a whole article about it for Strad mag in the 80s, that you can probably find if you look, says it's from the upper block up, on the front, based on scribe marks he saw on Strad ribs. I also found this on his site: http://www.roger-hargrave.de/PDF/Artikel/S...lanollo_PDF.pdf Search for the word "taper" in the article and it will pop up.

I also recollect that Gregg Alf wrote somewhere that he felt there was a bit of a tonal advantage to doing the taper from the upper blocks on the front as opposed to the back, but I found that the disadvantage to this is how it screws with neckset overstand, which you have to be aware of and compensate for. [Don't discount this comment at all: in my experience losing 2mm of overstand is a genuine big deal]

I don't know that it really matters much in the long run (OK, I don't really know that it *doesn't* matter either, which is why I do it as Hargrave has observed, since I figure doing things right can't hurt), as long as you know which method you're following, what plane or non-plane you're slapping your top on, and then adjust everything around that as dictated by what's really happening to the "plane" of the top.

[Darren Molnar]

"If you don't spring the bar, and you don't spring the top on the frame, but the ribs are arched, then you end up with a double negative unsprung bar, and that must be worse than bad. Double worse, in fact."

I don't think that the ribs are supposed to be "arched". Sure, on many instruments, the rib heights are different at the neck and tail, but, as far as I know, this is a straight taper, not an arch. I've also never heard of springing the top on the frame. Do you have any references to support this?

http://www.roger-hargrave.de/PDF/Book/Chap...e_Mould_PRN.pdf

page 17 3rd and fourth paragraph, I think

[stradofear]

I find this very difficult to follow.

Would you please expand.

Map it all out, and if you can think real good in 3D you'll see what the problems are. I'm not trying to be difficult: it's a complex spatial problem that's more visual than verbal. Just imagine a bar set for a flat top, then curl the ends of the top down over a non-plane ribset [imagine the violin sitting bridge-up, horizontally], which pulls the ends of the bar down, which pulls the middle of the bar down because the bar is a bar, and stiff, also (the bulk of the top being held UP by the corners and the ribs) and you can see that positive spring can quickly become negative as the middle of the bar is pulled downwards by the ends, and the bar, rather than pushing the top up in the middle, pulls it downwards, in the same direction as the bridge is pushing.

That's if you have not sprung the bar but have "sprung" the ribs. If your ribs are really flat and the top is, and the bar isn't sprung, you're home free, maybe. It all depends on what your various philosophies are for rib contour and bar spring. But it should be clear that about the worst thing you can do is follow the Strad pattern of tapering your ribs on the top from the upper block up, and putting on a top with an unsprung bar fitted on an unsprung bassbar frame. Which is probably why all of the old printed references from the 1700s--all two of them, Cozio and Marchi--refer to sprung bars as the norm.

[Darren Molnar]

Do you have a prefered working order when fitting your bassbars?

as in, fit the middle of the bar, then fit the ends by rocking the bar. Or fitting he whole bar, then planing the ends a bit to get the spring. Have you found an efficient method that works best for you?

Are you springing the whole bar, or the outer thirds, or other?

[Magnus Nedregard]

I've seen plenty of instruments that are virtually banana shaped, with the glue line for the table being both the convex and the concave part of the banana. Just had a cello in, where the c bout ribs were about 6 mm shallower than what they were in the widest parts on the instruments. So with the frame system, you may still end up wrong. In these extreme cases it is better to sort out the ribs though!

[David Burgess]

On old instruments, the ribs can be any old shape, and so can the top. I eventually came around to shimming a top on the frame to match the rib shape. Even if one isn't concerned about the stresses between the top and the bar when gluing a straight top to tweaked ribs, and any effect this might have on sound, such stresses often result in a noticeable "print-through", a ridge or deformation at some point at the outline of the bar. The top will bend easily to match the ribs and back. The bar won't.

[stradofear]

I was taught to force things around, but for the last 15-20 years have been shimming stuff up as David does. If you think the bar tension, whether you use any or not, is critical, you can't be gluing a bar into a top contour that doesn't exist in the finished instrument, then twisting the top into that shape with a bar that was intended for a different shape.

Two more things to be aware of if you want to avoid ripples in the top are too much water (when washing out squeeze-out) and putting all of the clamps on one side of the bar (something I saw on one maker's site the other day). Imagine the combination of soaking that top up real good with a lot of scrubbing to get that last bit of glue off, and then twisting it sideways with a couple of pounds of weight (the clamps) and a thin caul (the bar). The ends, where the top is endgrain, are particularly vulnerable to getting quickly soaked with a thin glue wash, being forced into a new shape, and being locked into that shape when the glue sets, which is where I think how a lot of the ripples at the ends of bars are formed, especially in new instruments.

The other way to make ripples (Darren) is to put the spring at the ends of the bar where it's not really needed. If you do that you run the risk of a little pooch right at the end of the bar where suddenly the top isn't being forced into a curve to meet the bar. I put most of it in the middle, where all of the support is needed, when I do it, and the ends click down flat for quite a distance in.