4 Bridge System Madness.

[Pete Jenner]

Kevlar body armour donned and securely fastened, I sit prepared for the ridicule, derision and ignominy that will doubtless ensue from this thread.

Some months ago I bought some cello tuning pegs from my local music shop and ordered some thumb wheel sets from Stewart MacDonald for the purpose of making these adjustable bridge posts.



This is an idea that has been brewing for about four years. Initially I thought it would be an interesting way of reducing crosstalk or phase additions and subtractions in the bridge. A respected scientist/luthier who is a member of these forums informed me that phase interactions in bridges is not significant enough to warrant such an exercise.

Not one to be easily deterred, I kept developing the idea with a different set of objectives in mind.

The driving motivation is simple curiosity. I want to see if this system results in timbral differences or changes the amplitude of an instrument compared to a 'normal' (in this case a Brekke) bridge.

Along the way, I'm hoping to dispel a few myths. A lot has been said about mandolin bridges over the years and I believe much of it is complete bunkum. For example, people talk about the tonal and timbral effects of 'wings' and strategically drilled holes placed in certain relationships to the string pairs.

If this system results in a mandolin that sounds and behaves no differently from one with a normal bridge, I think we can discount many of the popular misconceptions regarding bridges.

The mandolin I chose for this experiment is my No. 2 'The Heretic'. A mandolin that many people here have heard and/or played in real life and can attest that it's a fine sounding instrument.

One of the things that has held up this experiment is my belief that I needed to build a special 'non-string gathering' tailpiece in order that the lateral forces didn't pull the posts over. Yesterday it occurred to me that a simple string spreader would do the same job. ....well I was sort of right.

INFO: Each bridge post in this system weighs 3 grams for a total system weight of 12 grams. The original bridge weighs 17 grams.

More to come (I have to go to bed and change computers)...

[Timbofood]

Hmmmm.
Interesting, glad you are fishing this hole Peter, way too deep for me!
Have fun and put on the popcorn!

[sblock]

VERY interesting experiment. We all (well, at least some of us!) await your findings with bated breath. Please realize, though, that the added mass of the wooden "string spreader" you introduced between your new, separate bridge posts and the existing tailpiece will certainly affect the mandolin sound. What does this additional part weigh? To a good approximation, its extra mass can be expected to add a bit of extra sustain to the sound, just as a mute (usu. placed on or behind the bridge) would, and it may also 'roll off' some of the higher frequencies.

[Steve Sorensen]

And . . . ? What are your initial impressions of the Heretic 2x4?

Steve

[Atlanta Mando Mike]

Really cool experiment. I'm very interested to hear how it turns out. Another idea, and what would be cool to me-a player with absolutely zero ability in the lutherie department, would be to take your idea and attach it to a fixed base so that it would be easy to adjust the action of each string course separately.

[Pete Jenner]

VERY interesting experiment. We all (well, at least some of us!) await your findings with bated breath. Please realize, though, that the added mass of the wooden "string spreader" you introduced between your new, separate bridge posts and the existing tailpiece will certainly affect the mandolin sound. What does this additional part weigh? To a good approximation, its extra mass can be expected to add a bit of extra sustain to the sound, just as a mute (usu. placed on or behind the bridge) would, and it may also 'roll off' some of the higher frequencies.

Interesting you should say that Steven, the mass wasn't a factor but the spreader caused big problems.

When I first strung it up and played it, I was going to give the whole thing up as a bad joke. The mandolin sounded like a cross between a banjo and a violin being played pizzicato. It was dreadful, there were strange screechy overtones on every string but especially on the Es and Gs and nearly no sustain at all.

At first I thought it was the bridge posts but thinking about it more carefully, I realised it was the string spreader. The was no secure anchoring of the ends of the strings so I blocked up the ends of the spreader with some small blocks of wood and voila ...it sustain returned and banjo/violin sounds vanished. It sound pretty normal and not half bad. I still need to make or have made a special tailpiece but until then, all the tests I do with the Brekke bridge will also use the string spreader.



Every test will be recorded and if anyone has good ideas about what tests I should perfom, I'd be glad to hear them.

Apart from the mandolin, the 4 bridge system and the Brekke bridge, the equipment I shall be using is: A Shure 55SH MkII dynamic microphone even though I would prefer a condenser mic but I don't have one. A Digidesign Mbox2, Audacity for FFT analysis and 64 bit Cubase 7 elements for other more musical recordings running on Windows 7.

[Timbofood]

Mind of like Mike's solid base idea, might that help more than the spreader set up?
Mass of the spreader? How much contact with the "feet" on the spreader? Does that add up to contact size of a "single bridge foot"?
This is the kind of thing that really makes me happy to be a member! The ideas, both forward thinking and historically directed are just fascinating.
The little grey cells are having a field day!
Thanks Peter!

[Pete Jenner]

I've thought about that Tim but I'd be concerned that it would add too much mass and restrict the ability to adjust the intonation on each pair separately.

The spreader is about 2 grams. If I get good results, I'll get rid of the spreader and use a non-string gathering tailpiece.

[sblock]

Interesting you should say that Sean (Shaun?), the mass wasn't a factor but the spreader caused big problems.

When I first strung it up and played it, I was going to give the whole thing up as a bad joke. The mandolin sounded like a cross between a banjo and a violin being played pizzicato. It was dreadful, there were strange screechy overtones on every string but especially on the Es and Gs and nearly no sustain at all.

At first I thought it was the bridge posts but thinking about it more carefully, I realised it was the string spreader. The was no secure anchoring of the ends of the strings so I blocked up the ends of the spreader with some small blocks of wood and voila ...it sustain returned and banjo/violin sounds vanished. It sound pretty normal and not half bad. I still need to make or have made a special tailpiece but until then, all the tests I do with the Brekke bridge will also use the string spreader.



Every test will be recorded and if anyone has good ideas about what tests I should perfom, I'd be glad to hear them.

Apart from the mandolin, the 4 bridge system and the Brekke bridge, the equipment I shall be using is: A Shure 55SH MkII dynamic microphone even though I would prefer a condenser mic but I don't have one. A Digidesign Mbox2, Audacity for FFT analysis and 64 bit Cubase 7 elements for other more musical recordings running on Windows 7.

Hi there. Actually, my first name is Steven, not Sean. But no matter!

In your modified setup, the string spreader has sprouted two "feet" that contact the mandolin top directly. These can be expected to act as a kind of auxiliary (secondary) bridge, because any vibrations of the strings and 4 bridge posts will be communicated (although somewhat damped) right back to the spreader, and then down onto the top. My instincts are telling me that this experiment will not, therefore, offer a clean way to learn what you're trying to learn: namely, whether 4 separate string posts that inividually contact the top can replace, or improve upon, a single bridge with a saddle that either has two feet or is full-contact. You will get some results, of course, but these will be darned hard to interpret. Can't you figure a way to get rid of the top-supported spreader? Are you absolutely SURE you need it -- the four posts that you made, if fitted closely to the top, might just resist the lateral string forces perfectly well (given the high downbearing forces on the top and all that associated friction), and eliminate the need for a spreader. Regardless, I think you need to find a clever way to do this experiment without added additional elements that vibrate the mandolin top. But keep up the work!

[Pete Jenner]

Ah sorry Steven. I could have sworn I've seen you sign your name Sean in the past. I think I'm getting old timers disease. I just edited my post accordingly.

Yes I agree that the spreader is less than ideal but it shouldn't act as a bridge because the string vibrations all occur between the bridge and zero fret and the spreader is a long way back. This experiment will be ongoing and will eventually be done without the spreader.

I tried the posts without the spreader and while the G and E posts don't fall over under full string tension, the do tilt a bit. That's why I need to get a wide tail piece. I'd be interested to hear from anyone who can tell me how to make one.

[Timbofood]

What about a spreader between the adjuster posts, sort of a strap with holes for maintaining the interval?

[Pete Jenner]

I'd rather keep them completely separate but I understand where you are coming from.

[sblock]

I certainly don't mean to sound argumentative -- and I mean to encourage your experiment! -- but I believe that you are wrong about the spreader-with-feet not acting as a secondary bridge.

The tension in the strings causes them to bear down on the main bridge (usually through the saddle, or in your case, through 4 separate saddles), and after the strings are sent into vibration by the energy introduced by the pick, some of that vibration, and energy, is transmitted through the bridge and down into the top. But not all of it. Quite a bit of the remaining vibrational energy travels rearward, beyond the bridge saddle, and towards the tailpiece. In fact, that is the reason that many mandolinists use rubber grommet dampers, Weber "wood nymphs", or bits of felt or leather to dampen these vibrations, which can cause parasitic, unwanted sounds. And this same vibrational energy is also what causes loose tailpiece covers to rattle! So, it is a significant amount of energy. In your current setup, you are coupling the vibrational energy of the strings behind the bridge to the mandolin top through the feet of the string spreader. It will definitely act as a secondary bridge. Whether it turns out to be a small or a large perturbation, over-and-above the primary bridge, remains to be determined, but an effect is to be expected. Also, by bearing down, the two feet of the spreader will tend to produce a NODE line (a line of zero vibrational amplitude) in the complex modes of vibration of the top, thereby suppressing any of the (many) vibrational modes that happen to have high amplitude under the location of the spreader feet. (This is equally true of the regular mandolin bridge, which is one of the reasons why Lloyd Loar moved the F5 bridge more towards the center of the top, by lengthening the portion of the neck before it reached the body to 15 frets, compared to the earlier, shorter-neck F4). That will also change the sound, for sure.

Anyway, theory only tells us so much. You have to do the experiments! But, I hope you agree, these need to be "cleanly designed" to test what you think you really want to test -- and not something else. If you want to test the effect of 4 separate bridge saddles instead of just 1 common saddle, you ought not (1) mass-load and dampen the strings behind the bridge position with a spreader, and (2) have that spreader transmit vibrations to the top. It's still an interesting sonic experiment as you have it, of course, but I would not know how to interpret the results. Any change in sound or timbre could be due to the separate saddles, but it could equally well be be caused by the spreader. Or both! You will be left with a mystery. You want to design your experiment to change as few things at a time as possible.

[Pete Jenner]

Argument is good. Thanks for taking an interest. I'll read your post properly when I wake up and give it some proper thought.

[Jacob]

The mandolin equivalent of a banjo straight pull tailpiece might help.

[Pete Jenner]

The mandolin equivalent of a banjo straight pull tailpiece might help.

Yes. That's what I need Jacob.

[fscotte]

I'm a firm believer that current bridge technology is as good as it gets. There's only a few things that a bridge has to do, and the current Loar bridge does em all. But still, I like to see your results.

[sblock]

By contrast, I'm a firm believer that there's nearly ALWAYS some room for improvement when it comes to acoustic instrument design. How much room? It depends! Modern violins are now being built that are acoustically indistinguishable from Golden Age Cremona violins made by Stradivari and Guarneri -- even by some of the very best virtuoso violinists. So there are no "secrets" to these that cannot be replicated or even surpassed, it seems. Luthiery is a process, not an end-point!

Anyway, there is a whole lot not to like about the Loar-era mandolin bridge design! The saddle is known to be imperfectly compensated. Its shape also weakens it, and it is subject to breaking easily (many broken saddles). The two bridge feet on the base do not make optimal contact with the top (which is why some have elected to get full-contact bridge bases, instead), and the actual contact area of the two-foot base changes quite a bit from instrument to instrument, due to the need for sanding and fitting to the top. The bridge saddle cannot easily be adjusted for height under full string tension, as we all know. Bridge saddles, especially those that have to be raised higher than normal, tend to tilt progressively more forward under string tension or with string changes. The tone of the bridge seems to vary a lot with the woods used, so it's not all that reproducible. Strings can get easily stuck in the slots. And on and on -- that's just a short list of SOME of the shortcomings of the Loar-era two-piece design.

That said, it still might be the best design out there at the moment. It's like what Winston Churchill said about democracy: "It's the worst form of government, except when you consider all the rest!"

But that doesn't mean one can't do better in bridge design!

[fscotte]

I guess it depends on what some consider an improvement. If I had a special bridge that added a dramatic amount of sustain, the knee jerk reaction would be how great that is. But on second thought, more sustain isn't generally a good idea for a loar style mandolin. The things you mention above aren't necessarily the shortcomings of a two footed bridge, but rather sloppy work by the builder. Solid contact from the feet and the top can be achieved with patience.

I've yet to see an "improvement" in bridges. There's been a whole assortment of experiments for many years. But the job of a bridge is very basic, there isn't anything magical about what it does.

[sblock]

I guess it depends on what some consider an improvement. If I had a special bridge that added a dramatic amount of sustain, the knee jerk reaction would be how great that is. But on second thought, more sustain isn't generally a good idea for a loar style mandolin. The things you mention above aren't necessarily the shortcomings of a two footed bridge, but rather sloppy work by the builder. Solid contact from the feet and the top can be achieved with patience.

I've yet to see an "improvement" in bridges. There's been a whole assortment of experiments for many years. But the job of a bridge is very basic, there isn't anything magical about what it does.

Well, I entirely agree that there's nothing really magical about what the bridge does! And I'd be inclined to agree that the assortments of experiments that have been done have not necessarily led, yet, to a "better" bridge than the Loar-era two-foot design. At least not one that's widely embraced by the mandolin community. But then, popularity/widespread acceptance by this community (which tends to be VERY conservative about design, for better or worse!) is not perhaps the right measure for whether one design is "better" than any other. Still, we've had some pretty nice -- and fine-sounding -- alternatives tried, like the Vern Brekke adjustable (wood cam) bridge, and also the Weber "Traditional" bridge, which has a metal crossbar under the saddle, and can be adjusted under full string tension. But these bridges are not all that widely accepted, nor -- to my ear -- do they sound noticeably better, so they have not displaced the bridges with the traditional Loar-era designs. I also agree with you that it depends, to a great extent, on exactly what one means by "improvement." Amen to that.

The only part of your response that I'd quibble with is where you wrote this: "The things you mention above aren't necessarily the shortcomings of a two footed bridge, but rather sloppy work by the builder." Sorry, but I disagree completely with that! For example: (1) the inability to adjust the bridge height under full string tension is an intrinsic design drawback of the Loar-era bridge, and not evidence of sloppy work. (2) The fact that mandolin tops all have slightly different shapes, and therefore that the contact area of the two-foot bridge, which must be sanded and fitted to each and every top, is different from one mandolin to the next, is not a result of sloppy work, either! (3) The imperfections in the compensation of the saddle (which also vary with the choice of string gauge) are also not evidence of sloppy work -- they are intrinsic to the shape of the saddle (and vary slightly with different makers of bridges, too). (4) Also, the tendency of the saddle to lean progressively forward, especially when the thumbwheels have to be raised to a high position on some mandolins, is also a design flaw of this type of bridge, and not a product of sloppy work. So no, these shortcoming are NOT sloppy work: they are inherent flaws with the design, and they have been recognized as such (repeatedly) on this forum, and by individual luthiers.

That said, the Loar-era two-foot design with thumbwheels works in most cases, and it gives many of us the sound we enjoy. But it's also a source of many headaches for mandolinists and luthiers alike, and a better-sounding and better-adjusting alternative would be great to have.

To cite an analogous example of progress in design: I think the Bill James version of the mandolin tailpiece, with its hinged cover and integral noise dampers and solid attachment posts, represents a HUGE improvement over the earlier Loar-era design, with its flimsy stamped parts, separate cover (that rattles, comes loose, gets lost, etc.) and fragile, punched string posts (that break, cut into the strings, and so on). Yes, the James tailpiece LOOKS just about the same, from afar, and it SOUNDS just about the same (maybe even better?), but it's -- hands-down -- a better design!! And just look at how many of the really top-end, single-luthier mandolins are featuring James tailpieces. Or one-piece tailpieces, of the type first pioneered by John Monteleone (e.g. the Allen tailpiece). These are all improvements over the 2-piece Gibson tailpiece design, IMHO.

I'm just waiting for someone to do something similar with the mandolin bridge. Yes, there is ample room for design improvement, folks. And yes, the mandolin of tomorrow will be better than the mandolin of today. Mark my words. We're not done improving the mandolin.

[j. condino]

"Kevlar body armour donned and securely fastened, I sit prepared for the ridicule, derision and ignominy that will doubtless ensue from this thread....."

Quite possibly the most insightful thing ever stated here!!!!

Peter, good on ya' for the creativity and curiosity and even bolder for wilingness to stand naked and post it around here; keep it up! As for the crusty crusty whole lotta nuthin' nay sayers....a simple two word phrase comes to mind.....

j.
www.condino.com

[Tom Haywood]

Pete, I am very interested to hear the outcome of this experiment. My two pennies: If this weren't an iconic mandolin, you could tap some small nails into the end for loop holders to accomplish the straight pull. I think with the spreader you have in effect split a normal bridge into two locations touching the top, and that will not present the data you wish to test. Also, with the four posts you have, in effect, doubled the number of metal posts found in a standard adjustable bridge. I am interested to see if this causes any differences in the transmission of the overall string vibrations to the top. I agree with Mike that you may be onto something with the ability to easily adjust string height and intonation individually (shades of setting up electric guitars). You may have to design the mandolin tailpiece for this.

[Timbofood]

OK Peter, you have strung this rascal up, what does your ear tell you? First impression?

[Steve Lavelle]

It seems to me that one function of the traditional bridge is more even distribution of the significant downward pressure across the top plate. With the 4 independent supports in this experiment, what is the area of each bridge piece base? It looks like there may significantly more pounds per square inch at each of the contact points, making the top more susceptible to cracking or sinking in the long run.

[fscotte]

You would lose the normal rocking motion that a traditional bridge applies to the top plate. BTW, bridges don't transmit vibrations to the top. The bridge acts as a lever, more of a mechanical force, not like electricity.