Archtop deflection testing

[Robert Freemond]

I'm interested in starting to do some deflection testing on my archtops, but so far I haven't come up with any numbers. I've seen some cool looking gizmo's that I can make, but I don't know how much weight to use or any of the other stats. Does anyone know of books or articles dealing with this subject ?

[Dave Stewart]

There's very little out there re deflection testing archtops. Do a search at mandolin cafe for "deflection" in title & you'll find quite a bit (at least for mandos), much to do with Dan McRostie & his jig / approach. Every jig & setup is different, so you can't really compare one persons results to anothers. Whatever you use, you have to do things consistently & then track you own results, so you can compare. I use a cantilever arm and freeweights to generate a point force of 12.5 and 25 lbs. at the center of a self-adjusting bridge on mine.

[Michael Lewis]

Dave, what deflection readings are you looking for? Also, what has brought you to using these values? What gauge strings are you building for?

[Dennis Duross]

I'll pile on here too.

Dave: You gave a formula for determining downward force in a different thread just recently, which differs quite a bit from the one I'm using.

Your recent post:

The formula for downforce is Downforce = 2 x T x sin(breakover/2) where T is total string tension.

I've been using:

Downforce = T*2*SIN((breakover*(3.14/180))/2)

I can't remember where I got my formula, but assuming 12 gauge strings and a 13 degree breakover, mine gives 37 lbs vs 70 lbs with your formula.

Any thoughts?

[Mario Proulx]

It'd be simple and quick enough to just make a 'dummy' of a 4' length of 2x4 or something, 6 tuners, strings held to the two-by with 6 nails, and a scale under the bridge area. Set it up so that the scale length and afterlength are correct, as well as your chosen bridge break angle, string it up, tune to pitch, read the scale. Change the break angle up and down to take other readings. Scribble-down the notes, call it done.

The bigger question for me is, how much deflection do we want at pitch, regardless of the break angle?

Interestingly, after reading-up on Don's methods for the mandolin, I made a jig like his, and using his weights, found that all of my existing mandolins fell within the ranges he'd found on the Loar mandolins he'd measured. Yet, mine don't sound anything like a Loar. So I concluded that for me, it's a nice method to judge when you're still on the safe and proven side of structural integrity and when you might be treading toward dangerously underbuilt(or conversely, toward being overbuilt), but when it comes to tone, there's a -lot- more to it.... I'm sure Don knows that, too, but he ain't telling the rest of the story; the man builds a killer mandolin!

[Dave Stewart]

Dave, what deflection readings are you looking for? Also, what has brought you to using these values? What gauge strings are you building for?

Man, isn't that the Holy Grail of questions...I've been lookin' for someone to answer that for me for years. (Some numbers out there for mandos, but none for archtops that I've found)
Since deflection is only one piece of the puzzle (and can be the same with a low density plate vs a thin plate vs a thick plate/thin recurve for eg.) I'm reluctant to say for fear it might steer someone in the wrong direction.
But what the hell, I'll show you mine if you'll show me yours. After starting out, like most, fairly stiff, my most recent ones have been in the .045" to .050" range of deflection with my jig/methods. I've been pleased with the acoustics and fullness and am inclined currently towards this level of stiffness. (I typically string with 12-50's) What readings do you look for??

Dennis....ooops, you are correct! I've had that in my notes for a long time but, no longer having a working scientific calculator, anytime I need downforce I use the Mottola online version ( http://liutaiomottola.com/formulae/downforce.htm ) which uses your version of the formula. My bad. Sorry for any confusion.

[Michael Lewis]

Dave, I haven't been using deflection testing, I'm just curious. I can see that when a person gets a good set of 'numbers' to work to the results should be fairly consistent.

My approach has been by carving to more or less predetermined thicknesses, depending on the qualities of the wood. Maybe I should set up a rig to measure the deflection of one of my guitars to see how much it deflects. I had considered measuring deflection many years ago when I started making mandolins but never pursued the thought because I was getting good results shooting from the hip. I'm still shooting from the hip I guess, but I have developed some proven methods I have grown accustomed to.

[Dave Stewart]

Go with what works! And your approach certainly does. No doubt you do measure deflection, although more intuitively, with thumb pressure etc.
Whenever I go down such roads, I get an almost obsessive desire to learn from results that others typically get..... what deflections did the clearly acknowledged "greats" see? .... what graduations? ... what main modes? It's a journey.

[Dennis Duross]

Dave: THAT'S where I got the formula that I've been using. The Mottola website. I couldn't remember. I just have it plugged into an Excel spreadsheet.

[Robert Freemond]

Dave, Can you explain why you use two dial indicators.

By the way, there is an article on this subject with some numbers in American lutherie page #6 "Low Stress Archtop design by steve Grimes

Thanks for your help

[Dave Stewart]

I guess I just used the McRostie jig as the basis..... centerline load with deflection measurements either side. Seemed good to me. And you can "tune" things a bit as you go (bit stiffer on treble side?) if you want.
I have that article and have been influenced by it in building. Lots of great "pearls" in there for the archtop builder.

[Mario Proulx]

Hmmm.....

I've played around with very asymmetrical bracing on mandolins, to the point of trying a single tonebar/brace, and found that there was a lot of potential, but placement was so "touchy" that I couldn't justify pursuing it further.
But with two indicators, and working on a free plate(in other words, not fixed to the body yet, or fixed to the ribs but the back not being in place yet), I could perhaps have found that "happy place" quickly and repeatedly!

I love this place......

[Dave Stewart]

Dave: THAT'S where I got the formula that I've been using. The Mottola website. I couldn't remember. I just have it plugged into an Excel spreadsheet.

Dennis, I used one of several online scientific calculators http://web2.0calc.com/ to run some numbers.
To add more confusion, the Mottola downforce calculator interestingly does not use the formula it states (ie Downforce = T*2*SIN((breakover*(3.14/180))/2) which yields ridiculously small results).
It generates results which exactly correspond to my original formula -- Downforce = 2 x T x sin(breakover/2)
Frans Elferink wrote a great article for American Lutherie #74 called "Forces on Archtop Guitars" which included my original version of the formula as well as a vector diagram of the forces that confirm a typical downforce in the range of 1/3 total tension. (I downloaded this article long ago, which I guess is where I originally got the formula, but I can't seem to track it down currently...maybe no longer avail (except in the old AL) although it still appears on Elferinks page http://www.elferinkguitars.com/downloads.html )
The Steve Grimes article that Robert mentioned also had another version of the formula ...sin((angle*pi)/2) which again yields results far to high.
Anyway, I'm sticking with my original (and it looks like Mottola is too!)
(The 70lbs you mention leads me to think breakover is not being divided by two properly)

[Jeff Highland]

I think Mottola's formula includes conversion from degrees into radians for use in calculation tools such as excell which only work in radians.