Top Deflection rig

[DarwinStrings]

Well yous have made this job a easy one for me, thanks for sharing.

Jim

[DarwinStrings]

So with help from here I built a rig, took the measurements and did the maths. As I am not confident mathematician I sent my measurements to Jeff Highland and he did that maths for me to check if I was getting it right, we got the same results and I am happy to announce that I have a guitar with a monopole mobility of 15 x 10^-3 s/Kg. Thanks again Trevor.

Jim

[Edzard]

Hello All,

Please tell me/confirm to me that the spindle of the (analoge) gauge your all using goes "through" the housing of the gauge...

Other wise I'm going nuts how this works with the above jigs!

I'm asking because (a) I want to build one of my own and (b) I have a few digital gauges and there the spindle on top of the housing is fixed.



Thanks,


Edzard

[Craig Bumgarner]

Yes, the dial indicator I use goes all the way through the dial part on the housing. The load is applied to the spindle protruding from the top of the gauge and applies pressure to the top by way of the spindle coming out the bottom of the gauge. The top deflects and the gauge measures.

Many dial indicators have a cap on the top that protects the top of the spindle, but this can be removed by unscrewing it. All the dial indicators I have seen have the spindle protruding from both sides, otherwise it is much more complicated to support the spindle within the dial body.

[Edzard]

Thanks Craig!

Found this neat construction. It's from Michael Cone ( http://www.coneguitar.com/acoustics/a02.html)

Edzard

[Craig Bumgarner]

That's a nice looking jig, but it has a major flaw. The way it is set up with the base on the table, it will measure not only the top deflection, but also the guitar side deflection, the deflection (compression) of whatever the guitar is sitting on, plus any play in between. These will be considerable and likely affect the measurement substantially. (I know because this is how I made my first jigs). To get rid of all this, the jig needs to rest on the top of the guitar like some of the pictures earlier in this thread. Also, watch out for that pretty brass weight if you happen to bobble it

[jeffhigh]

Yes Craig , that is the problem and why I designed mine to sit on the edge of the soundboard.
It does not then matter whether the guitar is sitting on a towel, foam blocks, or surgically enhanced breasts.....

[JamesO]

I'm trying to understand why and how we test top mobility, but I'm having trouble finding it in the table of contents in Design and Build. Can someone point me to it? Thanks.

[Trevor Gore]

There are plenty of listings under mobility and monopole mobility in the Design index.

In Contents, Sections 1.7.1 and 1.7.2. The 2nd edition also has an Appendix which goes into the measurement process in a bit more detail (which is mostly covered in this thread).

Monopole mobility is important because it gives a measure of the sensitivity and responsiveness of a guitar, i.e. its potential to be able to produce a good deal of sound.

[Trevor Gore]

The traveling 3 day modal tuning course goes far and wide, which means hoping the gear at the destination works OK, or taking your own. The problem is my gear is all made of heavy hardwoods so easily exceeds a baggage allowance, so I thought I would re-do my top deflection rig to a lower mass version and make it multi-purpose so I can use it on a three point bending rig as well.

Here's the new top deflection rig:

DSCF0004s.jpg (75.48 KiB) Viewed 50747 times

This is all pretty self explanatory if you've read through to early part of this thread. The cross beam is now spruce, so less than half the mass of its Tassy oak predecessor. The gauge has gone low mass digital. It's basically a tyre tread depth gauge by iGaging (as used on the LMII string height jig). The loading is through the shaft as discussed above and the load is 1kg of lead shot in a plastic bag in the "velvet" bag (which is the lining material of Hiscox cases).

The loading platform is a push fit on the gauge shaft:

DSCF0033s.jpg (156.64 KiB) Viewed 50747 times

The gauge has all the usual digital benefits of "zero" and "hold" buttons and reads in mm, decimal inch and fractional inch for those who refuse to use SI units!

The shoe that sits on the guitar (see first pic) now has an extra bit of wood glued on the outside, so the cross beam can be used on my new low mass, knock-down, 3 point bending rig:

DSCF0016s.jpg (87.42 KiB) Viewed 50747 times

Cross grain stiffness can be measured as well by setting the beam diagonally across the test panel:

DSCF0013s.jpg (72.89 KiB) Viewed 50747 times

The top loading and top reading indicator is much easier to set up and use than my old rig, which was a troublesome top loading arrangement with a difficult to read dial indicator beneath.

The last bit of new gear is a low mass Chladni sig gen, but I'll start another thread for that.

[johnparchem]

Trevor, thanks for the post. I am going to convert mine to an igaging digital probe. Half the price of the modified LMI version for string action measurement.

[Dave M]

Trevor thanks for the description of your updated design. Your jigs always look so neat compared to ones own rather scruffy efforts.

It does still bother me that you would rely on a single measurement of the deflection of the top with an applied mass. A number of measurements with a series of increasing masses is surely a better experimental technique...?

You show a mid point deflection arrangement, presumably to measure Young's modulus of a half top. Does that imply that you are using this technique as well as the resonant frequency method for getting at this value?

All the best Dave M

[Trevor Gore]

It does still bother me that you would rely on a single measurement of the deflection of the top with an applied mass. A number of measurements with a series of increasing masses is surely a better experimental technique...?

Sure! I have three shot bags, but I also have very little problem with repeatability.

You show a mid point deflection arrangement, presumably to measure Young's modulus of a half top. Does that imply that you are using this technique as well as the resonant frequency method for getting at this value?

I nearly always use the tap test method. However, occasionally I have wood that can't be reduced to a tappable rectangle (think Euro spruce with corners cut off, or a Braz back that is "coffin" shaped). The 3 point bend comes into its own then because it is usually possible to isolate a rectangular shaped span and get a reading. Stuff that is cupped or twisted can be measured, too, by pre-loading to flatten the cupping/twisting and taking measurements with the cup/twist facing up then down (and taking multiple readings). Probably not as accurate as tapping flat stock, but it gets you a reasonable number to work with.

[GregHolmberg]

Here's my very crude rig. It can be adjusted for both height and width via slots in each member. It works, but I think there's room for improvement.

For a weight I used a 1 kg lead diving weight (1006 grams, actually).

On this Ibanez archtop (the top of which is greatly overbuilt), I got a deflection of 0.02 mm, giving a stiffness of 50,300 kg/m.

I'm concerned about the resolution of the iGaging depth gauge. If I can only get values of 0.01, 0.02 and 0.03 for this particular guitar, then I can only get stiffness values of 100,600, 50,300, and 33,533. It seems like I need a more accurate gauge.

What other gauge products are people using? Something that reads to 0.001 mm, I think.

Greg

[GregHolmberg]

Thinking about it more, the problem with this guitar is that it's just too darned stiff!

First, I should have done the calculation in newtons, not kg. In which case, this top has a stiffness of about 490,000 N/m. I think a typical stiffness for a steel-string flat-top is about 47,000. So this top is about 10X stiffer.

At a typical stiffness of 47,000, deflection would be about 0.21 mm. To get that deflection on my archtop, I'd need 10.6 kg!

Hmm. Maybe it's better to move onto testing a different guitar. I don't think I want to put that much weight on this rig or this top.

I can see that a resolution of 0.01 might be fine with a less stiff top. With my 1.006 kg weight, 0.21 mm would result in 47,000, and 0.22 would result in 44,800. That would be increments of 4.5%. Maybe that's close enough.

Greg

[hilokawika]

Aloha from Hawaii,

It's so interesting to me to see how mainstream measuring top deflections of various sorts has become since I first tried to popularize them more than 20+ years ago on my website ( https://ukuleles.com/applying-technolog ... mpliances/ ). It's terrific how generous everyone is on this site with respect to sharing their innovations and jigs...

aloha,

David Hurd

[kiwigeo]

Welcome to the forum David,

Yes I recall spending hours on your website back when I started building...the site was and still is a great source of information.

Aloha from Hawaii,

It's so interesting to me to see how mainstream measuring top deflections of various sorts has become since I first tried to popularize them more than 20+ years ago on my website ( https://ukuleles.com/applying-technolog ... mpliances/ ). It's terrific how generous everyone is on this site with respect to sharing their innovations and jigs...

aloha,

David Hurd

[GregHolmberg]

OK, the last one I showed was a bit crude. But I learned a lot, and have tried to improve it.





For one thing, I bought a better indicator (US$39), which claims 0.001 mm accuracy. Well, it reads to three decimal places anyway...

Two improvements.

1. Longer edge resting on the guitar prevents wobbling.

2. A lever arm amplifies the movement 10:1. It pivots on a small nail 50 mm to the right of the center, and the indicator is 500 mm from the nail.

I think there are still some opportunities for improvement.

I'm having troubles getting consistent readings. I can lift the weight and put it back down and get a reading that's 10% different. I'm not sure what's causing this. Maybe I should just use the traditional method without the lever arm.

I've also tried various weights. I expected the same stiffness, K = F/y. That is, I expected the force:deflection ratio to be constant as the force increased--i.e. if I double the force, I expected double the deflection. Instead I found K was smaller with more force--that is, I got more deflection than I expected with the bigger weights, based on the measurement with the smallest weight. Almost twice as much deflection than expected. This is telling me something, but I'm not sure what. Is it telling me something about the guitar, or something about the rig?

Greg

[kiwigeo]

I reckon that Ibanez would be built like a brick shit house....

[GregHolmberg]

I reckon that Ibanez would be built like a brick shit house....

It sure is. Despite being fully hollow, it's not really an acoustic instrument. Although it's louder than a solid body guitar.

I put 4.7 kg on the rig and only got 0.11 mm of deflection. So K is somewhere around 420,000 N/m and the top's main resonance is 360 Hz. MM about 5. Has two big parallel braces and the top is a 5 mm lamination. So yeah, way overbuilt. (By the way, I calculate, based on string tension and the break angle, that normal down force on the bridge is about 15 kg).

Very few arch-tops are actually acoustic instruments. Maybe a few built before the pickup was invented. But even those have little to no bass response (by design). There are a few acoustic arch-tops still built. Loar makes some. I suppose you could consider a Selmer-style to be an arch-top.

Most are really electric guitars, so I'm not sure why Jazz players don't just get a solid body guitar. Tradition, I suppose. They are fairly light-weight, but also so big it's hard to get your arm around them.

As Ken Parker says, the evolution of the acoustic arch-top ended in about 1935, and we don't know what the limits of that configuration really are. He's makes some excellent-sounding acoustic arch-tops. Here are some recordings. These definitely have bass response. Whether they sound like a flat-top or not, I'll leave to you to judge.


Greg

[Trevor Gore]

Is it telling me something about the guitar, or something about the rig?

Most likely the rig. The fewer moving parts (to minimise slop and backlash) and the lower the friction the better.

[edvinguitars.fi]

Thanks for all the pics for deflection jig implementations. I proudly copied yours to my jig.

After doing some repetitive measurements, I got a mobility result of ~20 for my falcate braced guitar. The measurement still has some error sources:

- I need to stiffen the horizontal bar to guarantee it does not bend at all by the weight applied. If it bends, I think I get "too" good results
- I played around with the excel sheet with the book equations : as my gauge guaranteed accuracy is +-0.01 mm, I entered such variance to my end result: seems that if I alter top deflection by 0.01mm to one direction or another, the mobility result changes from 20 to 23 or 17 , ie > 10%. For accurate results, I may have to weigh using different weights (like 250-500-750-1000g and use linear fit to find the exact value for K . My feeling is the bigger deflection I get, the smaller the tolerance impact of the gauge becomes.

[kiwigeo]

Orienting the horizontal bar with ply laminations running vertical would reduce risk of flexing.