String Stiffness Jig results

[kiwigeo]

I need to check some maths. Can some kind person run data from my string stiffness jig and see if they get the same results:

Inputs:

String (Nylon) - Savarez Corum Low E (506J)

Diameter: 0.64mm
tension: 7.4Kg (72.59N)
L1: 325.0mm
L2: 650.0mm
F1 = 80.14Hz
F2 = 160.5Hz
d = 4.0mm

My Results:

Unit Area = 2.83 x 10-7m2
deltaT = 0.1994N
deltaL = 4.92 x 10-5m (0.0492mm)
k = 2633.08
E = 9.31GPa

Thanks in Advance

[Trevor Gore]

Martin, your maths is basically correct in that I get k=2633 N when I use 325mm for L2. However, your diameter of 0.64mm doesn't look right for a low E. Also, check your L2. I use the hypotenuse of the 4mm and the 325mm, which gives a slightly different result for k, as L2 is now longer than 325mm.

Checking your result against what I have for a Savarez 500AR set (Corum basses) I have k=5760 N for the low E, and that's the lowest result I have for a low E. Most strings are quite a lot stiffer. You may need to let the string stretch out for longer.

[kiwigeo]

Thanks alot Trevor.

The string was on the jig under tension for 72 hours prior to testing.

Re L2....I suspected the figure wasn't quite right.

String diameter should be 1.11mm......High E figure got into the spreadsheet somehow. I'm assuming that the outside diameter of the wound basses is what you're using...it's a bit hard to micrometer the core as it consists of filaments.

Cheers Martin

[Trevor Gore]

I'm assuming that the outside diameter of the wound basses is what you're using...it's a bit hard to micrometer the core as it consists of filaments.

All I need for my intonation calcs is k and the tension at pitch. Usually, any reference to the diameter for "nylon" strings is just part of the string identification.

[kiwigeo]

Thanks Trevor

[kiwigeo]

Is there any great advantage in applying greater than specified tension to the string while it's stretching on the jig and then reducing tension to correct value prior to clamping the string?

[Trevor Gore]

I don't know any good shortcuts, Martin. A lot of "nylon" string materials have a memory effect, for want of a better phrase. If you tune a string high on a guitar (say the top E to F) leave it a while then tune down to E again, after a few minutes it will be sharp. The best I've come up with is to leave it loaded with weights at close to the correct pitch. As the string stretches out, the load remains constant (unlike with a constant displacement device like a tuning machine) and the pitch hardly changes. After a while, the rate of elongation slows. 3-5 days is usually enough, which means it seems to take forever to collect the data for a full set of nylons. The strings never stop stretching, as you can see on the wound strings on a guitar. The marks left by the frets just keep on moving towards the nut. I suppose this is good, because it means the windings don't wear through on the frets as the contact point keeps moving, but it also means that the string properties are still changing. I usually leave "nylon" strings a minimum of 3 days, then take the data and load another string when I get a chance between other jobs. Steel strings are easy. I can usually do a whole set in under an hour because they always remain in the elastic zone.

[kiwigeo]

Okay thanks Trevor..

[kiwigeo]

I put the low E string back on the jig for a few more days and re-tested. Results in table below.

[Trevor Gore]

That looks a lot closer to what I've been getting for k.

[kiwigeo]

That looks a lot closer to what I've been getting for k.

I should have the whole string set done by Christmas

[kiwigeo]

An update to an old thread.

The string stiffness jig is set up so I thought I'd re-do a set of Savarez Corum Strings. Strings were on the jig for 4 days prior to taking the readings used in calculations.

Results for the E6th string:

Scale Length = 650.0mm
Tuner to nut = 51mm

T = 7.7534Kg/76.061N to bring string to pitch
Mass/unit length = 6.629 x 10^-3 Kg/m
k = 3673.69 N (got 2633.08 N testing previous string set). Maybe a bit low and string needs longer on the jig??

Plugging the data into my spreadsheet I arrive at saddle compensation of 0.6mm (towards tail) and nut compensation of 3.0mm (towards saddle).

The saddle compensation looks a bit low. Nut relief is 0.1m and relief at 7th fret is 0.2mm. Action at 12th fret is 4.0mm

If some kind person could input my data and see if they get a similar saddle compensation value it would be greatly appreciated. Additional data as follows:

Fp= 0.75
Go = 0.5
alpha = 0.5
Beta = 1.14

[kiwigeo]

More sensible results after leaving the string on the jig for a few more days. Total of 7 days on the rig....

k value now 8901N and saddle compensation value works out at around 1.5mm.

[kiwigeo]

A question for Trevor, Jeff or the other gurus.

When I run a string on the testing jig I get as close as practically possible to the correct tension to get the string to target pitch. Once Ive got a steady mass/unit length and stiffness value I plug the mass/unit length value into Mersennes equation to derive the correct tension to achieve the target frequency. This tension value then gets plugged into the spreadsheet to calculate compensation.

Is the method above going to introduce sigificant error into the compensation calculations? Should I be busting my nuts and getting the tension on the test rig to exactly what is required to achieve target open string frequency??

[Trevor Gore]

No and no.

[kiwigeo]

No and no.

Next question...are two Nos equal to one NO!

[Nick]

No and no.

Next question...are two Nos equal to one NO!

One for each nut.

[kiwigeo]

No and no.

Next question...are two Nos equal to one NO!

One for each nut.

630x420_KickMeInTheNuts.jpg (160.17 KiB) Viewed 24484 times

[kiwigeo]

One more string to go and the testing on the Savarez Corum set is done....its taking around 5 days for the strings to settle down.