how to make a device to measure string tension?

Choice of classical guitar strings and technical issues connected with their use.
User avatar
rojarosguitar
Posts: 4792
Joined: Sat Sep 19, 2009 12:24 pm
Location: near Freiburg, Germany

Re: how to make a device to measure string tension?

Post by rojarosguitar » Wed Apr 30, 2014 8:47 am

Trevor Gore wrote:The "tool" is called a monochord. Pythagoras is reputed to have used one. A board, string fixed at one end, two moveable bridges, weights (adjustable) hanging off the string at the other end, usually after it's gone over a pulley. Frequency measurement using one of any number of free frequency measurement devices. If you're serious, make sure you can calibrate everything.

Google images for monochord to get a few ideas of the implementation. How I made and use mine is, of course, documented where you'd expect, including a neat way of measuring string longitudinal stiffness. Very useful for calculating compensation requirements.
This is one way of doing it, but it works from the assumption that the tension goes linear with the pitch, which might not be quite true for strings aimed at the end edge of Hookes' Law.
That's why I'd rather prefer to measure at the correct pitch - which you of course can achieve with weights as well, if you have enough different weight pieces.
Music is a big continent with different landscapes and corners. Some of them I do visit frequently, some from time to time and some I know from hearsay only ...
My Youtube Channel is: TheMusicalEvents

User avatar
Trevor Gore
Posts: 995
Joined: Sun Aug 07, 2011 12:57 pm

Re: how to make a device to measure string tension?

Post by Trevor Gore » Wed Apr 30, 2014 9:45 am

rojarosguitar wrote:This is one way of doing it, but it works from the assumption that the tension goes linear with the pitch...
Mersenne's law holds throughout. The reason I use the method I use is that I can hang weights on a string and leave it there for days, at constant tension, over which time it eventually settles to a (reasonably) constant pitch. What changes is the mass per unit length as the string stretches, but Mersenne's law always holds. If you are into the fully plastic region of a string's elongation characteristic all that happens is that the string keeps on stretching, its diameter reduces, the stress in it increases and it breaks. If you have the string tensioned with a tuning machine, and the string stretches plastically, the tension reduces and the pitch drops so tuning stability is never attained, which, of course, is fine if you want to use strings that won't stay in tune...

If one goes for an "instantaneous" tension/frequency reading, that's fine also, but it does not represent a service condition for an instrument that needs to stay in tune for at least the duration of a piece.
rojarosguitar wrote:That's why I'd rather prefer to measure at the correct pitch - which you of course can achieve with weights as well, if you have enough different weight pieces.
If one has an accurate set of scales, one always has enough different weights... :D
Trevor Gore: Classical Guitar Design and Build

User avatar
rojarosguitar
Posts: 4792
Joined: Sat Sep 19, 2009 12:24 pm
Location: near Freiburg, Germany

Re: how to make a device to measure string tension?

Post by rojarosguitar » Wed Apr 30, 2014 12:09 pm

Trevor, I wouldn't go for 'instantaneous' measurement, that's why I would leave the string on until it settles. If you look carefully at Mersenne's Law, it is more tricky than it lookes like, because this mass per unit length value is itself depending on the pulling force, and it will change in a nonlinear way. (So strictly speaking in and by itself it's not very useful unless you have determined the stable value of mass per unit length at the exact conditions you want to operate the string at. What it just states is what happens to frequency if one factor changes and all the other factors are kept constant - but only the string length remains constant to good accuracy all the time, which makes a guitar workable in first place).

Fortunately for all string instrument players this non-linearity is there and comes to help to let the tuning stabilize somehow. So if you want to know the actual tension of the string you really have to let the string settle at the required pitch and wait for a stable value of the pulling force.

This you can achieve with the method I have described, or if you prefer, adding little weights until the string stabilizes at the required pitch. Outcome same, but I suspect tuning peg is more practical (and your weight method potentially more accurate - if you really have enough different weights to achieve exactly the tuning :D)
Music is a big continent with different landscapes and corners. Some of them I do visit frequently, some from time to time and some I know from hearsay only ...
My Youtube Channel is: TheMusicalEvents

es335
Posts: 1524
Joined: Thu Aug 30, 2007 2:12 pm
Location: Germany

Re: how to make a device to measure string tension?

Post by es335 » Wed Apr 30, 2014 12:37 pm

From my understanding of plastics the strings are not operated in their plastic phase.
The stretching we observe during their settling is just a longitudinal reorganisation of the makromolecules. Some kind of relaxation process. It will just marginal reduce the diameter and has very little to no impact on the unit weight as the density goes up accordingly. This being said under the provision that the string's load is well below their yield strength!

Anyway from a pure logical perspective I support to measure the string's tension at the pitch they are intended for. Why not, when considering a new and standardized setup!?
Prominent Critic wrote:Perhaps D'Addario and the other string companies are already using some analogous device. It's certainly possible. It's also possible and probable that they're using some theoretical formula to calculate what should be the tension given all the factors they input. Since that method of calculation sends rockets to the moon, I don't see why it shouldn't be used - and successfully - for something as relatively simple as measuring the tension of a guitar string. The problem is to get all the string manufacturers to adhere to a uniform standard for the calculation.

Ramon
D'Addario has done that just recently by changing their tension values. The new values are now in full compliance with Mersenne's law with a density of 989 g/dm³ for ALL (!!!) treble strings.

Opposed to this the backward calculation of the old tensions results in different densitiy figures for every treble string which is not reasonable: 933 (e1), 952 (b2) and 1008 (g3)

Other brands are even worse: La Bella 2001 MT: 876 - 994, Hannabach 815 LT: 954 - 1169

So refering to the other thread which originated this one here, the answer is, nothing changed with the good old EJ45 strings except the calculation and most presumably the subjective tension perception influenced by different figures! :D :wink:

PS: BTW there are automatic tuners around which seem to operate very satisfactory. I suppose they would be perfect to bring and keep the strings automatically in tune! The keyword for google is tronical!

So the test bench could be something like a monochord with load cells or spring balances on one side and tronical tuners on the other side! :D

User avatar
Trevor Gore
Posts: 995
Joined: Sun Aug 07, 2011 12:57 pm

Re: how to make a device to measure string tension?

Post by Trevor Gore » Wed Apr 30, 2014 1:44 pm

es335 wrote:The stretching we observe during their settling is just a longitudinal reorganisation of the makromolecules. Some kind of relaxation process. It will just marginal reduce the diameter and has very little to no impact on the unit weight as the density goes up accordingly.
If the string gets, say, 10% longer, its unit mass (mass per unit length) has just reduced by 10% (by conservation of mass) whatever the visco-elastic mechanism might be.
rojarosguitar wrote:So if you want to know the actual tension of the string you really have to let the string settle at the required pitch and wait for a stable value of the pulling force.
Agreed. However, if one is testing, say, a 1st string and has a stable tension and frequency at, for example, 330Hz (rather than E at 329.638 Hz ) I don't think making a "Mersennian" correction for tension will introduce too much error due to non linearities. Normal production variation from string to string yields larger variations.
Trevor Gore: Classical Guitar Design and Build

es335
Posts: 1524
Joined: Thu Aug 30, 2007 2:12 pm
Location: Germany

Re: how to make a device to measure string tension?

Post by es335 » Fri May 02, 2014 7:11 am

Trevor Gore wrote:... If the string gets, say, 10% longer, its unit mass (mass per unit length) has just reduced by 10% (by conservation of mass) whatever the visco-elastic mechanism might be...
Fully agreed but at least d'Addario's latest tension chart revision seems to confirm the approach of constant unit mass respectively negligible unit mass variation when the string is tuned to pitch.

Statement of d'Addario taken from the other thread:
Prominent Critic wrote:Secondly, we remodeled the string Unit Weight calculation itself to allow for a more geometrically accurate number. While some tensions may seem to have changed drastically, (+/- lbs. in some cases) these changes have yielded a much more precise representation of our strings and the tension they yield when tuned to pitch, as well as a much more precise representation than the previous published tensions.
As already stated above that's nothing else than a flowery description of that they changed from a non-uniform to a uniform densitiy approach for the treble strings for instance! The only missing link is to see reliable test result confirming this calculation approach. :wink:

User avatar
Trevor Gore
Posts: 995
Joined: Sun Aug 07, 2011 12:57 pm

Re: how to make a device to measure string tension?

Post by Trevor Gore » Fri May 02, 2014 9:50 am

I had a pretty strong go at D'Addarrio about 6 months ago, when trying to make sense of their string data and finding that the charts supplied off their web site and the info on the packets for nominally the same strings was quite different. I was dealing with steel strings at the time. The "technical" guy I was swapping emails with did not seem to have much idea about the physics of things. I think he was basically a customer service rep. Anyway, sufficient people must have been asking questions for D'Addario to sort things out. When I read this:
Prominent Critic wrote:Secondly, we remodeled the string Unit Weight calculation itself to allow for a more geometrically accurate number. While some tensions may seem to have changed drastically, (+/- lbs. in some cases) these changes have yielded a much more precise representation of our strings and the tension they yield when tuned to pitch, as well as a much more precise representation than the previous published tensions.
my assumption was that D'Addario were making a better geometric approximation to the shape of their "hexagonal" steel cored strings. If they were making calculations on the basis on them being accurately round or accurately hexagonal they could get some significant errors. The actual shape is a rather rounded cornered hexagon.

I've no idea whether this was a correct assumption or not!
Trevor Gore: Classical Guitar Design and Build

es335
Posts: 1524
Joined: Thu Aug 30, 2007 2:12 pm
Location: Germany

Re: how to make a device to measure string tension?

Post by es335 » Fri May 02, 2014 11:00 am

I can't comment on their steel strings but it's quite obvious what they have done to their Nylon strings with special attention to the Pro Arte trebles.

The old tension values were based on different unit mass figures for each treble string, which really makes no sense, because it's difficult to understand why strings from the same material should have different densities just because their size is different?! :?

Their new tension values are now based on a uniform unit mass for all treble strings, which is what everybody would expect from a physics perspective.

Regardless of that, there are still some "mistakes" in their tension charts which reflects that it is still considered a chore task to provide this service. :(

cato33

Re: how to make a device to measure string tension?

Post by cato33 » Sun May 25, 2014 6:32 pm

I hope this bump isn't too big. Maybe some type of spoke tension meter would work or similar type of device? It measures tension by the amount of deflection against a constant pressure. A few more variables are needed for accurate results. Whether string is wound/unwound, width of string, and material. There are more precise variables as types of winding or special material, but results should be reasonably accurate.

http://ep1.pinkbike.org/p4pb7842781/p4pb7842781.jpg

User avatar
rojarosguitar
Posts: 4792
Joined: Sat Sep 19, 2009 12:24 pm
Location: near Freiburg, Germany

Re: how to make a device to measure string tension?

Post by rojarosguitar » Mon May 26, 2014 8:36 am

The idea is nice, though physically it (this device) rather measures some combination of stiffness and tension. Also when I think about orders of magnitude, there should be quite a big difference between guitar strings and bike spokes. So most likely you will have to work in the lowest margin of the range of this meter (I guess).
Music is a big continent with different landscapes and corners. Some of them I do visit frequently, some from time to time and some I know from hearsay only ...
My Youtube Channel is: TheMusicalEvents

User avatar
Trevor Gore
Posts: 995
Joined: Sun Aug 07, 2011 12:57 pm

Re: how to make a device to measure string tension?

Post by Trevor Gore » Mon May 26, 2014 11:11 am

cato33 wrote:Maybe some type of spoke tension meter would work or similar type of device?
There are similar devices used in the yacht racing world for measuring rig tension (tension in the wire stays). Having used them, they're accurate enough for that application but would have to be finely engineered to measure string tensions as accurately as I would want.
Trevor Gore: Classical Guitar Design and Build

es335
Posts: 1524
Joined: Thu Aug 30, 2007 2:12 pm
Location: Germany

Re: how to make a device to measure string tension?

Post by es335 » Mon May 26, 2014 1:48 pm

Trevor Gore wrote:
cato33 wrote:Maybe some type of spoke tension meter would work or similar type of device?
There are similar devices used in the yacht racing world for measuring rig tension (tension in the wire stays). Having used them, they're accurate enough for that application but would have to be finely engineered to measure string tensions as accurately as I would want.
Tension meters are available in an appropriate range and with sufficent accuracy. Just the standard big size guide rollers should most probably replaced by smaller ones. :wink:

PAustin
Posts: 156
Joined: Sat Feb 22, 2014 3:25 pm

Re: how to make a device to measure string tension?

Post by PAustin » Mon May 26, 2014 2:41 pm

Ahhhh. Rigging a bi-plane - now you're talking! :) flying is my other obsession. Anyway, you could use a small fish scale as well. A very old aviation mechanic taught me that when we needed to check some cable tension. It works in a pinch. Don't know why it wouldn't work for this too.

Paul


.
Paul Austin

2009 Pavan TP-30 Spruce
1980's ? Yamaha G-231 II

User avatar
rojarosguitar
Posts: 4792
Joined: Sat Sep 19, 2009 12:24 pm
Location: near Freiburg, Germany

Re: how to make a device to measure string tension?

Post by rojarosguitar » Tue May 27, 2014 8:09 am

PAustin wrote:... Anyway, you could use a small fish scale as well. A very old aviation mechanic taught me that when we needed to check some cable tension. It works in a pinch. Don't know why it wouldn't work for this too.

Paul


.
That's exactly what I suggested, with an electronic luggage scale that has the advantage of showing negative values...
Music is a big continent with different landscapes and corners. Some of them I do visit frequently, some from time to time and some I know from hearsay only ...
My Youtube Channel is: TheMusicalEvents

Return to “Classical Guitar Strings”