Curt's Experiment

b0b · Post by **b0b** » 8 Jul 2006 7:58 am

Curt Langston wrote:
-----
Joe
Your sketch gave me an idea. So, I thought what that heck, I'll try a little experiment. I had a few 0.011 gauge strings laying around, so I put them to good use.
Here is a crude sketch of what I tried. Now mind you, I do not have access to CAD, so this is a simple free hand for a visual. Not to scale! (pun intended)

I used a wooden drumstick for my keyhead to do the pulls. Jagwire strings were used for all pulls.
Pulls were applied with a slow take off as to not jerk the string and create extra tension. I drilled a tiny hole for the string to feed into. The strings were wrapped to make the different lengths. I did not overlap the strings, so there would be no sharp bends. Also the pulls were done in the barbell curling fashion.(I wore a long sleeve flannel shirt!)
The weights were held together by 14 guage utility wire. The strings were connected to the utility wire with standard electric fence wire, through the ball end.

Pull (C) was done first, using 35 lbs. and 28 1/2 TSL to simulate a keyed 24 in scale.
1st break was at the 14th pull. Broke at the ball end. 2nd break was at the 12th pull, at the drumstick end. 3rd break was at the 15th pull. Broke at the ball end again.

Pull (B) was next with a TSL of 26 inches, to simulate a keyless 25 inch scale, but with 5 pounds added. After 25 pulls, I quit. Not breakage.

Rested.

Pull (A) proved to be the same. A TSL of 25 inches was used to simulate a keyless 24 inch scale. I lasted another 25 pulls. Barely. No breakage.

Rested!.......Rested some more.

BTW, Eric. I got your message about my beautiful mind. That was a good movie.

[This message was edited by Curt Langston on 07 July 2006 at 08:15 PM.]

b0b · Post by **b0b** » 8 Jul 2006 8:02 am

It was a response to Joe Meditz, who had written:
-----
Below are two ASCII drawings of two systems of equal scale length. The string hangs over a roller o. The x anchors it. The weight provides the tension. The first system has more overhang than the second. Neglecting friction it is axiomatic that the tension in the string is the same in both cases since the weight is the same.
<pre> o----------x
|
|
|
|
|
|
|
|
|
_
/ \
---

o----------x
|
|
_
/ \
--- </pre>
Joe

ed packard · Post by **ed packard** » 8 Jul 2006 8:10 am

Whoops...missed the move!

Curt...nice descriptive sketches.
What I get from your experiment is that the strings tend to break at the ends.

The drumstick end probably because of the rolling motion I envision happening when you use the weightlifting curl motion...simulates to some degree what happens at the changer finger.

The weight end may simulate what happens if the ball/wrap end is not wrapped around a changer finger. The ball end and wrap was a problem in the past. It would tend to unwrap, and also to break where the wrap comes back to the string wire. You did not describe the happening at that end in enough detail to decide. This could be affected by rotation of the weight(s) = twist/torque forces on the string.

Earnest Bovine · Post by **Earnest Bovine** » 8 Jul 2006 8:13 am

Here is a simple way to demonstrate that string tension doesn't depend in any way on the amount of string on the left side of the roller nut. It requires no special tools, but you will use some strings.

Loosen all the strings except #1 and #6
Put 2 new identical gauge strings on #1 and #6
Tune them to the same pitch
Compare the tension by balancing something across the strings. If the tension is different, one string will sag more than the other.

b0b · Post by **b0b** » 8 Jul 2006 8:20 am

I moved this topic out of Ed Packard's topic about frequency response experiments.

Curt wrote: "Pull (C) was done first, using 35 lbs. and 28 1/2 TSL to simulate a keyed 24 in scale."

That's not what it simulated. There's no 24" scale in the experiment.

In Curt's experiment, total string length (TSL) and scale length are the same thing because there's no nut. There was also no pitch measurement, so we really don't know what note the string was sounding.

He seems to have proven that a longer scale length (28.5") will break strings at a lower tension. If true, that would be bad news for anyone using a 28.5" scale. Does anyone besides Ed Packard have such a beast?

------------------
<img align=left src="http://b0b.com/b0bxicon.gif" border="0"> Bobby Lee
-b0b- quasar@b0b.com 
System Administrator 
My Blog

b0b · Post by **b0b** » 8 Jul 2006 9:31 am

Here's a good string tension calculator.

Curt Langston · Post by **Curt Langston** » 8 Jul 2006 9:46 am

<BLOCKQUOTE>quote:<HR>In Curt's experiment, total string length (TSL) and scale length are the same thing because there's no nut. There was also no pitch measurement, so we really don't know what note the string was sounding.

He seems to have proven that a longer scale length (28.5") will break strings at a lower tension. If true, that would be bad news for anyone using a 28.5" scale. Does anyone besides Ed Packard have such a beast?

<HR></BLOCKQUOTE>

b0b:

I used a 28 1/2 inch length of string to simulate the TSL on a 24 1/4 inch scale.(since a 24 1/4 has a TSL of 28 1/2)
I did not explain this, as I thought it to be common sense. I should have went into more detail. Of course no one has a 28 1/2 inch scale. Also, the other two lengths 25" and 26", were to be understood that they were representing a 24 inch keyless scale, and a 25 inch keyless scale respectively.(allowing 1 inch over hang)

This should be apparent by the writing on the string in each example.

Drawing Key: 28 1/2 TSL= 24 1/4" KEYED guitar.
26 TSL= 25" KEYLESS guitar.
25 TSL= 24" KEYLESS guitar.

This was just a little experiment I did, actually more for myself than others. I just posted it because I have been told to "Do the experiments" or "Run the numbers"

This was the easiest way for me. (and cheapest)

All the pulls were done in the same manner. (same speed, stroke length and such) I realize that this is not as scientific as someone like Ed or the other engineers could do, but it proved to ME the results I suspected all along, that TSL does factor in string breakage. Longer TSL's will not hold as much tension as shorter ones. That is why I say that a keyed 24 1/4 inch scale guitar, (with a TSL of 28 1/2") will break the 0.011 quicker than a shorter TSL guitar. (keyless 24 or 25)

<BLOCKQUOTE>quote:<HR>There was also no pitch measurement, so we really don't know what note the string was sounding.
<HR></BLOCKQUOTE>

There wasn't supposed to be.

I was not concerned with pitch. I was concerned about breaking points on strings, at different TSL's.

<BLOCKQUOTE>quote:<HR>The weight end may simulate what happens if the ball/wrap end is not wrapped around a changer finger. The ball end and wrap was a problem in the past. It would tend to unwrap, and also to break where the wrap comes back to the string wire. You did not describe the happening at that end in enough detail to decide. This could be affected by rotation of the weight(s) = twist/torque forces on the string.
<HR></BLOCKQUOTE>

Good question Ed. Sorry I left that out. I did think of that beforehand. The two breaks at the ball end were about 7-8 mm above the wrapping. Not quite a centimeter. More like 3/4 of a centimeter.

I was careful to not allow the barbell weights to rotate. I thought if they did, it would creat more tension. On each downstroke, I allowed the weight to LIGHTLY land on the carpet, as to not promote the twisting of the string. When the barbells would turn, it would only be 1/5 of a turn, in which I turned my body to correct it when the barbells rested on the floor.

I tried to make each pull uniform, smooth and easy. By the time I got halfway through the 25" TSL, I was becoming fatigued, and the movements were not as smooth. Yet the 25" TSL withstood the great poundage in spite of my having to jerk up more on the drumstick. By the 20 repetition on the 25" TSL, I was actually just trying to get to the 25th pull, and was sloppily jerking up on the drumstick. Yet no breakage. [This message was edited by Curt Langston on 08 July 2006 at 03:06 PM.]

Eric West · Post by **Eric West** » 8 Jul 2006 9:59 am

Good.

Now maybe we can move on to how much more tension it takes to get a string up to pitch on two exact scale lengths, One with a longer "overhang", and one with none..

One thing at a time.

EJL

David Doggett · Post by **David Doggett** » 8 Jul 2006 10:15 am

Earnest, your test wont work. Technically, tension is the force at the string end. The force required to deflect the string sideways includes tension, but also is affected by the amount of stretch in the string, both over the scale, and beyond the scale. The longer the length is beyond the scale, the more the string has to be pulled sideways to cause the same pitch change, because that extra length adds extra stretch. So the strings with different overhangs beyond the scale will feel different when pressed or pulled sideways, but the tension at the end will be the same, if the scale length, gauge and pitch are the same. We could call the sideways force the deflectivity to differentiate it from tension (at the string end).

If you are not pulling sideways, but lengthways, the way a changer finger pulls, the string with more overhang will require a longer pull to get to pitch, even though the tensions before and after the pull are the same. On a typical changer that rolls the string around the top of the changer, the longer overhang guitar will have a longer finger throw, will wrap more string around, will have more bending heat and stress, and will probably break sooner. It's not because of any tension difference, but because of more stretch, and more bending at the changer. Curt keeps thinking the more breakage proves more tension. It doesn't. There is another more plausible explanation - more string bending at the changer.

As b0b points out, Curt keeps confusing scale length (SL) and total string length (TSL). Joe's drawing is the simplest depiction of the situation. The horizontal line is the scale length, it is the same in the two figures. Those two scales will play the same note, because they are the same length and have the same tension. The vertical line is the overhang. It doesn't matter how long the ovehang is; if you put a 30 lb. weight at the end of each overhang, by definition there will be 30 lb. of tension everwhere in the overhang (in spite of their different lengths), and everywhere in the scale, and at both ends of the strings, in spite of their different total lengths. That is the definition of "tension," the amount of force at the end of the string.

What Curt and Charlie are talking about is something else. It is something that has a fixed value per inch of string, and so sums to a greater value for a longer string. Tension does not do that. It is not fixed per inch of string, but by definition is fixed at the end of the string, regardless of length. By definition it does not change with string length. 30 lb. of tension at the end of a 24" string is the same as 30 lb. of tension at the end of a 28" string.

Maybe Curt and Charlie are talking about stretch. That is fixed per inch of string, and sums to a greater value for a longer string. They just need to get their terminology straight. Whatever they are thinking about is not called tension technically. By calling it tension, they are confusing the issue.

I am starting a new thread with a descriptive title, so it will be searchable. I'll call it "Longer string, same tension." It will summarize the relevant physics and will explain the problems with the various quotes used to support the myth that greater TSL causes greater tension.

------------------
Student of the Steel: Zum uni, Fender tube amps, squareneck and roundneck resos, tenor sax, keyboards

Earnest Bovine · Post by **Earnest Bovine** » 8 Jul 2006 10:32 am

oops, you are right; the balance method will not compare the 2 tensions accurately. (But I will leave my stupid idea up there anyway.) I guess that using hanging weights is the simplest way to do the comparison.

Curt Langston · Post by **Curt Langston** » 8 Jul 2006 11:04 am

As b0b points out, Curt keeps confusing scale length (SL) and total string length (TSL).

No I don't. What about a situation where the scale length and the TSL are ALMOST equal, say, Ed's Beast. Can he pull a 0.011 G# up to an A?
Not hardly. Thats 31+ inches. A keyed 24 1/4 guitar is around 28 1/2. Each inch will add about 2 pounds tension. Keyed or keyless. On a 24 1/2 keyed guitar, your G# pulled to an A is very close to it's breaking point. Thats why it will wear out sooner. You can pull 25 or 26 inches of string (for a 24 and 25 inch keyless scale) easier and with less tension than 28 1/2 inches of string.(for a keyed 24 1/4 guitar)

David Doggett · Post by **David Doggett** » 8 Jul 2006 11:31 am

Curt, you are wrong again. Ed’s numbers show that there is an extra 2 lbs. of tension per inch of SCALE LENGTH. Each inch of total string length adds no more tension. Ed’s beast would have a hard time pulling to A without breaking because his SCALE length is 30”. If he keeps his total string length THE SAME, but moves his nut to shorten his scale length to 24”, he will have lowered the tension required to pull to A, and will be able to do so just fine, even with 31” of total string length. However, the 7 inches of overhang does introduce more stretch to reach A than a 1” overhang would. That extra stretch requires a longer changer finger throw, and more string wrapped over the top of the changer. That causes more bending, stress and heat. And that could cause the string to break sooner. It’s not the tension, it is the extra stretch and bending.[This message was edited by David Doggett on 08 July 2006 at 12:35 PM.]

Curt Langston · Post by **Curt Langston** » 8 Jul 2006 12:09 pm

Each inch of total string length adds no more tension.

Not so David. Remember, we have already established that the scale length and the overhang have the same tension.

However, the 7 inches of overhang does introduce more stretch to reach A than a 1” overhang would.

RIGHT! And what is needed for more stretch?
Answer: More TENSION.

If he keeps his total string length THE SAME, but moves his nut to shorten his scale length to 24”, he will have lowered the tension required to pull to A, and will be able to do so just fine, even with 31” of total string length.

Wrong....Imagine this: Ed's Beast has a scale of 30 inches. His total length is 31 inches. NOW, if he could somehow separate the anchor points on the nut, from the actual nut, and have that nut on sliding rails that would allow for movement to ANY point on the scale, he could shorten his scale, WITHOUT lessening the tension on the anchor points.(total length)

Or this scenario: Tune his guitar to E9th. (without using his pedals, he probably could)
Then place a bar underneath, halfway on the 12th fret or so. NOW raise that G# up to an A.... POP!

He did not change the tension by placing the bar underneath at the 12 fret.(if he did change the tension, it would be a slight INCREASE due to the height of the bar) So tension on both sides of the bar would be equal.

Curt Langston · Post by **Curt Langston** » 8 Jul 2006 12:11 pm

Is it just me, or do I sense a lot of tension in here?

David Doggett · Post by **David Doggett** » 8 Jul 2006 12:12 pm

Curt, you need to relax your defenses a little and think carefully about your terminology. I admit that originally I was half wrong, and you were half right. I believe you are right that shorter overhang and shorter total string length will result in less string breakage on a pedal steel. That is because there is less stretching in the overhang, and therefore, less bending at the changer. But you are half wrong and I am half right, because the reason for the less string breakage is not less tension. The tension only changes with the scale length, not the overhang or total string length. If you shorten the scale length and hold pitch and gauge the same, tension decreases, regardless of overhang and total string length. If you lengthen the scale length and hold pitch and gauge the same, tension increases , regardless of overhang and total string length. Your terminology is tripping you up. Get your terminology in agreement with the rest of us, and you will see where we agree. Tension refers to force at the string end. If moving the string to a further post increased the tension at the string end, it would also increase it over the scale, and so would change the pitch. But if the pitch stays the same (A), then the tension has not changed over the scale, and therefore cannot have changed at the string end either. Think about that carefully please.

Tony Prior · Post by **Tony Prior** » 8 Jul 2006 2:07 pm

I'm glad you guys are doing this so I don't have to.

I am practicing Gary Stewart tunes aka: Steve Palousek

Someone has to actually practice the dang thing instead of just talk about it

Actually, I'm gonna re-phrase my first sentance:

I don't really care about all that tension vs how many pulls before it breaks stuff..

It's a 50 cent string on a $4000 Guitar

t[This message was edited by Tony Prior on 08 July 2006 at 03:09 PM.]

Curt Langston · Post by **Curt Langston** » 8 Jul 2006 2:11 pm

Wish I were practicing instead of being at work!

Jim Peters · Post by **Jim Peters** » 8 Jul 2006 2:48 pm

Curt, reconsider your position, please. Try to follow this logic:
Take any TSL(say 30") and tune it to A440, it will have a certain tension,agreed? Now put a nut on it, make a scale length of whatever, let's say 25". Whatever that note becomes, it will have the same tension as the original TSL, we've changed nothing else. With me? Let's measure the note on a tuner, let's just say it is an E.
Now take that nut, make the scale length longer, let's just say 27". Remember, we've changed nothing else. We measure the note, now it is lower than the E, for the sake of arguement, let's just say it is now a D. The tension is still the same as when we started, agreed? We've changed only the scale length, nothing else.
Now to make that longer scale an E, what would we do? Tighten it, right? We add tension. The longer scale needs more tension to achieve the same note. Still with me?
It is a scientific fact that any 25" scale ever made will require the same tension to get an E, whether the TSL is 25" or 25', or 1 mile. (string gauge the same)
The tension is increased because of the longer scale length, no other reason.
Whatever the reasons for a longer guitar breaking strings(if they do), it is not tension caused by a longer TSL, the tension for the same scale length is the same for a given note, regardless of TSL.
Whatever part of this you don't follow, ask questions, it does become clearer as you think about it. JP

Steve Hinson · Post by **Steve Hinson** » 9 Jul 2006 7:46 am

..you fellows certainly have a lot of leisure time...

------------------
http://home.comcast.net/~steves_garage

Bobby Lee · Post by **Bobby Lee** » 9 Jul 2006 11:27 am

Longer scale length requires more tension to bring a string up to pitch.

Longer total string length has no relation to the amount of tension required to bring a string up to pitch.

Longer total string length may, however, increase string breakage on a pedal steel because of increased changer travel.

The tension required to tune a .011 string to A is less than 35 lbs. on all popular steel guitar scale lengths.

Did I get everything right?

------------------
<img align=right src="http://b0b.com/b0b2005.gif" width="78 height="78">Bobby Lee (a.k.a. b0b) - email: quasar@b0b.com - gigs - CDs, Open Hearts
Williams D-12 E9, C6add9, Sierra Olympic S-12 (F Diatonic)
Sierra Laptop S-8 (E6add9), Fender Stringmaster D-8 (E13, C6 or A6) My Blog [This message was edited by Bobby Lee on 09 July 2006 at 12:30 PM.]

Curt Langston · Post by **Curt Langston** » 9 Jul 2006 11:33 am

Did I get everything right?

Depends on which opinion you subscribe to.

<BLOCKQUOTE>quote:<HR>Longer scale length requires more tension to bring a string up to pitch.
<HR></BLOCKQUOTE>

Agreed.

<BLOCKQUOTE>quote:<HR>Longer total string length has no relation to the amount of tension required to bring a string up to pitch.
<HR></BLOCKQUOTE>

Now, think about that for a moment. Ed's Beast has a scale that is almost equal to his TSL.(almost... what, about 3/4 inch overhang on each end?) His guitar cannot pull a 0.011 G# tuned string up to an A. If he were to shorten up his scale,(and subsequently his TSL, since they are about equal) to 26 inches(thus making his scale 25 inches) he could do it.
A keyed 24 1/4 inch scale guitar has about 28 1/2 inches of TSL. A keyless 24 or 25 inch scale guitar has about 25 or 26 inches of TSL respectively. Those 4 or 5 inches of less string length, reduce the breaking point of that 0.011 string greatly. Remember that 1 inch of string amounts to right at 2 pounds of tension. And we have all agreed that that overhang part of the string is under the SAME tension as the scale.[This message was edited by Curt Langston on 09 July 2006 at 01:32 PM.]

Curt Langston · Post by **Curt Langston** » 9 Jul 2006 12:31 pm

b0b,

There are a lot of keyless players that realize their guitars do not break strings as often as the keyed ones do.
They just can't explain exactly why.

I have just explained it above.

If that makes me unpopular, then so be it.

"Being right does not always mean being popular, but then again, being popular does not always mean being right"

CL

David Doggett · Post by **David Doggett** » 9 Jul 2006 1:04 pm

b0b, I agree with everything you said.

Curt, you are getting very close. What you seem to have trouble with is that the extra travel required to bring the longer TSL up to A does not require more tension than required to bring the shorter TSL up to A. More stretch, and more travel, yes. If two guitars have the same scale length and string gauge (or if you take one guitar and attach two of the same gauge string, one to the first post and the other to the 5th or 6th post), and one has 1” overhang, and the other has 6” overhang, because the scale and gauge are the same, the two starting G# notes will have the same tension on both guitars (or both strings), and the final A notes will have the same tension. But it takes more travel to bring the string up to the A tension on the guitar (or string) with the longer overhang. That extra travel at the changer may well cause more string breakage. And a very short overhang, such as on a keyless, may improve string breakage to the point that one can add an extra ¾” to the scale (say from 24 ¼” to 25”) without getting more string breakage, even though the tension on the 25” scale will be a little higher. In other words you may be able to use the savings in string breakage with the shorter overhang to increase the scale and the tension. This may be what has happened with 25” scale keyless guitars. They can use the longer scale, which actually has more tension (Ed measured it), but have the same breakage, or maybe even less. So it is all because of less stretch and changer travel. But stretch and travel are not the same thing as tension. This is simply the technical definition of tension. It is a property measured statically (without motion) at the string end. It does not measure stretch, travel or total string length. You measure it with the string at rest for the G#, and you measure it again after the pull motion has stopped at the A.

As for Ed’s guitar, he could keep his 31” TSL, but move his nut to shorten the scale to 24”, thus lowering the tension required to get an A, and he could get both G# and A, the same as any other 24” scale guitar. With 7” of overhang, there would be a lot of stretch and travel to raise to that A, even though there is less tension required than with the 30” scale. So he might get more string breakage than if he reduced both the scale length and overhang length and ended up with a 24” scale and 1” overhang and TSL of 25”.

To your credit, by forcing us to think this through so many different ways, we have a plausible explanation (less changer travel) for your claim that keyless guitars have less string breakage (even with longer scales and more tension), and why maybe that BMI with the 3rd string on the 4th key-post position might have more string breakage (more changer travel). The only thing we object to is assigning the alleged string breakage to tension, when the known laws of physics say that overhang distance does not affect the technical definition of tension. Please think about the technical definition of tension, and lets move on.
[This message was edited by David Doggett on 09 July 2006 at 02:08 PM.]

Jim Peters · Post by **Jim Peters** » 9 Jul 2006 1:21 pm

"I have just explained it above."
No, you haven't!
" If he were to shorten up his scale,(and subsequently his TSL, since they are about equal)"
That's what "we" want you to see- it is the scale length that matters to tension. If the TSL is the scale length, then a longer one would have more tension. If they be different, the scale length sets the tension, whether the TSL be 1 mile, or 1 foot. So close! Thanks for keeping it civil! JP

Curt Langston · Post by **Curt Langston** » 9 Jul 2006 2:35 pm

Jim, David, Eric, Charlie, b0b, Ed et al :

Actually, I think I see your point. I believe that I have been wrong all this time.

Hmmm.

Oh well.

What can I say.

Guess I'm ready to drop it.

So, basically, the scale length on a guitar has to be under 25 inches, because if it is much longer, then you have too much tension and the G# won't hold up.

Maybe, the 24 1/4 inch scaled guitars that broke the G#, were doing so because of some sort of resistance created by the roller nut. Or, perhaps too sharp a bend at the nut. (seems like I heard that somewhere)

And come to think of it, 25 inch scale keyless guitars seem to be a little tighter.

Eureka!

It all seems so logical now.

I knew you guys would finally get it!

I'm convinced that you guys are right, and I was wrong.

Totally wrong all along!

I don't understand the results of the barbell test, but I do see the logic in your posts, and I now agree.

You guys did a fine job convincing me.

I appreciate all the input. If nothing else, I learned something.

And heck, now I think I'll go out and buy me another Carter.

I always did like them.

Go ahead and give me a good razzing. I deserve it!

This is b0bs thread. He can close it as he pleases.[This message was edited by Curt Langston on 09 July 2006 at 04:08 PM.]